Test CircleCount PRO now!
Login now

Not your profile? Login and get free access to your reports and analysis.

Mark Bruce

Mark Bruce 

A transhumanist and technophile residing in Australia, loving life and avidly looking forward to the future.

Location: Adelaide

Followers: 15,530

Following: -

Views: 123,739,715

Cream of the Crop: 04/19/2012

Added to CircleCount.com: 03/21/2012That's the date, where Mark Bruce has been indexed by CircleCount.com.
This hasn't to be the date where the daily check has been started.
(Update nowYou can update your stats by clicking on this link!
This can take a few seconds.
)

Tags

Sign in

The following tags have been added by users of CircleCount.com.
You can login on CircleCount to add more tags here.

  • Scientists

Are you missing a tag in the list of available tags? You can suggest new tags here.

Login now

Do you want to see a more detailed chart? Check your settings and define your favorite chart type.

Or click here to get the detailed chart only once.

Shared Circles including Mark Bruce

Shared Circles are not available on Google+ anymore, but you can find them still here.

The Google+ Collections of Mark Bruce

Activity

Average numbers for the latest posts (max. 50 posts, posted within the last 4 weeks)

10
comments per post
12
reshares per post
48
+1's per post

4,494
characters per posting

Top posts in the last 50 posts

Most comments: 71

posted image

2016-04-12 13:52:17 (71 comments; 16 reshares; 70 +1s)Open 

Ultrasociety: How 10,000 Years of War Made Humans the Greatest Cooperators on Earth by Peter Turchin

This book changed my conception of war, culture, society, cooperation, competition, religion, and memes - to a lesser or greater extent.

The main thrust of Ultrasociety concerns support of new theories of evolutionary group selection to explain the rise and nature of the complex societies we currently find ourselves in, while challenging the standard criticisms of conventional evolutionary group selection. While genetics is of course mentioned, these advanced theories of group selection are based on Multilevel Cultural Selection, the selection of cultural traits, of memes or memeplexes, that lead to the evolution of ever-larger groups better able to compete, better able to cooperate, better able to produce more resources and more equitable societies. And this selection for the... more »

Most reshares: 43

posted image

2016-04-07 12:18:17 (15 comments; 43 reshares; 137 +1s)Open 

The Accelerating Feedback Loop Between Brain Mapping and Machine Learning

Quanta has yet another great article providing a relatively detailed overview of the current state of brain mapping advances, machine learning advances, and the accelerating feedback between the two helping to drive progress forward https://www.quantamagazine.org/20160406-brain-maps-micron-program-iarpa/.

It covers the Machine Intelligence from Cortical Networks program run by the Intelligence Advanced Research Projects Activity, one of whose goals is to Revolutionize machine learning by reverse engineering the algorithms and computations of the brain. Indeed, machine learning algorithms they will develop based on the neural connection diagrams they uncover will be tested against pattern recognition tasks.

One challenge will be dealing with the enormous amounts of data the research produces —1... more »

Most plusones: 137

posted image

2016-04-07 12:18:17 (15 comments; 43 reshares; 137 +1s)Open 

The Accelerating Feedback Loop Between Brain Mapping and Machine Learning

Quanta has yet another great article providing a relatively detailed overview of the current state of brain mapping advances, machine learning advances, and the accelerating feedback between the two helping to drive progress forward https://www.quantamagazine.org/20160406-brain-maps-micron-program-iarpa/.

It covers the Machine Intelligence from Cortical Networks program run by the Intelligence Advanced Research Projects Activity, one of whose goals is to Revolutionize machine learning by reverse engineering the algorithms and computations of the brain. Indeed, machine learning algorithms they will develop based on the neural connection diagrams they uncover will be tested against pattern recognition tasks.

One challenge will be dealing with the enormous amounts of data the research produces —1... more »

Latest 50 posts

posted image

2016-05-28 08:17:41 (4 comments; 8 reshares; 29 +1s)Open 

Inflammatory Alarmism and Other Nonsense Against Human Genome Editing

I came across a few articles recently that promoted outrageous inflammatory alarmism over human genome editing. The sentiment, pessimism, and authoritarianism present in all cases annoyed and offended me deeply and I couldn't help but get a little rant off my chest. This little rant turned out a little longer than anticipated.


We Need a Species-Wide Conversation About the Future of Human Enhancement
http://techcrunch.com/2016/05/01/homo-sapiens-2-0-we-need-a-species-wide-conversation-about-the-future-of-human-genetic-enhancement/

This piece discusses the impending future of human genome editing and enhancement via CRISPR and other tools, but leans on the side of allowing the dictates of the uninformed majority to guide development and finishes with outright alarmist... more »

Inflammatory Alarmism and Other Nonsense Against Human Genome Editing

I came across a few articles recently that promoted outrageous inflammatory alarmism over human genome editing. The sentiment, pessimism, and authoritarianism present in all cases annoyed and offended me deeply and I couldn't help but get a little rant off my chest. This little rant turned out a little longer than anticipated.


We Need a Species-Wide Conversation About the Future of Human Enhancement
http://techcrunch.com/2016/05/01/homo-sapiens-2-0-we-need-a-species-wide-conversation-about-the-future-of-human-genetic-enhancement/

This piece discusses the impending future of human genome editing and enhancement via CRISPR and other tools, but leans on the side of allowing the dictates of the uninformed majority to guide development and finishes with outright alarmist conclusions that cataclysmic outcomes will result if we don’t do this. I think it is well written but basically fanciful in its main thrust.

We get a discussion of how over nearly 40 years the first test tube baby went from being a moral abomination to passing a very routine 5 millionth IVF birth, and how human genome editing will probably go through a similar phase of aversion and acceptance. This will be unavoidable and start with fighting disease. There are many bits I have a problem with:

Not everyone will be comfortable with genetic enhancement based on some people’s understandable ideological or religious beliefs or for real or perceived safety concerns. Life is not just about science and code. It involves mystery and chance and, for some, spirit. This erroneously presumes you won’t still have those things when genome editing is common, which is just wrong and misleading.

69 percent of Americans felt that genetically altering unborn babies to reduce their risk of developing serious diseases should be illegal. The sheer level of ignorant sadism in this statement, shared by more than two thirds of the population, is just breathtaking. When reducing the suffering of another human is trivially easy, and you seek to use state force to prevent people from doing so, and so cause needless human suffering, you are most definitely part of the problem.

Eighty three percent felt that genetic alterations to improve the intelligence or physical characteristics of unborn babies should be banned. Again, what right do people think they have to dictate what other people can and can’t do to improve the life of their children? How is their desire to use state force to stop people from doing this any different than those people using the state to enforce the adoption of these techniques on all people whether they want them or not?

No matter what the intention of parents, might genetic selection of children become a form of liberal or not-so-liberal Eugenics that challenges the moral core of our humanity? Might it encourage us to devalue the critically important and varied contributions everyone makes in a diverse society? There is more than a hint of regressive political correctness here and ill-considered virtue signalling, assuming one’s own morals and values are the moral core and collective values of humanity and seeking to impose this on everyone else, and all while using shaming name-calling (Eugenics) to shut down rational debate.

If a relatively small number of even very well intentioned people unleash a human genetic revolution that will ultimately touch most everyone and alter our species’ evolutionary trajectory without informed, meaningful, and early input from others, the backlash against the genetic revolution will overwhelm its monumental potential for good. Ending on a rather alarmist and pessimistic note. The whole piece almost becomes contradictory here, calling for a species-wide conversation on our future use of these technologies, which would not only be infeasible in principle, but it sure as hell wouldn’t be informed or meaningful with most of those people.


A Secret Meeting on Creating Synthetic Human Genomes
TechnologyReview coverage: https://www.technologyreview.com/s/601540/big-ideas-big-conflicts-in-plan-to-synthesize-a-human-genome/
Engadget coverage:
http://www.engadget.com/2016/05/13/scientists-held-a-secret-meeting-to-debate-creating-synthetic-hu/
Cosmos criticism:
https://cosmosmagazine.com/society/should-we-synthesise-human-genome

So a bunch of geneticists and technologists met at Harvard recently to discuss and plan the Human Genome Project Write, focused on the technical ability to synthesise very large whole genomes from scratch, basically doing for human cells what Craig Venter has already done for minimal microbes. This amounts to pitching a mega-project to help drive technological evolution and drastically reduce the price of DNA synthesis. But they’ve been slammed for making the meeting “secret”.

The TechnologyReview article is quite good and includes rational responses from George Church towards some of the myopic criticism the proposal received.

The Engadget piece has just terrible science reporting:
At the top of their worry pile is that the project will end up innovating new ways to make synthetic human genes, which could lead to them being used to artificially create humans. Nope, we can already create synthetic human genes if we want, and we already have IVF and other techniques for artificially creating humans. This is misleading alarmist BS.

And if the meeting was about synthetic human genes, there should be a public debate on the morality of creating or editing human genes. Human genes are already being edited as part of standard gene therapies; the debate has already been had. But no, let’s hinder and block the technology again and force the morals of some segment of the population onto everyone else and block free expression and choice.

The original criticism in Cosmos is just backwards and full of the same authoritarian moralistic impositions as other pieces here:
A project that made polio virus from scratch in 2002 generated such fear that public funding for improving DNA synthesis tools was cancelled, unwittingly harming research across diverse and unrelated fields while policy makers struggled to imagine how such tools could ever be controlled. So yes let’s bow down again to the ignorant and the uninformed.

Would it be OK, for example, to sequence and then synthesise Einstein’s genome? Yep, it’d be irrelevant.

We note that the narrative of creation of the human is the central narrative for many religious communities. How is this even remotely relevant to the matter at hand?

To create a human genome from scratch would be an enormous moral gesture whose consequences should not be framed initially on the advice of lawyers and regulators alone. Nope, in the grand scheme of things it would be a trivial and quite irrelevant step, just doing for larger genomes what we’ve already done for smaller, and opening up a range of more interesting applications that would be consequently cheaper and easier to pursue. Sure, if you’re a religious nut believing in a creator santa claus then it might seem morally relevant but to those unblinded by such then it is morally irrelevant. The value of a human lies in more than just their DNA code.

The perspectives of others including self-identified theologians, philosophers, and ethicists from a variety of traditions should be sought out from the very beginning. Yes let’s prostrate ourselves before theologians of all things and self-appointed morally superior bioethicists, and follow their dictates because the moral compass and ethical outlook of those actually driving and developing the technology, those who understand it best, are obviously inferior. Simply unfriggenbelievable that in this day and age there are calls for theologians to be involved in setting science policy and guidelines.


Let People Most Affected by Gene Editing Write CRISPR Rules
https://www.newscientist.com/article/2086548-let-people-most-affected-by-gene-editing-write-crispr-rules/

A fantastic example of self-righteous bioethicists seeking to arrogantly impose and enforce their morals on everyone else in order to justify their existence. Far be it from those most versed in the technology, those most expert, those most willing to devote time, effort and resources to the technology, no; instead apparently the disabled and the third world should write the rules of how CRISPR is used that everyone else should be forced to follow.

And all because some disabled people might like being disabled and might not like the idea of gene editing to cure their disability. Aside from the fact that use of such technology would be a choice, no; they might not like it so they should be able to control the availability of that choice for everyone else.

And also because this will be an elitist treatment that won’t benefit the poor in the third world and has to be shared equally. Aside from the fact that the massive amount of investment involved dictates first generation CRISPR therapies will be costly, not least because those same bioethicists dictate onerous and incredibly expensive clinical trials forcing this to be so, and ignoring the fact that not much more than a decade at most after first market entry cheap generic versions of these same therapies will be manufactured and available to everyone just like current drugs today.

And even because women will suffer because they are the primary carers of people with disabilities, and this risks putting them out of a job. Seriously.

I find sentiment like this unbelievably disappointing. Let’s hinder CRISPR, slow it down, make it harder for therapies to be developed, tie it up in regulation, suffocate innovation, drive it underground. Just get out of the way and let the technology evolve; the benefits will come. Attempts at prohibition will just deliver nasty unexpected side effects and costs. Just like they do every, single, time.

The one positive was specific mention of and support for DIY at-home CRISPR hacking. So it wasn’t a complete waste.


The above is just so damn depressing that I have to finish on an optimistic note: BioViva ups the ante on medical tourism by planning an age-reversal clinic in Fiji.
https://www.inverse.com/article/15895-bioviva-plans-an-age-reversal-clinic-in-fiji-as-medical-tourism-gets-weird

The trailblazing Liz Parish, CEO of BioViva, who last year performed the first life-extension gene therapies on herself, has partnered with Sierra Sciences and announced the creation of the world’s first age reversal clinic in Fiji. Fiji has implemented very progressive policies with regards to medical regulation, branding itself as a medical tourism destination, and provides tax incentives to medical institutes.

Looks like the partnership hopes to kick things off with a launch to medical tourists in 2017 or 2018 at the latest, and using revenues to bring costs down. This is great, optimistic news and a real cause for hope. Even if the worst comes to pass, as hinted at in the above list of myopic attempts at control of biotechnology, then we should still be able to jump on a plane to Fiji and other places in order to access the tools and treatments that we want.

As always, prohibition never works. ___

posted image

2016-05-25 13:16:55 (12 comments; 36 reshares; 87 +1s)Open 

How Technology Hacks People's Minds

This article is just too damn good not to share - easily the best I've read in recent memory. I forget who first shared it here but thank you.

https://medium.com/@tristanharris/how-technology-hijacks-peoples-minds-from-a-magician-and-google-s-design-ethicist-56d62ef5edf3#.x953hbytl

It discusses 10 psychological hacks used in designing (typically) digital products and services to better grab and hold your attention, to the point of inducing a form of addiction if possible. In summary these are:

1. Controlling the menu to control the choices.
2. Making the experience like a slot machine.
3. The fear of missing something important.
4. The need for social approval.
5. Social favours and reciprocity.
6. Infinite feeds and autoplay.
7. Instant, "urgent" interruption. ... more »

How Technology Hacks People's Minds

This article is just too damn good not to share - easily the best I've read in recent memory. I forget who first shared it here but thank you.

https://medium.com/@tristanharris/how-technology-hijacks-peoples-minds-from-a-magician-and-google-s-design-ethicist-56d62ef5edf3#.x953hbytl

It discusses 10 psychological hacks used in designing (typically) digital products and services to better grab and hold your attention, to the point of inducing a form of addiction if possible. In summary these are:

1. Controlling the menu to control the choices.
2. Making the experience like a slot machine.
3. The fear of missing something important.
4. The need for social approval.
5. Social favours and reciprocity.
6. Infinite feeds and autoplay.
7. Instant, "urgent" interruption.
8. Making your reasons their reasons.
9. Offering choices that are inconvenient.
10. Forecasting errors via foot-in-the-door tactics.

The author Tristan Harris worked for Google as a Product Philosopher / Design Ethicist exploring and recommending ways to protect and defend people's minds from being hijacked and their time from being wasted.

We need our smartphones, notifications screens and web browsers to be exoskeletons for our minds and interpersonal relationships that put our values, not our impulses, first. People’s time is valuable. And we should protect it with the same rigor as privacy and other digital rights.

Other Resources

Tristan's TEDx talk: https://www.youtube.com/watch?v=jT5rRh9AZf4

Tristan's resource site for better design and time saving: http://timewellspent.io/___

posted image

2016-05-25 11:50:32 (9 comments; 2 reshares; 38 +1s)Open 

Thanks for the Free Upgrade Google

YouTube Red finally launched in Australia a week or so ago and I was happy to note that my existing $10 per month Google Play Music subscription automatically granted me access upon opening the newly updated YouTube app for the first time after the announcement. So this was like a free bonus as no extra $10 per month was needed.

The updated YouTube app is damn slick and smooth. With the Red option active, little differences that seem minor, such as (i) no adverts to skip or sit through, (ii) the ability to play audio in the background while using other apps, and (iii) the ability to play audio with the screen off, make a very noticeable and exceptionally welcome impact when using YouTube. The one thing I don't expect to use much is the "download to watch offline" function.

I listen to a lot of YouTube talks, lectures,... more »

Thanks for the Free Upgrade Google

YouTube Red finally launched in Australia a week or so ago and I was happy to note that my existing $10 per month Google Play Music subscription automatically granted me access upon opening the newly updated YouTube app for the first time after the announcement. So this was like a free bonus as no extra $10 per month was needed.

The updated YouTube app is damn slick and smooth. With the Red option active, little differences that seem minor, such as (i) no adverts to skip or sit through, (ii) the ability to play audio in the background while using other apps, and (iii) the ability to play audio with the screen off, make a very noticeable and exceptionally welcome impact when using YouTube. The one thing I don't expect to use much is the "download to watch offline" function.

I listen to a lot of YouTube talks, lectures, and debates and the video imagery is usually optional. Being able to just listen to the audio with screen off while driving is great and also saves on mobile data consumption. I know there have been specialised apps for years that have allowed you to do this but having the ability in the native YouTube app just makes things that much easier and convenient. Also: similar benefits on desktop and Chromecast.

Anyone else using and enjoying YouTube Red? Any complaints? ___

posted image

2016-05-24 13:45:17 (6 comments; 17 reshares; 90 +1s)Open 

Vysor

I've been playing around with Koush's new Vysor app, which you can grab via http://www.vysor.io/

Basically Vysor lets you control your phone from your computer and allows you to generate a shareable link to the session for other people or devices to control your phone from anywhere. It's all done in the browser and the execution is pretty slick.

Install the Chrome webapp from the Chrome Store, plug your phone in (my Nexus 6P here), launch the Vysor app, a Chrome window opens to allow mouse control of the phone and keyboard entry, etc. Click a button to share access, grab the link, send to someone else, they click and can execute the same control of your phone in their browser (example here of wife logged into Chromebook).

Potentially useful to troubleshoot a friend's or family member's phone remotely and quickly. Or perhaps... more »

Vysor

I've been playing around with Koush's new Vysor app, which you can grab via http://www.vysor.io/

Basically Vysor lets you control your phone from your computer and allows you to generate a shareable link to the session for other people or devices to control your phone from anywhere. It's all done in the browser and the execution is pretty slick.

Install the Chrome webapp from the Chrome Store, plug your phone in (my Nexus 6P here), launch the Vysor app, a Chrome window opens to allow mouse control of the phone and keyboard entry, etc. Click a button to share access, grab the link, send to someone else, they click and can execute the same control of your phone in their browser (example here of wife logged into Chromebook).

Potentially useful to troubleshoot a friend's or family member's phone remotely and quickly. Or perhaps remotely control a device that the phone has been linked to and has control over, e.g. LIFX lights, a Lego robot, etc. Or even set up as a remote video camera.

Next step will be full wireless capabilities to untether from the computer if possible; it should be given I played with basic apps with that level of functionality about 4 years ago, e.g. https://plus.google.com/u/0/+MarkBruce/posts/hscDMkAnsh6.

Note I paid the $10 for the Pro version.___

posted image

2016-05-24 11:53:32 (14 comments; 4 reshares; 28 +1s)Open 

Your Brain Does Not Store or Process Information
Or so claims Robert Epstein in this piece: https://aeon.co/essays/your-brain-does-not-process-information-and-it-is-not-a-computer

There are quite a few problems with this article and its thesis but I still think it is worth a read for those interested in the brain, cognition, and consciousness. The article is ultimately an appeal to the field of Embodied Cognition.

Some of the worthwhile bits include passages like the following and just simply recognising the fundamental importance of metaphor in our thinking, our understanding, and how we see the world:

The information processing (IP) metaphor of human intelligence now dominates human thinking, both on the street and in the sciences. There is virtually no form of discourse about intelligent human behaviour that proceeds without employing this metaphor, just as... more »

Your Brain Does Not Store or Process Information
Or so claims Robert Epstein in this piece: https://aeon.co/essays/your-brain-does-not-process-information-and-it-is-not-a-computer

There are quite a few problems with this article and its thesis but I still think it is worth a read for those interested in the brain, cognition, and consciousness. The article is ultimately an appeal to the field of Embodied Cognition.

Some of the worthwhile bits include passages like the following and just simply recognising the fundamental importance of metaphor in our thinking, our understanding, and how we see the world:

The information processing (IP) metaphor of human intelligence now dominates human thinking, both on the street and in the sciences. There is virtually no form of discourse about intelligent human behaviour that proceeds without employing this metaphor, just as no form of discourse about intelligent human behaviour could proceed in certain eras and cultures without reference to a spirit or deity. The validity of the IP metaphor in today’s world is generally assumed without question.

But the IP metaphor is, after all, just another metaphor – a story we tell to make sense of something we don’t actually understand. And like all the metaphors that preceded it, it will certainly be cast aside at some point – either replaced by another metaphor or, in the end, replaced by actual knowledge. The IP metaphor is ‘sticky’. It encumbers our thinking with language and ideas that are so powerful we have trouble thinking around them.

Prevailing metaphors adopted to understand the world are heavily influenced by and at times dictated by the technological paradigm of the time. Hindsight allows us to see the errors and simplicity in old, outdated, metaphors. If nothing else the article forces us to ask: is the prevailing metaphor of our times, that of computation, the final metaphor? If not then improved future understanding and metaphors will look at us as we do our forebears. This forces the author to give an account as to why it isn’t the final metaphor but I don’t think they achieve this.

As an example, the dollar bill-in-memory test doesn’t appear to offer a satisfactory explanation:

But she hadn’t made a deliberate effort to ‘memorise’ the details. Had she done so, you might argue, she could presumably have drawn the second image without the bill being present.

And also:

But neither the song nor the poem has been ‘stored’ in it. The brain has simply changed in an orderly way that now allows us to sing the song or recite the poem under certain conditions . . . We simply sing or recite – no retrieval necessary.

The author argues that even if she draws the dollar bill perfectly from memory, she doesn’t actually have the bill in memory in her brain. This seems like a tautology of sorts or else I have something very wrong here. My understanding, to put it simply, is that the act of memorisation of a detailed object involves the brain forming an ever-more-accurate pattern in the brain representing the object and this accurate pattern can be re-experienced, i.e. re-membered, in order to recognise or reproduce it in future; surely this counts as storing the memory of that thing and this pattern has been computed by the neural networks of the brain?

Throughout the piece I kept wanting a clear and coherent alternative to be presented. The author claims to do so, but as far as I can tell either fails, demands too much prior jargon from the reader, or otherwise dispenses with any clarity. The closest to clarification they come to concerns a description of catching a flying ball:

The IP perspective requires the player to formulate an estimate of various initial conditions of the ball’s flight – the force of the impact, the angle of the trajectory, that kind of thing – then to create and analyse an internal model of the path along which the ball will likely move, then to use that model to guide and adjust motor movements continuously in time in order to intercept the ball.

That is all well and good if we functioned as computers do, but McBeath and his colleagues gave a simpler account: to catch the ball, the player simply needs to keep moving in a way that keeps the ball in a constant visual relationship with respect to home plate and the surrounding scenery (technically, in a ‘linear optical trajectory’). This might sound complicated, but it is actually incredibly simple, and completely free of computations, representations and algorithms.

With this and other descriptions (see below) I can’t help but think the author is blinded or hindered by an incredibly constrained understanding or definition of computation. To me this explanation manages to explain very little, and certainly not how it is free of computation. Fortunately the author makes reference to and recommends other prominent proponents of Embodied Cognition and after searching a bunch of their blog posts I found what appears to be their best explanation here http://psychsciencenotes.blogspot.com.au/2015/07/brains-dont-have-to-be-computers-purple.html.

While their explanation may be even more obtuse, one of the key examples or analogies they are relying on concerns the Polar Planimeter. Knock out the Planimeter and you knock out one of the foundations of their argument. They claim the Planimeter doesn’t actually compute the area of the shape it traces out, despite the fact that it takes an input - moving the needle around the edge of the shape - and produces an output - the area of the shape thus traversed. It seems to me as though the Planimeter does indeed compute the area of the shape or am I missing something here?

The computation, or algorithmic function, for determining the area of the shape transcribed is encoded in the design of the device and its gears, or so it appears to me. Am I wrong or missing something here? They seem to claim something along the lines of “simply by interacting with its environment the Planimeter naturally produces a suitable response to that environment” but it all seems terribly hand-wavy and imprecise, and again suffers from a restricted definition of computation. Others have referred to Planimeters as analogue calculating devices; surely they can also be referred to as analogue computing devices?

A problem with this passage and those that precede it:

Fortunately, because the IP metaphor is not even slightly valid, we will never have to worry about a human mind going amok in cyberspace; alas, we will also never achieve immortality through downloading.

This seems to claim that the human mind / cognition / consciousness is not physical, that it exists apart from matter and physical law. There are few people who have time to entertain such simplistic dualism. All evidence points to these things having a physical basis and as such claiming substrate independence for the phenomena is a reasonable claim. I suspect the author is caught up in holding only the most basic of computational substrates as a possible alternative, when other substrates can easily be posited to address arguments he has against these.

Finally, for this passage:

Worse still, even if we had the ability to take a snapshot of all of the brain’s 86 billion neurons and then to simulate the state of those neurons in a computer, that vast pattern would mean nothing outside the body of the brain that produced it.

I would ask if fMRI studies are now allowing us to partially determine what someone is thinking then surely running a complete connectome simulation and likewise measuring activity would allow us to determine what that brain was thinking, even without a body?

Planimeters
[1] https://www.youtube.com/watch?v=_W35iDhRfZg
[2] https://www.youtube.com/watch?v=kdxPEZnv-U0
[3] https://www.youtube.com/watch?v=l_k_0hRpOA4
[4] https://en.wikipedia.org/wiki/Planimeter ___

posted image

2016-05-23 11:55:52 (7 comments; 0 reshares; 34 +1s)Open 

Serving up slices of home-made bread to go with our home-made soup last night I couldn't help but notice it looked like the cross-section of a brain. A little bit of frontal lobe damage on the left hemisphere perhaps but damn tasty nonetheless. 

Serving up slices of home-made bread to go with our home-made soup last night I couldn't help but notice it looked like the cross-section of a brain. A little bit of frontal lobe damage on the left hemisphere perhaps but damn tasty nonetheless. ___

posted image

2016-05-22 07:02:02 (23 comments; 32 reshares; 86 +1s)Open 

SciTech #ScienceSunday Digest - 21/2016.
Permalink here: http://www.scitechdigest.net/2016/05/massive-photon-theories-chiral-mirror.html

Massive photon theories, Chiral mirror enzymes, TensorFlow ASIC, Efficient solar cells, Carbon computing, Better autonomous vehicles, Fast wireless Internet, Fixing mitochondria, 3D printed hair, Smartphone urine test.

1. Dark Energy & Massive Photons
A new theory suggests that if photons have a mass of about 10^-70kg then the way that photons interact with different fields and matter in the Universe would lead to a repulsive effect that looks a lot like dark energy causing the Universe to expand http://www.sciencealert.com/heavy-light-could-explain-dark-energy. This replaces the current unexplainable cosmological constant with an unmeasurable property in the tiny mass of photons. Of course it also goes... more »

SciTech #ScienceSunday Digest - 21/2016.
Permalink here: http://www.scitechdigest.net/2016/05/massive-photon-theories-chiral-mirror.html

Massive photon theories, Chiral mirror enzymes, TensorFlow ASIC, Efficient solar cells, Carbon computing, Better autonomous vehicles, Fast wireless Internet, Fixing mitochondria, 3D printed hair, Smartphone urine test.

1. Dark Energy & Massive Photons
A new theory suggests that if photons have a mass of about 10^-70kg then the way that photons interact with different fields and matter in the Universe would lead to a repulsive effect that looks a lot like dark energy causing the Universe to expand http://www.sciencealert.com/heavy-light-could-explain-dark-energy. This replaces the current unexplainable cosmological constant with an unmeasurable property in the tiny mass of photons. Of course it also goes against all known physics that demands light have zero mass. Still interesting as it ties in with recent work on the EM Drive and the Unruh radiation explanation for acceleration that is dependent on photons having a small inertial mass. In related news light has been discovered to have another property thought to be impossible and not in line with current physics, that of half-integer angular momentum https://www.tcd.ie/news_events/articles/physicists-discover-a-new-form-of-light/6815#.Vzv11d8rJR0.

2. Synthetic Chiral Mirror Enzymes
All of life is built on and deals with handed or chiral molecules: all amino acids are left-handed while all DNA twists like a right-handed screw. For the first time a synthetic chiral mirror-image polymerase protein that can both copy mirror-DNA and transcribe it to mirror-RNA has been engineered http://www.nature.com/news/mirror-image-enzyme-copies-looking-glass-dna-1.19918. This part of a wider research program to make a complete mirror-image cell in which all proteins, polynucleotides, and other molecules are chiral, mirror images of a normal cell. This is interesting because mirror-image chemistries are incompatible with each other and so such a biological system would have natural resistance to normal viruses, degradation enzymes, and potentially have potent effects.

3. Machine Learning AISC by Google
Google revealed that it developed, tested, deployed, and used - over the last year - a custom ASIC chip called a Tensor Processing Unit (TPU) specifically for machine learning and tailored to their TensorFlow platform https://cloudplatform.googleblog.com/2016/05/Google-supercharges-machine-learning-tasks-with-custom-chip.html. TPUs deliver 10x better performance per watt for machine learning, leaping ahead 7 years with regards to Moore’s Law. They are currently used by 100 teams in Google and power things like RankBrain, StreetView, and AlphaGo and are being made available to third party developers. Meanwhile machine learning is being applied to controlling Bose-Einstein Condensates http://www.anu.edu.au/news/all-news/artificial-intelligence-replaces-physicists and discover new materials http://www.nature.com/news/can-artificial-intelligence-create-the-next-wonder-material-1.19850.

4. Solar Cells Reach 34.5% Efficiency
A new four-junction solar cell demonstrates non-concentrated light-to-energy conversion efficiency of 34.5%, significantly improving on the previous record holder at 24% http://newsroom.unsw.edu.au/news/science-tech/milestone-solar-cell-efficiency-unsw-engineers. As a comparison recent technology roadmaps aimed to achieve 35% by 2050. Multi-junction architectures are more complex and costly than simpler alternatives but the team are working to reduce manufacturing complexity and reduce the cost of such cells to help facilitate broader uptake.

5. Carbon Computing Advances
Akhan Semiconductor is set to announce a diamond integrated circuits, and has demonstrated diamond devices running at 100 GHz with 100nm feature size, due mainly to diamond’s vastly superior thermal conductivity compared to silicon http://www.eetimes.com/document.asp?doc_id=1329663. While they wish to revitalise the processor clock race they are focusing initially on power electronics and heat management. Meanwhile a new bilayer graphene-based transistor design also claims the possibility of achieving 100 GHz processing speeds while achieving low-voltage switching http://www.nanowerk.com/nanotechnology-news/newsid=43427.php.

6. Autonomous Vehicles Getting Better
New dynamic planning and control algorithms allow autonomous cars (at least scale models) to powerslide and drift around dirt tracks http://spectrum.ieee.org/cars-that-think/transportation/self-driving/autonomous-mini-rally-car-teaches-itself-to-powerslide. It’ll be interesting to watch this demonstrated in a full-size rally car; such control algorithms will help make autonomous cars safer too. Meanwhile Otto is developing an autonomous self-driving system for the purpose of retrofitting trucks, an attractive, cheaper option for converting existing trucking fleets http://nextbigfuture.com/2016/05/billion-dollar-startup-otto-developing.html.

7. Super-Fast Wireless Internet
Starry, a new Internet service provider from the guy behind failed TV rebroadcaster Aereo, is solving the problem of last-mile high-speed Internet access using their new Starry Router (installed outside the customer’s home) and their own fibered transmitter that is able to serve 600 - 900 routers within a 1km range at speeds of 300Mbps to 1Gbps https://www.technologyreview.com/s/601442/wireless-super-fast-internet-access-is-coming-to-your-home/. As an infrastructure play customer acquisition is $25 instead of $2,500, and this is all made possible due to newer active phased array technologies solving a range of problems including costs, range, and interference. My Australian government National Broadband Network should consider something like this to reduce costs and accelerate roll-out.

8. Compensating for Loss of Mitochondrial Function
Engineering allotropic expression of remaining mitochondrial genes takes another step forward with the recent demonstration of the relocation of another two mitochondrial genes to the cell nucleus with proven targeting to mitochondria in the cell and able to compensate for mitochondria in which these genes have been damaged https://www.fightaging.org/archives/2016/05/announcing-success-in-the-mitosens-project-crowdfunded-at-lifespanio-in-2015/. Only 8 more genes to go to address this mechanism of aging. In related - and utterly fascinating news - we have the discovery of a eukaryotic single-celled organism that appears to have dispensed with its energy-producing mitochondria and replaced it with a different system acquired from bacteria via lateral gene transfer http://www.cell.com/current-biology/abstract/S0960-9822%2816%2930263-9; I’d love to see whether a multicellular organism like a nematode could be engineered with this replacement.

9. 3D Printed Hair
3D printed polymer strands about 100 microns thick produce a surface of well-defined hair that not only provides a soft, different texture for printed objects but can be used to produce surprising effects http://www.engadget.com/2016/05/16/3d-printed-hair-finds-practical-uses/, as can be seen in the video. By making the hairs vibrate for example you can make objects placed on the surface move over it and follow a specific path, or folded onto itself it can produce motors to make objects perform lateral or rotational movement. Of course opposing hairy surfaces can also function as a type of velcro for certain applications. The video is the killer here - really surprising and very clever.

10. Smartphone Urine Test Device
A new microfluidic urinalysis system ensures that paper-based dip-tests receive the correct amount of urine and consistent lighting levels to enable a smartphone (camera) to analyse colour changes over 30 seconds to 2 minutes and provide an accurate diagnosis for any disease detected https://news.stanford.edu/2016/05/16/stanford-engineers-design-home-urine-test-scan-diseases/. At-home, do-it-yourself, point-of-care diagnostic and medical systems are a hot space undergoing a lot of development and this simple urine analysis system using paper dip sticks with 10 colour-changing pads looks like a very convenient and cheap platform that could take advantage of people developing a range of other paper based tests for other disorders.

SciTech Tip Jar: http://www.scitechdigest.net/p/donate.html___

posted image

2016-05-19 12:43:41 (26 comments; 14 reshares; 53 +1s)Open 

Genetically Engineering Super-Intelligent Humans

Point
Nautilus has an interesting discussion on the thousands (10k) of genetic variations that contribute to or are correlated with human intelligence, and with single variants being responsible for less than 1 point of IQ http://nautil.us/issue/18/genius/super_intelligent-humans-are-coming. Statistical analysis indicates that when an individual possess just 100 more of these positive variants above the population average then they will enjoy an IQ boost of one standard deviation, or an extra 15 points. Swapping out all 10,000 variants in an individual for the optimal positive variants might, in theory, result in that person having an IQ of 1,000 - if you just go by the statistical analysis. It isn’t really clear whether an IQ this high really has much meaning, especially considering human IQ is quoted as just double or triplet... more »

Genetically Engineering Super-Intelligent Humans

Point
Nautilus has an interesting discussion on the thousands (10k) of genetic variations that contribute to or are correlated with human intelligence, and with single variants being responsible for less than 1 point of IQ http://nautil.us/issue/18/genius/super_intelligent-humans-are-coming. Statistical analysis indicates that when an individual possess just 100 more of these positive variants above the population average then they will enjoy an IQ boost of one standard deviation, or an extra 15 points. Swapping out all 10,000 variants in an individual for the optimal positive variants might, in theory, result in that person having an IQ of 1,000 - if you just go by the statistical analysis. It isn’t really clear whether an IQ this high really has much meaning, especially considering human IQ is quoted as just double or triple that of chimpanzees. Regardless, this analysis makes it seem conservative that we can pass the IQ 200 or 300 mark.

The piece discusses support for this in more detail but if it turns out to be true then well within 10 years we will have the technology to fairly easily ensure any particular human baby conceived via IVF will have a 15 point IQ boost. There will be very real consequences for any country that bans such technology.

Counterpoint
PZ Myers disagrees however and attempts to dismiss the proposal in a thoroughly disparaging and ad hominem manner http://freethoughtblogs.com/pharyngula/2016/04/05/superbrains-will-not-come-out-of-a-test-tube/. Resisting the urge to dismiss Myers as a buffoon, at a basic level he simply isn’t convinced that boosting absolute human intelligence is a desirable thing, and would rather indirectly boost intelligence as a whole by improving global nutrition and education. His objections boil down to two things. First, that such a thing would be technically impossible; but this seems utterly unrealistic given the rapid pace of technology development in genetics, CRISPR, and embryo editing.

Second, that such a thing would be theoretically impossible, and quotes evolutionary arguments that (i) if it was possible evolution would have already done it, and (ii) humans are so multidimensional that tweaking one variant for one trait invariably involves compromise for other traits; basically that other things in the organism would suffer.

Rebuttal
Stephen Hsu, the author of the original Nautilus post concisely responds to Myer’s counter-arguments with a rebuttal on his private blog here http://infoproc.blogspot.com.au/2016/04/this-is-for-pz-meyers.html. The rebuttal presents some very straight-forward arguments from basic population genetics that support the proposal, clears up Myer’s confusion between genes and variants, states the completely non-controversial fact that cognitive ability is highly heritable, and presents data supporting the fact that there are many thousands of variants responsible for IQ. Myer’s response to this rebuttal was to dismiss Hsu as a dilettante.

Bonus Coverage
Finally, Scott Alexander from SlateStarCodex weighs in to take apart the multidimensionality claim of Myers http://slatestarcodex.com/2016/05/04/myers-race-car-versus-the-general-fitness-factor/. Turns out it is hard to find traits that are traded off against increases in intelligence, which would otherwise be easy to find if the multidimensionality claim were true. People with high IQ live longer, are taller, healthier during childhood, commit less crime, are fitter, have lower rates of stroke, diabetes, and heart disease, and are possibly more attractive. As usual Scott provides additional, substantial, and fascinating thoughts and analysis that are well worth reading.

I’m reminded of a Nick Bostrom quote:
Far from being the smartest possible biological species, we are probably better thought of as the stupidest possible biological species capable of starting a technological civilization—a niche we filled because we got there first, not because we are in any sense optimally adapted to it.

If it is almost trivially possible to grant a person an additional 15 points of IQ with little to no risk of downside, is it unethical not to do so? ___

posted image

2016-05-17 12:43:44 (14 comments; 6 reshares; 26 +1s)Open 

The Mystery of Consciousness = The Mystery of Matter

This article flips the conventional hard problem of consciousness on its head by suggesting that it is not so much the hard problem of consciousness that is the issue as is the hard problem of matter.

http://www.nytimes.com/2016/05/16/opinion/consciousness-isnt-a-mystery-its-matter.html

This is a refreshing perspective on this most fundamental of all questions and one that I sympathise with.

The hard problem concerns how conscious experiences and raw conscious sensations can arise from physical matter, when the more our knowledge of the brain and its mechanistic function grows, and the more detailed our knowledge of matter becomes, the harder it is to see “where” consciousness resides or “what” role it has or “how” it exists “atop” a physical substrate that appears to have no need of it.

As ... more »

The Mystery of Consciousness = The Mystery of Matter

This article flips the conventional hard problem of consciousness on its head by suggesting that it is not so much the hard problem of consciousness that is the issue as is the hard problem of matter.

http://www.nytimes.com/2016/05/16/opinion/consciousness-isnt-a-mystery-its-matter.html

This is a refreshing perspective on this most fundamental of all questions and one that I sympathise with.

The hard problem concerns how conscious experiences and raw conscious sensations can arise from physical matter, when the more our knowledge of the brain and its mechanistic function grows, and the more detailed our knowledge of matter becomes, the harder it is to see “where” consciousness resides or “what” role it has or “how” it exists “atop” a physical substrate that appears to have no need of it.

As the article points out we’re caught in an incredibly strong belief that we actually know more about the physical, about matter, than we really do. As powerfully descriptive and predictive as our neuroscience, biology, chemistry, and ultimately physics is, none of it tells us anything about the intrinsic nature of the matter that comprises physical structures. Of course, there is a separate debate here about how we interpret, and how we trust the tools and models with which we probe reality. The hard problem is not what consciousness is, it’s what matter is, what the physical is.

I disagree with the article where the author claims there are two groups of people: dualists who believe consciousness is non-physical stuff and eliminativists who believe everything is physical and deny the existence of consciousness. For years I’ve thought both positions are absurd and that there is a third option: everything is physical but our knowledge of the physical, of the fundamental properties of matter, is woefully incomplete, that there is something else about matter, about the physical that is responsible for - or just is - consciousness.

Of course this third option is that which is promoted by the author and the reason this piece resonated with me. And it was this third option that I was getting at, that I was attempting to articulate (poorly) in my previous essays on consciousness here https://plus.google.com/u/0/+MarkBruce/posts/3FA1C5xWg1B and here https://plus.google.com/u/0/+MarkBruce/posts/gEce5bdDjGt and which some people thought worthy of derision if not ridicule.

The piece doesn’t mention panpsychism at all but it is certainly a related philosophical concept. I’ll finish with a few choice quotes from the piece:

We know what conscious experience is because the having is the knowing: Having conscious experience is knowing what it is.

The nature of physical stuff is mysterious except insofar as consciousness is itself a form of physical stuff.

“We know nothing about the intrinsic quality of physical events, except when these are mental events that we directly experience.” - Bertrand Russell___

posted image

2016-05-16 14:22:15 (7 comments; 11 reshares; 33 +1s)Open 

Swarm: A Blockchain Contender for the Distributed Web

Swarm is an Ethereum implementation for a decentralised, distributed version of the Web, facilitating key services such as content storage and distribution via an inherently peer-to-peer storage and serving system http://swarm-gateways.net/bzz:/swarm/.

Swarm is designed to be DDOS resistant, zero downtime, fault tolerant, censorship resistant, and self sustaining, and uses the Ethereum blockchain for domain name resolution and other components of necessary infrastructure. Much like the InterPlanetary FIle System (IPFS) implementation that attempts to provide a similar capability, systems like this just look like the future. The ideas, benefits, and underlying technology are just too damn good, innovation and technical evolution is currently extremely rapid, and such platforms possess transformative capacity once they start... more »

Swarm: A Blockchain Contender for the Distributed Web

Swarm is an Ethereum implementation for a decentralised, distributed version of the Web, facilitating key services such as content storage and distribution via an inherently peer-to-peer storage and serving system http://swarm-gateways.net/bzz:/swarm/.

Swarm is designed to be DDOS resistant, zero downtime, fault tolerant, censorship resistant, and self sustaining, and uses the Ethereum blockchain for domain name resolution and other components of necessary infrastructure. Much like the InterPlanetary FIle System (IPFS) implementation that attempts to provide a similar capability, systems like this just look like the future. The ideas, benefits, and underlying technology are just too damn good, innovation and technical evolution is currently extremely rapid, and such platforms possess transformative capacity once they start to mature.

___

posted image

2016-05-16 13:52:14 (10 comments; 11 reshares; 33 +1s)Open 

Recent Natural Selection in Humans

+Stephen Hsu presents a recent paper demonstrating evidence for very recent natural selection of various traits in humans http://infoproc.blogspot.com.au/2016/05/evidence-for-very-recent-natural.html. This shows humans have been subject to strong selection pressure in just the last 2,000 years, which seems pretty incredible.

Adult height, birth size, female hip size, and birth weight are all traits that seem to be closely related such that selecting for just one of these factors selects for the others as well.

Recent Natural Selection in Humans

+Stephen Hsu presents a recent paper demonstrating evidence for very recent natural selection of various traits in humans http://infoproc.blogspot.com.au/2016/05/evidence-for-very-recent-natural.html. This shows humans have been subject to strong selection pressure in just the last 2,000 years, which seems pretty incredible.

Adult height, birth size, female hip size, and birth weight are all traits that seem to be closely related such that selecting for just one of these factors selects for the others as well.___

posted image

2016-05-15 11:20:10 (5 comments; 14 reshares; 65 +1s)Open 

SciTech #ScienceSunday Digest - 20/2016.
Permalink here: http://www.scitechdigest.net/2016/05/industrial-synthetic-enzymes.html

Industrial synthetic enzymes, Mimicking protein evolution, Topological spintronics, Smooth telepresence robots, Silk preservation technique, Regrowing cartilage, Improving prosthetic hands, Parsey McParseface, Graphene light harvesting, RNA life origins.

1. New Enzymes for Industrial Synthesis
A bacterium with enzymes that make it resistant to heavy metals has been co-opted to produce a bacterium that secretes quantum dots into solution that can then be purified by simple centrifugation http://www1.lehigh.edu/news/quantum-dots-nature. These are cadmium- or lead-sulfide quantum dots, with plans to create more varieties for a range of applications in electronics, photonics, and sensing, and produced via conventional... more »

SciTech #ScienceSunday Digest - 20/2016.
Permalink here: http://www.scitechdigest.net/2016/05/industrial-synthetic-enzymes.html

Industrial synthetic enzymes, Mimicking protein evolution, Topological spintronics, Smooth telepresence robots, Silk preservation technique, Regrowing cartilage, Improving prosthetic hands, Parsey McParseface, Graphene light harvesting, RNA life origins.

1. New Enzymes for Industrial Synthesis
A bacterium with enzymes that make it resistant to heavy metals has been co-opted to produce a bacterium that secretes quantum dots into solution that can then be purified by simple centrifugation http://www1.lehigh.edu/news/quantum-dots-nature. These are cadmium- or lead-sulfide quantum dots, with plans to create more varieties for a range of applications in electronics, photonics, and sensing, and produced via conventional fermentation processes with fine control over nanometer sizes. In other new a new enzyme AbyU has been discovered in bacteria living at the bottom of the ocean that is able to catalyse the industrially important Diels-Alder reaction http://www.bristol.ac.uk/news/2016/may/enzyme-antibiotic-discovery.html.

2. Mimicking Evolution to Develop Novel Proteins
In related protein engineering news a new technique called SEWING has been developed that recombines portions of existing proteins to produce new structures and functions https://news.unchealthcare.org/news/2016/may/unc-school-of-medicine-scientists-digitally-mimic-evolution-to-create-novel-proteins. This is a tool for creating proteins that don’t exist in nature and have structures that allow new functions and reactions that weren’t previously possible. The evolutionary process starts with computer simulations, and in the latest demonstration mapped 50,000 variably stitched proteins to arrive at an optimised sample of 21 structures that were then synthesised in the lab and experimentally verified to have the predicted structure.

3. Spintronics from New Topological Insulators
New topological insulators made from bonding layers of bismuth selenide with magnetic europium sulfide manages to retain all of the properties of a topological insulator but with strong magnetic properties, at room temperature http://news.mit.edu/2016/unexpected-magnetic-effect-thin-film-materials-0509. Such new materials have promising applications in realising spintronics as well as new magnetic memories with possible molecular scale information storage. In related news, graphene has been made magnetic by inserting hydrogen atoms into specific locations of the carbon lattice, and resulting in graphene now also functioning as a promising spintronic material http://www.eetimes.com/document.asp?doc_id=1329572.

4. Silky Smooth Telepresence Robots
Disney research impresses yet again with the demonstration of a telepresence robotic system that is powered by a hybrid hydrostatic transmission system that enables the user to feel immersed in the remote environment via (i) stereoscopic cameras and (ii) head and arm actuation that responds with multiple degrees of freedom and force-feedback https://www.disneyresearch.com/publication/haptic-telepresence-robot/. The system demonstrated physical human interaction and very fine & detailed object manipulation. Meanwhile RE2 Robotics also demonstrated a very impressive remote robotic operation system http://www.gadgetify.com/re2-imitative-controller-robots/.

5. Silk Preservation Technology
Tufts University has demonstrated a couple of silk preservation technology applications. First, an ultra-thin coating of biocompatible silk proves very effective at extending the room temperature shelf-life of a range of different fruits, prolonging freshness and slowing ripening https://now.tufts.edu/news-releases/silk-keeps-fruit-fresh-without-refrigeration. Second, blood samples can be stabilised for long periods of time, even after 84 days, and even at high (45C) temperatures after being encapsulated in air-dried silk protein http://now.tufts.edu/news-releases/silk-stabilizes-blood-samples-months-high-temperatures.

6. Regrowing Cartilage
Regrowing cartilage has proven to be one of those persistent problems people have struggled with. A new bio-glass material that mimics the shock-absorbing and load bearing qualities of real cartilage has the potential to encourage cartilage cells to regrow to help reform and repair damaged areas of cartilage, while the material degrades and disappears over time http://www3.imperial.ac.uk/newsandeventspggrp/imperialcollege/newssummary/news_12-5-2016-9-57-13. The material can be 3D printed to customer or personalised dimensions and lead applications include replacing damaged intervertebral discs and knee cartilage and could be transformational given the limitations of current materials and implants. Meanwhile a new injectable gel helps generate blood vessels https://www.technologyreview.com/s/601416/injectable-gel-generates-new-blood-vessels/.

7. Improvements for Robotic & Prosthetic Hands
Yet another DARPA project is demonstrating natural control of a prosthetic hand and arm by a human amputee, this time utilising Myo armbands to pick up signals from the upper arm and translat them into precise hand movements and facilitated by an osseointegration surgical procedure that secured a metal utility rod to the upper arm bone http://gizmodo.com/darpas-mind-controlled-arm-will-make-you-wish-you-were-1776130193. In related news that may feed into prosthetics like the above, machine learning techniques are helping to produce robotic hands that are approaching human-like dexterity http://www.washington.edu/news/2016/05/09/this-five-fingered-robot-hand-learns-to-get-a-grip-on-its-own/.

8. SyntaxNet & Parsey McParseface
Google opensourced SyntaxNet, a neural network implemented in TensorFlow that provides a foundation for Natural Language Understanding systems, as well as an English language parser called Parsey McParseface (the most accurate such model in the world with 94% accuracy) that can be used to analyse English text http://googleresearch.blogspot.com.au/2016/05/announcing-syntaxnet-worlds-most.html. These tools analyse the linguistic structure of language, explain the functional role of words in sentences, and can be used to automatically extract information, translate text, and better determine the meaning of text.

9. Efficient Light Harvesting with Graphene
By creating a Morie superlattice of layered graphene and boron-nitride researchers created material states that are very efficient at converting light into electrcity, with up to five electrons being kicked to flow through the material with each photon http://www.washington.edu/news/2016/05/13/uw-researchers-unleash-graphene-tiger-for-more-efficient-optoelectronics/. I really liked the exploitation of geometry in this work as it reminded me of this great Numberphile video discussing dot patterns and the surprising superstructures that can result https://www.youtube.com/watch?v=QAja2jp1VjE.

10. RNA Life Origins Gains More Evidence
While it has been previously shown that 2 of the 4 bases in RNA could be made via natural chemical reactions on early Earth, it has only recently been demonstrated that the other 2 bases can also be made via similar natural chemical reactions on early Earth http://www.sciencemag.org/news/2016/05/rna-world-inches-closer-explaining-origins-life. In this case the only requirements were hydrogen, cyanide, water, and a little bit of acid; it turned out to be far simpler than previously thought. The next piece of the puzzle will be finding out how the 4 bases linked together to form the first RNA molecules whose autocatalytic self-replicating behaviour could begin being selected for by evolution.

SciTech Tip Jar: http://www.scitechdigest.net/p/donate.html___

posted image

2016-05-10 12:28:46 (7 comments; 1 reshares; 12 +1s)Open 

[Aristotle], On Trolling

Great share / find by Anders on the nature and categorisation of the phenomenon of Trolling and those who practice it, and written in an eclectic, engaging style that is hard to find these days. Found via Anders' G+ share of the summary of his blog, here http://aleph.se/andart2/academia/aristotle-on-trolling/.

As the OP requires a couple of extra clicks to get the full text, a more direct link to the On Trolling article is here: http://journals.cambridge.org/action/displayAbstract?fromPage=online&aid=10293503&fulltextType=RA&fileId=S2053447716000099 and just click PDF or HTML to read the whole thing.

And as helpfully reminded by Anders, Harry Frankfurt's classic article On Bullshit is also a worthwhile read in this context: http://www.stoa.org.uk/topics/bullshit/pdf/on-bullshit.pdf

I think Aristotle got trolling right: http://aleph.se/andart2/academia/aristotle-on-trolling/___[Aristotle], On Trolling

Great share / find by Anders on the nature and categorisation of the phenomenon of Trolling and those who practice it, and written in an eclectic, engaging style that is hard to find these days. Found via Anders' G+ share of the summary of his blog, here http://aleph.se/andart2/academia/aristotle-on-trolling/.

As the OP requires a couple of extra clicks to get the full text, a more direct link to the On Trolling article is here: http://journals.cambridge.org/action/displayAbstract?fromPage=online&aid=10293503&fulltextType=RA&fileId=S2053447716000099 and just click PDF or HTML to read the whole thing.

And as helpfully reminded by Anders, Harry Frankfurt's classic article On Bullshit is also a worthwhile read in this context: http://www.stoa.org.uk/topics/bullshit/pdf/on-bullshit.pdf

posted image

2016-05-08 07:34:05 (11 comments; 27 reshares; 69 +1s)Open 

SciTech #ScienceSunday Digest - 19/2016.
Permalink here: http://www.scitechdigest.net/2016/05/towards-artificial-womb-printed.html

Towards artificial wombs, Printed terahertz lens, Autonomous robot surgeons, Cloud quantum computing, Machine learning development, Rejuvenation updates, Biomimicking bee navigation, Modular microfluidics, Femtosecond clock synchronisation, Holographic displays.

1. Towards an Artificial Womb
An artificial placenta has been developed that successfully kept alive extremely premature baby lambs outside the womb http://labblog.uofmhealth.org/health-tech/artificial-placenta-holds-promise-for-extremely-premature-infants. The device utilises extracorporeal membrane oxygenation to help oxygenate the tissues when the lungs are not fully developed; in humans this might allow the current limit of 24 weeks to be pushed back... more »

SciTech #ScienceSunday Digest - 19/2016.
Permalink here: http://www.scitechdigest.net/2016/05/towards-artificial-womb-printed.html

Towards artificial wombs, Printed terahertz lens, Autonomous robot surgeons, Cloud quantum computing, Machine learning development, Rejuvenation updates, Biomimicking bee navigation, Modular microfluidics, Femtosecond clock synchronisation, Holographic displays.

1. Towards an Artificial Womb
An artificial placenta has been developed that successfully kept alive extremely premature baby lambs outside the womb http://labblog.uofmhealth.org/health-tech/artificial-placenta-holds-promise-for-extremely-premature-infants. The device utilises extracorporeal membrane oxygenation to help oxygenate the tissues when the lungs are not fully developed; in humans this might allow the current limit of 24 weeks to be pushed back significantly. Meanwhile new protocols allow human embryos to reach the 2 week stage of development for the first time, and was only cut short due to established ethical policies that will hopefully be overturned http://www.theverge.com/2016/5/4/11591318/cambridge-rockefeller-university-human-embryo-study-ethics-limit. I can see this gap continuing to shrink.

2. 3D Printed Terahertz Lens
A lens for focusing the terahertz spectrum of light has been developed by 3D printing a metamaterial structure with a gradient refractive index https://3dprint.com/132220/northwestern-terahertz-lens/. This method used a curable photopolymer to build up the precise features of the lens, which might be useful in imaging, security, and biological applications. In other 3D printing news a better recipe has been developed for 3D printed bones http://inbt.jhu.edu/2016/05/04/building-a-better-recipe-for-3d-printed-bones/.

3. Autonomous Robot Surgeons
In certain surgical tasks robots are now outperforming humans, as demonstrated in recent work in which a robot stitched up a pig’s small intestine http://spectrum.ieee.org/the-human-os/robotics/medical-robots/autonomous-robot-surgeon-bests-human-surgeons-in-world-first. This was done on both intestine samples and also on living anesthetised pigs; the intestines repaired by robots had more consistent stitches and were more resistant to leaks. This is considered a breakthrough as the first time a robotic system has outperformed human surgeons on soft tissue repair work.

4. Quantum Computing in the Cloud
IBM launched a new cloud service to provide users everywhere with access to their 5-qubit quantum computer processor http://www-03.ibm.com/press/us/en/pressrelease/49661.wss. While the current system is not a universal quantum computer the main purpose here is to provide the scientific community with access to the system in order to help accelerate innovations in the field and discover new applications for the technology.

5. Latest Machine Learning Developments
Machine learning is well on the path to ubiquity with Qualcomm’s announcement of a new deep learning SDK for their Snapdragon 820 processors, which will find their way into many if not most mobile devices launched this year and next https://www.qualcomm.com/news/releases/2016/05/02/qualcomm-helps-make-your-mobile-devices-smarter-new-snapdragon-machine. Nvidia demonstrates a deep learning system that taught an autonomous car how to drive after watching humans drive for just 72 hours, and which continuously learns after deployment http://www.digitaltrends.com/cars/nvidia-gpu-driverless-car/. And another deep learning system suggests autonomous cars would respond to new situations in a similar manner to humans https://www.kuleuven.be/english/news/2016/machines-can-learn-to-respond-to-new-situations-like-human-beings-would.

6. Latest Rejuvenation Biotech Updates
FightAging! had a trio of interesting rejuvenation developments this week. First, another senolytic drug candidate working via a different mechanism to kill senescent cells https://www.fightaging.org/archives/2016/05/abt-737-is-another-new-senolytic-drug-candidate-working-via-bcl-w-bcl-xl-and-induced-apoptosis/. Second, a new study shows how macrophages repair broken and leaky blood vessels in the brain, suggesting another benefit of boosting macrophage activity https://www.fightaging.org/archives/2016/05/macrophages-repair-broken-capillaries-in-the-brain/. Finally, an interesting new drug that enhances autophagy in the cell, boosting cellular cleanup of damage and wastes https://www.fightaging.org/archives/2016/05/auten-67-as-an-example-of-an-autophagy-enhancing-drug-candidate/.

7. Bee Navigation for Drones
The latest improvements in computer models of how bees navigate the world using vision to detect movement in the world around them and avoiding collisions look set to provide superior navigation and object avoidance abilities to robots and autonomous drones http://www.sheffield.ac.uk/news/nr/bee-model-1.573018. The new work provides a system able to detect both motion direction and motion speed using the optical flow of the visual world around them. Definitely looking forward to seeing autonomous drones using this to zip effortlessly through a forest.

8. Modular Plug-n-Play Microfluidics & Lablets
Microfluidics takes a step forward with the development of standard modular components including microvalves and micropumps that can be reliably used plug-n-play style to develop and evolve different microfluidic circuits and applications http://www.research.a-star.edu.sg/research/7504/small-devices-make-a-big-impact. Drilling down to smaller scales electronic microlabs (lablets) measuring 140 micrometers can control specific chemical reactions with voltage codes designed to enable an interesting platform for chemical evolution http://rubin.rub.de/en/featured-topic-when-science-and-science-fiction-merge/electronic-micro-labs-control-chemical; this is a fascinating chemical synthesis and control platform.

9. Femtosecond Clock Synchronisation
DARPA has a femtosecond clock synchronisation project (QuASAR & PULSE) underway aimed at facilitating advanced applications in ultraprecise satellite formations and radio astronomy http://nextbigfuture.com/2016/05/femtosecond-clock-synchronization-for.html. The most recent work demonstrated the most accurate clock ever with uncertainty measuring 2 parts in 10^18, or 1 second in 14 billion years, giving 10,000 times better performance than existing GPS clocks, and with novel time and motion compensations in both free space and atmospheric conditions.

10. Flexible Holographic Displays
HoloFlex is a new flexible smartphone with a holographic display http://spectrum.ieee.org/tech-talk/consumer-electronics/gadgets/holoflex-a-flexible-smartphone-with-a-holographic-display. The display is a 1080p flexible OLED screen with a layer of 3D printed flexible lenses - over 16,000 individual lenses - that enable a holographic lightfield display and 3D glasses-free images over a 35 degree field of view. An interesting display platform to keep an eye on.

SciTech Tip Jar: http://www.scitechdigest.net/p/donate.html___

posted image

2016-05-07 07:04:27 (7 comments; 10 reshares; 29 +1s)Open 

The Psychology of Procrastination & The Fear of Not Trying

A couple of recent excellent TED talks discuss procrastination and related quirks of psychology. First, the main talk below The surprising habits of original thinkers by Adam Grant discusses the three key habits of original thinkers and the benefits of trying to learn from or imitate them:

1. Being late to the party, and the benefits of both learning from others and doing it better, and also procrastinating to drive creativity.
2. Feeling doubt and fear, but in the sense of doubt in ideas and not doubt in the self, and being more afraid of not trying than trying and failing.
3. Having lots and lots of bad ideas, because the greatest success tends to accrue to those who fail the most because they try the most.

Other key take-aways include (i) vuja de and the knack for looking at a common thing... more »

The Psychology of Procrastination & The Fear of Not Trying

A couple of recent excellent TED talks discuss procrastination and related quirks of psychology. First, the main talk below The surprising habits of original thinkers by Adam Grant discusses the three key habits of original thinkers and the benefits of trying to learn from or imitate them:

1. Being late to the party, and the benefits of both learning from others and doing it better, and also procrastinating to drive creativity.
2. Feeling doubt and fear, but in the sense of doubt in ideas and not doubt in the self, and being more afraid of not trying than trying and failing.
3. Having lots and lots of bad ideas, because the greatest success tends to accrue to those who fail the most because they try the most.

Other key take-aways include (i) vuja de and the knack for looking at a common thing and seeing it afresh with possibility, and (ii) our biggest regrets are not of our actions, but of our inactions.

The second, related, and excellent talk is Inside the mind of a procrastinator by Tim Urban of the WaitButWhy blog https://www.youtube.com/watch?v=arj7oStGLkU. The key take-away concerns identifying the two types of procrastination:

1. Procrastination when there are deadlines, which can generally be managed and dealt with.
2. Procrastination on things for which there are no deadlines, which are often ignored and chronically neglected by most people. This includes things such as developing a career, seeing family, exercising, taking care of personal health, working on relationships, and other important life goals. These commonly extend out forever via the typical “I’ll get around to it one day” delusional throw-away line.

One of the strategies to consider when you catch yourself procrastinating on type 2 goals that lack deadlines is simply to erect artificial deadlines. ___

posted image

2016-05-06 12:32:20 (5 comments; 10 reshares; 45 +1s)Open 

360 Degree SpaceX Rocket Landing

Just saw one of the best and most inspiring 360 degree videos yet: SpaceX rocket re-entry and landing on a drone ship at sea. Looking around the deck, then up at the rocket in the sky, and following it down as it gently lands on the deck . . . just wow. Gave me shivers.

And then I watched it again in tilt-your-head VR mode on mobile with Google Cardboard. Do yourself a favour and strap on a VR headset if you have one. Click and drag on desktop with your mouse if you don't :)

360 Degree SpaceX Rocket Landing

Just saw one of the best and most inspiring 360 degree videos yet: SpaceX rocket re-entry and landing on a drone ship at sea. Looking around the deck, then up at the rocket in the sky, and following it down as it gently lands on the deck . . . just wow. Gave me shivers.

And then I watched it again in tilt-your-head VR mode on mobile with Google Cardboard. Do yourself a favour and strap on a VR headset if you have one. Click and drag on desktop with your mouse if you don't :)___

posted image

2016-05-02 13:54:07 (2 comments; 4 reshares; 22 +1s)Open 

Only The Dead

Honest, brutal, and confronting. I watched this documentary by chance last night, never having heard about it or the main journalist involved in making it. Michael Ware (https://en.wikipedia.org/wiki/Michael_Ware) spent years in Iraq covering the war; he was there before it began, gained the trust of insurgents, was fed video and material direct from al-Zarqawi, toured with US Marines, lived in some of the most hostile cities in existence, and bore witness to the inane brutality of a very different kind of war. Invaluable as a perspective, narrative, and recent history I'd not seen before. 

Only The Dead

Honest, brutal, and confronting. I watched this documentary by chance last night, never having heard about it or the main journalist involved in making it. Michael Ware (https://en.wikipedia.org/wiki/Michael_Ware) spent years in Iraq covering the war; he was there before it began, gained the trust of insurgents, was fed video and material direct from al-Zarqawi, toured with US Marines, lived in some of the most hostile cities in existence, and bore witness to the inane brutality of a very different kind of war. Invaluable as a perspective, narrative, and recent history I'd not seen before. ___

posted image

2016-05-01 07:52:15 (4 comments; 17 reshares; 83 +1s)Open 

SciTech #ScienceSunday Digest - 18/2016.
Permalink here: http://www.scitechdigest.net/2016/05/exoplanet-imaging-dna-nanothermometers.html

Proteins per gene, Exoplanet imaging, DNA nanothermometers, Protein assembly tools, Versatile optogenetics, 3D printing robot spiders, Deep learning, Sensory prosthetic hand, Superfluid helium blackholes, Tissue regeneration.

1. One Gene, Many Proteins
It used to be thought that each gene encoded for and produced just one single protein; this latest ambitious study has blown that paradigm apart and made it very apparent that there is far more nuance and complexity here than first thought https://www.quantamagazine.org/20160426-one-gene-many-proteins/. This thorough study looked at 1,500 human genes, found how many produced multiple proteins, and ran binding studies to 15,000 other proteins to determine whether... more »

SciTech #ScienceSunday Digest - 18/2016.
Permalink here: http://www.scitechdigest.net/2016/05/exoplanet-imaging-dna-nanothermometers.html

Proteins per gene, Exoplanet imaging, DNA nanothermometers, Protein assembly tools, Versatile optogenetics, 3D printing robot spiders, Deep learning, Sensory prosthetic hand, Superfluid helium blackholes, Tissue regeneration.

1. One Gene, Many Proteins
It used to be thought that each gene encoded for and produced just one single protein; this latest ambitious study has blown that paradigm apart and made it very apparent that there is far more nuance and complexity here than first thought https://www.quantamagazine.org/20160426-one-gene-many-proteins/. This thorough study looked at 1,500 human genes, found how many produced multiple proteins, and ran binding studies to 15,000 other proteins to determine whether different proteins from the same gene shared the same or different functions; they generated surprisingly variable results. Different proteins can be formed from the same gene by combining different gene segments (exons) in different sequences. This will ideally be repeated for all genes and proteins. I think the take-away here is simply recognising such seemingly chaotic complexity as a measure of evolutionary robustness.

2. Imaging Exoplanets at 1km Resolution
A new proposal for a space-based telescope positioned 11 times further away than Pluto utilises the gravitational lensing of our Sun to achieve kilometer scale resolution of candidate exoplanets orbiting other stars in our local vicinity https://www.technologyreview.com/s/601331/a-space-mission-to-the-gravitational-focus-of-the-sun/. All you need is (i) a means to block out the Sun’s light, (ii) account for the Sun’s corona, (iii) improve pointing accuracy by an order of magnitude to 0.1 nanoradians, (iv) design a propulsion system able to account for orbital motion, (v) better software and optics to account for blurring, and (vi) filtering light from the planet’s parent star. Do this and you’ll get 10,000 times more light from the exoplanet. Sounds like a worthwhile project. I only wonder about such a telescope being overtaken by technological development during a lengthy commute to 600AU or so.

3. DNA Origami Nanothermometers
Specific DNA sequences can now be used to produce DNA origami structures that are programmed to function as nanothermometers http://www.nouvelles.umontreal.ca/udem-news/news/20160427-chemists-use-dna-to-build-the-worlds-tiniest-thermometer.html. The technique produces DNA structures that fold and unfold at very specific temperatures and adding optical reporter molecules results in 5nm wide structures that produce an easily-detectable signal as a function of temperature. Applications in intra-cellular biology, testing biological machines and enzyme “overheating”, and in nanoelectronics to measure temperatures of very small areas.

4. Advanced Protein-Based Tools
First, a new set of modular proteins has been designed and tested that can be customised to specifically bind arbitrary RNA sequences, and so allowing a versatile mechanism to control and image specific RNAs in the cell http://news.mit.edu/2016/controlling-rna-living-cells-0425. This is a versatile modular code for generating custom proteins able to bind specific RNA sequences from 6 to 18 bases long, with applications in future molecular assembly lines and precise measurements of how often RNA is being translated in the cell. Second, another engineered protein naturally self-assembles carbon buckminster fullerene molecules into ordered lattices and suggests a pathway to proteins able to organise nanomaterials by design http://www.eurekalert.org/pub_releases/2016-04/dc-rca042516.php.

5. More Versatile Optogenetics
Optogenetics is increasingly being used to control pain in test animals by using viruses to functionalise neurons responsible for conveying pain and sensation signals, and then using light - either implanted or in this case external to the skin for peripheral neurons - to turn pain transmission on or off in very localised and specific areas https://www.mcgill.ca/newsroom/channels/news/brighter-prospects-chronic-pain-260515. New optogenetics tools now also include the ability to be activated by red light that can penetrate deeper into tissues, and also be combined with other proteins and receptors to drive other cellular processes with light http://ist.ac.at/news-media/news/news-detail/article/red-light-controls-signaling-in-human-cells/6/.

6. 3D Printing with Robot Spiders
A new prototype 3D printing technology involves the use of robotic spiders able to move around with an in-built portable 3D printer, extruding plastic instead of silk in specific patterns to collaboratively build up printed structures - accuracy of localisation is a key hurdle https://www.technologyreview.com/s/601330/robot-spiders-weave-products-from-plastic-in-a-new-spin-on-3-d-printing/. In related 3D printing news the rise of custom-made, personalised, 3D printed medical implants is accelerating and increasing in sophistication https://www.technologyreview.com/s/601305/the-key-to-repairing-your-bones-may-come-out-of-a-printer/.

7. Interesting Deep Learning Developments
A new platform called OpenAI Gym has been launched as a toolkit for developing and comparing reinforcement learning algorithms for applications such as teaching agents to play games and navigate environments https://gym.openai.com/. Movidius has released a neural net accelerator called Fathom on a USB stick that uses only 1 watt of power to run powerful, typically computationally intensive image recognition neural networks with wide applications including allowing every robot to have cutting-edge vision capabilities http://spectrum.ieee.org/tech-talk/computing/embedded-systems/movidius-puts-neural-network-on-a-usb-stick. Meanwhile Drive.ai launched from Stanford’s AI Lab to test autonomous vehicle systems based on deep learning http://spectrum.ieee.org/cars-that-think/transportation/self-driving/driveai-brings-deep-learning-to-selfdriving-cars, and talking of autonomous vehicles self-driving trucks are really building momentum https://www.eutruckplatooning.com/News/495554.aspx?.

8. HAPTIX: The Prosthetic Hand that Can Feel
Here’s a good overview of efforts within DARPA’s HAPTIX program to develop prosthetic hands that allow amputees to regain a sense of touch and sensation, at least through some of the most recent prototypes http://spectrum.ieee.org/biomedical/bionics/creating-a-prosthetic-hand-that-can-feel. The latest prototypes incorporate direct neural interfaces that convey tactile sensory information from sensors located on the prosthetic hand, and resulting in the patient consciously perceiving sensations from those areas as if it were their own hand, and drastically improving sensitive tactile manipulation tasks from 43% to 93% success rate. Slowly getting towards a system that makes the person momentarily forget they lost the hand.

9. Blackholes, Superfluid Helium, & Phonons
New insights into the existence and behaviour of Hawking radiation at Blackhole event horizons are being made with related phenomena involving rapidly rotating superfluid helium and phonons http://www.sciencealert.com/physicists-have-created-a-black-hole-in-the-lab-and-it-could-finally-confirm-the-existence-of-hawking-radiation. The rapidly rotating superfluid helium forms a barrier through which sound waves should not be able to leave, yet the experiment detected phonons, small packets of sound wave energy, leaking out of this sonic blackhole as a sonic analogue to Hawking radiation leaking from a conventional blackhole. The work is undergoing peer review, confirmation, and debate.

10. Regeneration of Brain and Other Tissues
Recent experiments demonstrate that simply inserting a microneedle into the hippocampus of mice with Alzheimer’s Disease helps induce the hippocampus to regenerate, repair damage, and reduce the beta-amyloid plaques characteristic of the disease http://www.eurekalert.org/pub_releases/2016-04/ctco-mii042616.php. Another recent study demonstrates the use of a cocktail of nine different chemicals able to transform skin cells into beating heart or neural stem cells (different cocktail for each), that when transplanted into animals helped to regenerate damage and restore normal function to those organs https://www.sciencedaily.com/releases/2016/04/160428152117.htm.

SciTech Tip Jar: http://www.scitechdigest.net/p/donate.html
___

posted image

2016-04-28 14:17:10 (25 comments; 4 reshares; 45 +1s)Open 

Halo 5: The Clunky Compromised Crapshoot
An overdue rant.

I was running towards where I thought the fighting would be, across a huge expansive map hosting a 24-person multiplayer game when I just stopped, in the middle of a wide open area, and just stared at the screen. Conflicting thoughts swirled through my head as I felt dull, listless, and wondering why the fuck I was bothering. I recognised a profound feeling of boredom, in what was supposed to be a fast-paced next-gen first person shooter. I finally admitted to myself that Halo 5 was simply a shit game in so many obnoxious little ways. 343 Industries had achieved the seemingly impossible by actually managing to make a first person shooter that induced boredom in the player. I put the controller down, leaned over, and turned the console off.

Halo has been a big part of my life over the years and has been my go-to... more »

Halo 5: The Clunky Compromised Crapshoot
An overdue rant.

I was running towards where I thought the fighting would be, across a huge expansive map hosting a 24-person multiplayer game when I just stopped, in the middle of a wide open area, and just stared at the screen. Conflicting thoughts swirled through my head as I felt dull, listless, and wondering why the fuck I was bothering. I recognised a profound feeling of boredom, in what was supposed to be a fast-paced next-gen first person shooter. I finally admitted to myself that Halo 5 was simply a shit game in so many obnoxious little ways. 343 Industries had achieved the seemingly impossible by actually managing to make a first person shooter that induced boredom in the player. I put the controller down, leaned over, and turned the console off.

Halo has been a big part of my life over the years and has been my go-to game ever since the series first launched 15 years ago. I’ve spent uncounted hours in campaign and orders of magnitude more in multiplayer and enjoyed some of the most thrilling, hilarious, and socially enjoyable gaming of my life. No other game comes close to the hold this series and its idiosyncratic gameplay mechanic has held over me. I love the Grand Theft Auto series too, but my time spent playing in those worlds runs a very distant second to Halo.

The final installment on last generation consoles, on the Xbox 360, was Halo 4 and while I still managed to get a lot of hours in and enjoyment out of the game I did feel it was a step backwards for the series and for the core gameplay mechanic in general. Skill was still required, but less so than before, with random events and factors such as encountering opponents with unexpected weapons and armour abilities tending to introduce a large element of luck into encounters in which right place, right time / wrong place wrong time played a noticeable role in deciding outcomes rather than being able to use skill and cunning to make the best of most novel situations.

I’ve played many games on the Xbox One since it launched and in particular fell in love with a competitive first person shooter, Titanfall, for which I clocked up many enjoyable, thrilling, and thoroughly amusing hours. But I maintained a soft spot for Halo and eagerly awaited the release of the latest version - the first for the Xbox One - in 343 Industries’ much anticipated Halo 5. I hoped that 343 would learn from some of the things they did in Halo 4 and engineer Halo 5 to be a significant improvement.

Unfortunately, after clocking up a great many hours I’m left with the inescapable conclusion that they embraced Halo 4’s failings with gusto, massively expanded all the things Halo 4 suffered from and introduced a bunch of new mechanics that together ensure that Halo 5 is and will always be a clunky compromised crapshoot.

For the first time ever a Halo game actually drove me away. I went on hiatus a couple of times for a month or more, but came back to see if anything had changed or been fixed or just to berate myself for being so judgemental and to give it another chance. But over six months since first release of Halo 5 I can finally say with certainty that I will never play the game again.

I actually snapped the game disc in two in order to prevent me from ever even trying to play it again. As bad as it was Halo 5 had a funny addiction quality about it, taunting and luring me back to play it, and yet always leaving me frustrated, annoyed, and disappointed every time I played. I put this down to my history with the franchise and many years enjoyment causing me to keep hoping beyond hope that it would get better. To break this destructive and wasteful cycle the disc itself had to be destroyed.

Observations and General Frustrations

➤ Couldn’t play for 2 days after purchase. To even start playing the game the installation process requires a 9 GB download and for whatever reason, given the time I was able to download and the general strain on servers at launch it was over two days before it finally finished and I was finally able to start playing.

➤ Gameplay just feels clunky and constricted. Of course you’re portrayed as a supersoldier, but in reality have to run around as if you’re a fragile little petal that breaks in a slight breeze. Aside from that the aiming mechanic feels . . . off . . . and wrong for some reason. Some people put this down to the deadzones on the triggers being larger. It feels like you have a much narrower field of view than before, a boxed-in feeling, and much reduced situational awareness - all crucial factors for any FPS. Many people have said that you just need to get used to the new mechanic but this amounts to just easy excuses when the changes made to such a crucial aspect of gameplay were deliberate and worthy of criticism and complaint. Personally I suspect that these myopic changes to the fundamental feel and mechanics of the game were a result of compromises to ensure the game always ran at 60 frames per second; that the entire game was compromised just to achieve this silly and irrelevant technical feat.

➤ The radar or motion sensor has been completely changed and limited to the point of near uselessness. It is very short range and results in constant surprise and should either be turned off completely or restored to the previous Halo setting. What is worse/weird is that it works perfectly through floors and ceilings: somehow the “sensor technology” can’t detect enemies rushing directly towards you yet can perfectly sense those beneath a foot of concrete under you.

➤ Other gameplay annoyances include the vehicle boarding mechanic and hijacking zone making it harder and more random to tackle vehicles, the vehicles themselves suffer from pathetic physics resulting in a light and floaty feeling (worse than previous titles) and the propensity to roll at the drop of a hat, grenades are overpowered mini-nukes, the utterly stupid inclusion of gaps in levels that you can fall to your death through, melee kills after dying (serve lag?), difficulty in knowing from where and which direction you were killed from (especially from distance), and a mongoose quad bike that happens to pack insta-kill cannons. And the invisibility power-up perk is so stupidly overpowered that it beggars belief it was actually included in the game.

➤ Spawning is awful. You can literally spawn after dying and die again within 2 seconds. I’ve seen grenades land at people’s feet as they spawned. I’ve had people spawn right behind me and kill me.

➤ This is the first title to lack a split-screen option for 2 - 4 player action at home in the same room. Previously this feature alone granted the game a wonderfully social aspect, inspired a great many hours of laughter and fun times with friends. It’s absence is profoundly disappointing and is, I suspect, another compromise to eek out a guaranteed 60fps from the game.

➤ My preference for multiplayer gameplay is Big Team Battle (8 vs 8) games and yet the game launched without a BTB list at all. This was finally rectified 4 or so months later.

➤ Some playlists - typically the ones I preferred - were frequently empty or just refused to find me a game to play. This is an example of the game designers compromising on the time to get people into a game vs the risk of latency or lag while playing. But the fact that BTB very quickly became unplayable for me, and occasionally even conventional Team Slayer, refusing to find any players or join me into any game indicates that (i) the matchmaking mechanic is sorely broken or (ii) players had pretty much abandoned the game and the numbers available in the pool were just too low to make a game.

➤ Warzone mode is an epic crapshoot. The maps are too big and you can regularly encounter long stretches of time without finding any opposing players. REQ bonus points (for “purchasing” power weapons, vehicles, and upgrades) accrue to those who need these perks the least. Typically results in games regularly becoming terribly unbalanced that finish with one team double the score of the other; I’ve regularly been in games in which one team has dominating advantages of two tanks plus two banshees and the other team has nothing - at this point there is no point playing and quitting out of the game is the only sensible option. Another consequence is that Warzone results in no predictability for vehicles or weapons on maps, making it impossible to plan any meaningful strategy or tactics.

➤ Warzone maps themselves have generally poor layouts; they have terrible choke points (Stormbreak is the worst) that invariably and regularly result in pointless, repetitive, boring stalemates. But you’re damned if you do and damned if you don’t go for these points. Warzone intro and ending cinematic sequences are a terribly boring waste of time; novel and interesting the first three games and then the. same. boring. sequence. and voice. every. single. time. and. killing. the. same. NPC enemies. in. same. base. every. single. time.

➤ The pre-game lobby is crap. The only easily accessible information you have on players are emblems, which mean and convey nothing. Previous titles had player names, which you learned to recognise after a while and could judge good or poor players and also convey a more social aspect. Not anymore. There is also no indication of who is talking or playing music and no way to mute them. At least at first: 4 or so months after launch they finally introduced a mute option.

➤ I quit the campaign “story” after completing just two levels due to it being uninspired, boring, and generally insipid. It was the same old simplistic gameplay: run and shoot and jump through levels and kill fodder enemies and then kill bigger enemies and then wash, rinse, and repeat. Predictable boring AIs. The campaign possessed as much linear appeal as a platformer like Mario on the Super Nintendo but lacked the “richness” of storyline.

➤ At the end of the day Halo 5 is a crapshoot: incorporating random elements and features that make player performance and experience far more contingent on luck than skill compared to previously, and result in players less in control of situations. It is clunky: enforcing poor UI elements such as field of view and aiming mechanics. And it is compromised: pursuing 60fps and other goals at the expense of better more fluid UI and gameplay, local sociality, empty matchmaking lists, and player engagement.

Some people love Halo 5 however.

If you don’t have a fulltime job and play very regularly with a close team that you communicate constantly with via a headset to coordinate cohesive team strategies then Halo 5 works very well. But this is no longer a game for good, casual players looking to wind down and have some fun; for many, only frustration and boredom await. I can’t possibly see how this series can recover.

___

posted image

2016-04-28 11:12:39 (8 comments; 1 reshares; 22 +1s)Open 

Aggregate Stats Since Joining G+

I logged into Circle Count yesterday out of curiosity and found some surprising aggregate stats since first joining G+ in July 2011 nearly 5 years ago. Turns out I've made 958 public posts and written 459,673 words, which just blows my mind. Could I have written 3 or 4 books in that time? Other stats below include how many +1s, reshares, and comments I've received over that time.

http://www.circlecount.com/ 

Aggregate Stats Since Joining G+

I logged into Circle Count yesterday out of curiosity and found some surprising aggregate stats since first joining G+ in July 2011 nearly 5 years ago. Turns out I've made 958 public posts and written 459,673 words, which just blows my mind. Could I have written 3 or 4 books in that time? Other stats below include how many +1s, reshares, and comments I've received over that time.

http://www.circlecount.com/ ___

posted image

2016-04-24 11:17:48 (4 comments; 25 reshares; 81 +1s)Open 

SciTech #ScienceSunday Digest - 17/2016.
Permalink here: http://www.scitechdigest.net/2016/04/engineered-artificial-ribosomes.html

Engineered artificial ribosomes, Tantalising EmDrive, Nanomagnet Holograph displays, Hafnium oxide memristors, Rejuvenation genetherapy confirmed, Improved CRISPR, Seawater uranium extraction, Implantable ultrasound communications, Biomimicry brighter LEDs, Actin memory role.

1. Artificial Synthetic Ribosome
Ribosomes in cells take in genetic code and sequentially synthesise a complete protein from specific amino acids that correspond to that code. A new, autonomous molecular machine based DNA has been developed as an artificial synthetic ribosome able to take specific sequence instructions and make new synthetic polymer materials out of different molecules in a similar way... more »

SciTech #ScienceSunday Digest - 17/2016.
Permalink here: http://www.scitechdigest.net/2016/04/engineered-artificial-ribosomes.html

Engineered artificial ribosomes, Tantalising EmDrive, Nanomagnet Holograph displays, Hafnium oxide memristors, Rejuvenation genetherapy confirmed, Improved CRISPR, Seawater uranium extraction, Implantable ultrasound communications, Biomimicry brighter LEDs, Actin memory role.

1. Artificial Synthetic Ribosome
Ribosomes in cells take in genetic code and sequentially synthesise a complete protein from specific amino acids that correspond to that code. A new, autonomous molecular machine based DNA has been developed as an artificial synthetic ribosome able to take specific sequence instructions and make new synthetic polymer materials out of different molecules in a similar way http://www.rsc.org/chemistryworld/2016/04/ribosome-mimic-dna-molecular-machine-polymer-production. The full paper is also worth a look http://sci-hub.io/10.1038/nchem.2495. The system relies on (i) chemistry DNA hairpin sequences attached to specific reactant molecules, and (ii) instruction DNA hairpin molecules with sequences that control the order in which the different monomer reactant molecules are assembled. Autonomous, programmable, atomically precise synthesis of large molecules. In related news another engineered ribosome functions in bacteria to create proteins with beta-amino acids that has never been achieved in live cells before http://news.yale.edu/2016/04/21/new-protein-making-factory-promises-better-medicines.

2. EmDrive and Unruh Radiation
To date six independent experiments have replicated the original EmDrive results by building their own device and measuring a thrust from a hollow cone when microwaves are bounced inside it, all without a suitable explanation for a seeming violation of conservation of momentum. A new theory of inertia suggests inertia is the pressure Unruh radiation exerts on an accelerating body, an effect predicted by General Relativity, and which also explains the acceleration discrepanies in falyby anomalies https://www.technologyreview.com/s/601299/the-curious-link-between-the-fly-by-anomaly-and-the-impossible-emdrive-thruster/. The theory suggests inertia is quantised, predicts flyby anomalies and their discrepancy, predicts the magnitude of thrusts measured in all experiments done so far, and makes two predictions yet to be tested: (i) a dielectric in the cavity will enhance thrust, and (ii) changing cavity dimensions will reverse thrust. It also assumes photons have inertial mass and the speed of light changes in the cavity. Will be interesting to see where this goes.

3. Nanomagnet Pixels for Holographic Displays
Wide-angle 3D holographic displays have been developed that are powered by nanomagnets http://www.eurekalert.org/pub_releases/2016-04/tuot-npt041916.php. The nanomagnets in these displays are referred to as magneto-optic spatial light modulators; a laser focused on the surface defines the display pixel size, pixel switching occurs in 10 nanoseconds, with pixel size and pitch in this demonstration being 1 micrometer, and enabling a 30 degree viewing angle. A nice advance in spatial light modulators towards the goal of glasses-free holographic video displays.

4. Hafnium Oxide Memristors as Synapses
New memristors made of thin-film hafnium oxide and via conventional production processes show promise as artificial synapses in brain-like chips http://phys.org/news/2016-04-physicists-electronic-synapses-neural-networks.html. Demonstrations show the memristors reproducing electrical signalling behaviour as observed in biological synapses, including spike-timing-dependent plasticity, long-term potentiation, and long-term depression. Next step will be to incorporate these hardware prototypes into larger brain-like chips.

5. Rejuvenation Gene Therapy Confirmed Against Aging
BioViva has confirmed Liz Parish’s experimental gene therapy, undertaken last year for disabling myostatin and extending teleomeres, successfully extended the telomeres of the cells that were analysed (white blood cells) from 6.71kb to 7.33kb and so effectively removing an average of 20 years worth of telomere shortening http://bioviva-science.com/2016/04/21/first-gene-therapy-successful-against-human-aging/. As always an N=1 should be taken with a grain of salt; the company has received an injection of funding and repeatability will be key in convincing skeptics of the result. The same techniques could be used to target the Per2 gene, leading to rejuvenation of the mammalian immune system and prolonged lifespans http://www.alphagalileo.org/ViewItem.aspx?ItemId=163209&CultureCode=en.

6. Single DNA Base Read/Writes Via CRISPR
A newly developed version of CRISPR can be targeted to specific sites in the genome and instead of cutting the DNA, precisely change one DNA base for another in order to correct precise single-base mutations or misspellings https://www.statnews.com/2016/04/20/clever-crispr-advance-unveiled/. This again reduces the risk of off-site target effects and provides an elegant way to make edits, provided the group can further improve the tool to be capable of all 12 basepair swaps (currently does 2). A protein nanopore array has been used for real-time single-base electronic DNA sequencing http://www.eurekalert.org/pub_releases/2016-04/cuso-cet042116.php. The nanopores incorporate a DNA polymerase protein that synthesises a complementary DNA strand as the strand of interest is passed through the pore; each of the four different nucleotides added incorporates a distinct tag that facilitates clearer and more accurate electrical signals that allow the sequence to be reconstructed.

7. Improvements in Extracting Uranium from Seawater
There are lots of efforts to extract useful elements from seawater with Uranium being of particular interest given the oceans collectively hold 4 billion tons of Uranium. Significant advances are being made with seawater extraction of Uranium via novel adsorbent materials that can now achieve 5.2 grams of Uranium per kilogram of adsorbent after 49 days in seawater, and with more recent tests showing 6 grams after 56 days https://www.ornl.gov/news/advances-extracting-uranium-seawater-announced-special-issue. There would be many benefits to achieving economical Uranium (and other metals) extraction from seawater.

8. Ultrasonic Data Transmission Through Flesh
Ultrasonic signals can now be used to transmit data through meat at 30 mbps, enough for HD video http://spectrum.ieee.org/tech-talk/biomedical/devices/ultrasonic-signals-transmit-data-through-meat-at-hd-video-quality. Current implanted medical devices are usually limited to much lower data rates, but having wireless data transmission via ultrasound able to deliver HD video offers a number of benefits including live-streaming video from swallowed camera-pills, easily managing large firmware updates to implanted devices, and others. Next step is animal studies and confirmation of the effect of bone structures on data rates.

9. Biomimicry for Brighter LEDs
The amount of light emitted by LEDs was boosted by 60% by carefully shaping the outer surface of the LED to mimic the structure of a firefly’s lantern http://gizmodo.com/scientists-made-leds-60-percent-brighter-by-copying-fir-1771979185. This essentially amounts to forming a nanostructured lens on top of the LED and these structures reduce the difference in refractive index between the material body and air, allowing more light to be emitted. This opens up the possibility for more power-efficient LEDs, but I’m also wondering if they can flip it and improve the efficiency of photodetectors and photovoltaics?

10. The Role of Actin in Memory
New models suggest that actin, the protein that helps to control cell shape, is responsible for the formation of long-term memories http://news.rice.edu/2016/04/18/thanks-actin-for-the-memories/. This concerns studies of the energy landscape of proteins and how actin filaments pull upon and stabilise certain proteins to form longer, more stable, and insoluble prion-like fibers. This offers another piece of evidence for the biological role of prion-like proteins, and also suggests a healthy role for some types of protein aggregates in cells, which are often considered a sign of disease and malfunction. This provides a mechanism for synaptic structures to last many years, if not decades, although the transition from short-term memory to this form of long-term memory is not yet known.

SciTech Tip Jar: http://www.scitechdigest.net/p/donate.html ___

posted image

2016-04-17 08:26:20 (15 comments; 16 reshares; 68 +1s)Open 

SciTech #ScienceSunday Digest - 16/2016.
Permalink here: http://www.scitechdigest.net/2016/04/neocortical-sequence-memory-neural.html

Neocortical sequence memory, Neural bypass for movement, 3D printed ovaries, Human tissue textiles, Noisy CPUs, Smartphone satellite messaging, CRISPR edited mushrooms, Sticky wearable display, Full duplex radio, SpaceX ship landing.

1. Neocortical Sequence Memory
Numenta have presented a tantalising new theory of how networks of neurons in the neocortex learn sequences that explains (i) why neocortical neurons have thousands of synapses that are segregated onto different parts of the cell, and (ii) how neocortical column arrangement can form a powerful sequence memory http://numenta.com/press/numenta-researchers-discover-how-the-brain-learns-sequences.html. The new models are vastly different to current artificial... more »

SciTech #ScienceSunday Digest - 16/2016.
Permalink here: http://www.scitechdigest.net/2016/04/neocortical-sequence-memory-neural.html

Neocortical sequence memory, Neural bypass for movement, 3D printed ovaries, Human tissue textiles, Noisy CPUs, Smartphone satellite messaging, CRISPR edited mushrooms, Sticky wearable display, Full duplex radio, SpaceX ship landing.

1. Neocortical Sequence Memory
Numenta have presented a tantalising new theory of how networks of neurons in the neocortex learn sequences that explains (i) why neocortical neurons have thousands of synapses that are segregated onto different parts of the cell, and (ii) how neocortical column arrangement can form a powerful sequence memory http://numenta.com/press/numenta-researchers-discover-how-the-brain-learns-sequences.html. The new models are vastly different to current artificial neuron models and deep learning models, and exhibit powerful properties including high fault tolerance, continuous unsupervised learning, and the ability to learn complex sequences. Such models, if validated, will greatly accelerate the development of machine intelligence. In related news certain neural circuit mapping tools are now 20 times more powerful http://medicalxpress.com/news/2016-04-brain-tool-powerful-previous-version.html.

2. Neural Bypass Links Brain to Hand
A new neural bypass system uses a brain implant to record signals from the motor cortex, then decodes and sends these to a sleeve of electrodes that stimulate the appropriate forearm muscles, and can do this with an accuracy that allows the person with spinal injury to pick up a bottle or hit the right chord on guitar hero http://spectrum.ieee.org/the-human-os/biomedical/bionics/now-theres-a-way-around-paralysis-neural-bypass-links-brain-to-hand. With just 96 electrodes sampling 30,000 times per second, machine learning algorithms separate signal from noise in order to correctly encode the correct signals to be sent to the arm electrodes and muscles. With higher-density electrodes and better interfaces with distant muscle neurons we might be well on the way to providing natural, remote brain-body and brain-machine movement.

3. 3D Printed Ovaries Work in Mice
When new ovaries 3D printed with a new technique were implanted into mice they were able to successfully give birth to live healthy pups http://singularityhub.com/2016/04/10/this-3d-printed-prosthetic-ovary-restores-female-fertility-in-mice/. Comprised of a cross-linked biomaterial scaffold and seeded with ovarian follicles the implants were completely innervated by blood vessels, restored the animal’s hormone cycle, and produced eggs that could be fertilised. Human trials and therapies will commence once induced pluripotent stem cell technology can reliably produce the necessary oocytes and support cells needed to seed the scaffold.

4. Textile Techniques for Human Tissue Fabrication
Leading on from 3D printed ovaries, studies of different textile manufacturing techniques have determined which processes are ideal for engineering tissues needed for organ repair and implant http://munews.missouri.edu/news-releases/2016/0407-methods-used-to-create-textiles-also-could-help-manufacture-human-tissues-2/. The processes tested included the standard electrospinning technique, as well as meltblowing, spunbonding, and carding, all of which exceeded the cost-performance of electrospinning. The main driver for exploring these techniques is the requirement for industrial scale up of these engineered tissues, for example the need to produce 100s of meters of tissue material with consistent and reliable properties.

5. Engineering Noisy CPUs
DARPA-funded Singular Computing is producing CPU chips that deliberately perform mathematical operations incorrectly to a small extent, outputting 2.01 when performing 1 + 1 for example https://www.technologyreview.com/s/601263/why-a-chip-thats-bad-at-math-can-help-computers-tackle-harder-problems/. This fuzzy output turns out to be an asset when processing noisy, messy data and resulting in lower-power computation while software is still able to produce a useful result. Applications include radar imaging, 3D information from stereo photos, deep learning, and others. Demonstrations of object tracking in video performed 100 times faster on these error prone chips and consumed only 2% of the power of a conventional processor.

6. Smartphone Satellite Messaging
Higher Ground has developed a smartphone case called StaPaq that communicates to the phone via bluetooth while packing a satellite communications antenna that allows basic text messaging and email via satellite from remote locations lacking standard network connectivity http://spectrum.ieee.org/tech-talk/telecom/wireless/could-satellite-messaging-startup-higher-ground-bring-down-the-911-system. There are concerns with spectrum interference and antenna power density at close range, but there is a genuine use-case and benefits for people in remote areas, and also for remote sensors and intermittent reporting monitors as part of the Internet of Things. In related news Facebook unveiled ARIES, a wireless antenna array that can deliver data communications to devices dozens of miles away http://www.wired.com/2016/04/facebooks-massive-new-antennas-can-beam-internet-miles/.

7. CRISPR Edited Mushroom Available for Sale
Mushrooms whose genomes have been edited with CRISPR can be cultivated and sold without further regulatory interference http://www.nature.com/news/gene-edited-crispr-mushroom-escapes-us-regulation-1.19754. This is the first CRISPR-edited organism to receive such a green light; the mushroom was edited to disable 6 genes and this resulted in an increased resistance to browning. The mushroom made it through regulatory oversight because the only edits involved disabled genes and didn’t include the introduction of foreign or novel genes from other organisms.

8. Wearable Display Adheres to Skin
A skin-like microns-thick organic LED display can stick comfortably to the skin and display information from similarly wearable sensors http://spectrum.ieee.org/the-human-os/biomedical/devices/this-wearable-digital-display-just-sticks-on-your-skin. With further development one might imagine higher resolution displays capable of images and more dense information such as a replacement for a smartwatch as a form of temporary stick-on tattoo, a full-body version of which would be a pretty impressive / magical sight.

9. Low Power Full Duplex Radio
Building on work and prototypes first demonstrated last year a low-power full-duplex radio chip has been demonstrated that can transmit and receive signals on the same frequency at the same time using a single antenna http://spectrum.ieee.org/tech-talk/telecom/wireless/new-full-duplex-radio-chip-transmits-and-receives-wireless-signals-at-once. The promise is that mature rollout of such a chip would double the data capacity of existing technology. This breakthrough required a novel hardware solution involving the exact arrangement of transistors on the chip as well as an echo-cancelling receiver. The current chip should work for WiFi but will need to have its power capacity boosted to facilitate longer distance mobile communications.

10. SpaceX Lands Rocket on Drone Ship
At the start of the week we got to bear witness to SpaceX achieve yet another remarkable milestone in drastically reducing the cost to reach orbit by successfully launching a Falcon 9 rocket 200km into space, deploying an inflatable habitat in space, then re-entering the atmosphere and landing on a drone ship out at sea http://arstechnica.com/science/2016/04/like-a-boss-falcon-soars-into-space-and-lands-in-the-ocean/. Landing out at sea is more fuel efficient. If it passes firing tests the same rocket will be relaunched by June with the company offering a 30% price saving to customers.

SciTech Tip Jar: http://www.scitechdigest.net/p/donate.html ___

posted image

2016-04-16 02:50:41 (46 comments; 5 reshares; 25 +1s)Open 

Recent Politics

Perhaps the best, regular, and most engaging political coverage I’ve ever found bar none is Dave Rubin’s +The Rubin Report the YouTube channel of which has quickly become one of my favourites and a regular go-to source. I’d like to share why that is and also provide some insight into my current political views and how they came to be that way - the first such post I’ve done on politics in over 5 years. Decent chance I'll regret it. I agonised over which of Dave’s videos to share here but chose this one because (i) the call to action at the end inspired me to be more vocal, (ii) the critiques of the far left RE terrorism are applicable to critiques of the far left RE other things, and (iii) it’s just a great public statement in general.

Some of the things I absolutely love about the show include the incredible diversity of people he has on tointerview, ... more »

Recent Politics

Perhaps the best, regular, and most engaging political coverage I’ve ever found bar none is Dave Rubin’s +The Rubin Report the YouTube channel of which has quickly become one of my favourites and a regular go-to source. I’d like to share why that is and also provide some insight into my current political views and how they came to be that way - the first such post I’ve done on politics in over 5 years. Decent chance I'll regret it. I agonised over which of Dave’s videos to share here but chose this one because (i) the call to action at the end inspired me to be more vocal, (ii) the critiques of the far left RE terrorism are applicable to critiques of the far left RE other things, and (iii) it’s just a great public statement in general.

Some of the things I absolutely love about the show include the incredible diversity of people he has on to interview, the mostly balanced nature of the discussions, the fact that the guests are allowed to talk and say their piece, the exposure to different opinions and viewpoints that I wouldn’t otherwise encounter, the useful rhetoric and argumentative tools that you pick up, the fact that Dave helps a person open up and I find myself both agreeing with some of what they say and disagreeing with other bits, and that Dave is just a really likeable, honest guy who’d be cool to hang out and have a drink with. Although he is just as likely to prefer a joint instead.

It probably helps that my political alignment is similar to Dave’s in the sense of a classical liberal with libertarian leanings. Freedom of speech and expression is sacrosanct, fundamental, and the foundation upon which all we hold dear is built.

I first came across the channel when he interviewed Sam Harris. This was a while after the Ben Affleck affair on the Bill Maher show in which Affleck absolutely embarrassed himself with obnoxious, misleading, and utterly ludicrous attacks on Harris. I’m a fan and supporter of Harris and have long considered him one of the most rational, courageous, and carefully precise thinkers currently serving the public sphere. The interview with Dave is excellent and probably Harris’ best response to the Affleck affair and other matters.

Interestingly the Affleck abuse of Harris (like similar treatment by The Young Turks that I now can’t tolerate), a well-intentioned liberal attacking a well-grounded intellectual liberal, marked the first time I became acutely aware of a stark and ugly division in the left, something that had apparently been building for years.

A lot of the focus of the show over the past 6 months has been to hammer home and expose issues to do with the radicalised far left, which is replete with special interest groups who espouse inherently authoritarian and fascist ideologies but paradoxically believe they are progressive. Nothing could be further from the truth of course and Rubin recasts these efforts with a more suitable term first used by Maajid Nawaz, that of the Regressive Left, and making a very good case for political horseshoe theory with the extreme left and extreme right having far more in common with each other (like authoritarianism) than anyone else along the spectrum including classical liberals, libertarians, or other moderates.

Some examples of regressive left, fascist ideologies include things like third wave feminism, social justice warriors, Islamophobia, rape culture, modern identity politics, safe spaces, trigger warnings, microaggressions, and political correctness in general, which is perhaps the foundation for them all.

On political correctness the late George Carlin said it best: Political correctness is fascism pretending to be manners, https://www.youtube.com/watch?v=hkhUivqzWv0

What these movements, these ideologies all have in common is the authoritarian desire to control, censor, and block free speech, while sharing rhetoric that involves rampant use of logical fallacies including anecdotal evidence, black and white examples, cherry-picked data, and most commonly an appeal to emotion used to drive a baseless culture of outrage and offense. At its worst this results in a culture of shaming and witch-hunting, and a deplorable infantilisation of both our youth and our discourse in general. Such ideologies serve only to make society less equal, and less equitable while boosting societal divisiveness, bizarrely threatening to reverse the hard-won progress we’ve collectively made previously in making everyone more equal and more equitable.

In general everyone along the spectrum has good hearts and means well; everyone wants a better world. Yet as the regressive left seeks to censor debate and reach for that elusive pseudo moral superiority with which to bash everyone else with, it instead grasps fascism in all its intolerant glory. What annoys me most at times with this growing facet of our political spectrum are things such as a denial of biology, a confirmation bias that recasts everything to fit the narrative, and the unshakable faith-based conviction that all who disagree are bigots that must be silenced. If nothing else it has opened my mind to the moral and intellectual bankruptcy of moral and cultural relativism. Few things sadden me as much as seeing free speech blocked, censored, and silenced on University campuses, those supposed bastions of the enlightenment.

Some memorable interviews include:
Note all interviews are also available in shorter sections addressing specific topics.

Sam Harris
On Affleck fallout, Islam, and being continually misrepresented
https://www.youtube.com/watch?v=zQqxlzHJrU0

Christina Hoff Sommers
One of the most prominent feminist academics on the planet
https://www.youtube.com/watch?v=2RNaspc5Ep4

Mike Cernovich
Convinced me that Trump is a good bet for President over Hillary purely on establishment shake-up grounds. Although Sanders is still my first choice and I hope he succeeds in his bid.
https://www.youtube.com/watch?v=K03gRy6qjKY

Milo Yianopolous
Provocateur; but I tend to agree with his message and methods
https://www.youtube.com/watch?v=1FvADt-mJ_o and
https://www.youtube.com/watch?v=DiA0P9iELAA

Ben Shapiro
Confrontingly conservative, hammers the irrelevancy of feelings in public / social policy decisions
https://www.youtube.com/watch?v=3Qrlnn35gBo

John McCain
Presented surprisingly well and articulately
https://www.youtube.com/watch?v=JSB8CYXcchw

Gad Saad
Jovial evolutionary psychologist and liberal political commentator
https://www.youtube.com/watch?v=Jq1utGg0PUY

Douglas Murray
Political commentator on free speech and Islam
https://www.youtube.com/watch?v=oX914A6dbbs

All of this was playing in the back of my mind recently while reading Peter Turchin’s excellent book Ultrasociety concerning the rise of cultures, states, and societal institutions that are ever-more cooperative, ever-more equitable, ever-more trusting, productive, and wealthy. And all driven by competition between different cultural groups that has, historically, been unfortunately characterised by war. The key passages that come to mind here are those concerning the common features of empires (that their dominant cultures) that fall: (i) enrichment of elites, (ii) rampant inequality in wealth, and (iii) rising demands of special interest groups - all of which contributes to the evaporation of trust, a decline in cooperation and the weakening of the state and culture that helped make the existence of productive wealth and special interests even possible in the first place.

I fear that there are real parallels here to our current situation, that the rise of special interest groups, small distinct subcultures that don’t integrate with and are divisively toxic to the broader culture that supports them, are like tumours slowly festering and sending out toxic signalling molecules to the surrounding support tissue. Hyperbole? Sure. But in this view the rising influence of far left fascist interest groups are as potentially damaging to our society’s future well being as are the depredations of extreme elite enrichment and inequality. Dealing with this intolerance is becoming doubly important given the growing technological pressures (such as pervasive surveillance) for our society to develop values and institutions that boost wisdom and enshrine forgiveness and tolerance.

To finish off this part Rubin Report recommendation, part rant that some are sure to judge me by, I’ll share a good recent piece by Status 451, All Aboard The Shame Train, that overviews some of the recent growing intolerance and fascism that I’ve alluded to: https://status451.com/2016/04/05/all-aboard-the-shame-train/

For now, I'm going back to the science and tech I love so dearly.

Edit: Clarification over Sanders / Trump confusion___

posted image

2016-04-14 05:15:08 (4 comments; 4 reshares; 18 +1s)Open 

Just attended the Rodney Brooks Public Lecture: The Future of Robotics and AI

Rodney Brooks, of Roomba, Rethink Robotics, Baxter, Sawyer fame gave a public lecture at Flinders University this afternoon and I was lucky enough to attend. Rodney received his Bachelors degree from Flinders before travelling to the USA to complete a PhD and launch his stellar career, and still has family in Adelaide.

Sawyer's design was influenced by Apple apparently: Apple dictates that employees at Chinese factories where iPhones are made must have a 2-foot by 2-foot by 5-foot workspace, and so Sawyer was designed to fit that workspace.

He enjoyed the latest The Martian movie because the little robot he worked on Sojourner made an appearance o_0

Rodney believe's that the future is one of human-machine partnership; 600,000 manufacturing jobs in USA remain... more »

Just attended the Rodney Brooks Public Lecture: The Future of Robotics and AI

Rodney Brooks, of Roomba, Rethink Robotics, Baxter, Sawyer fame gave a public lecture at Flinders University this afternoon and I was lucky enough to attend. Rodney received his Bachelors degree from Flinders before travelling to the USA to complete a PhD and launch his stellar career, and still has family in Adelaide.

Sawyer's design was influenced by Apple apparently: Apple dictates that employees at Chinese factories where iPhones are made must have a 2-foot by 2-foot by 5-foot workspace, and so Sawyer was designed to fit that workspace.

He enjoyed the latest The Martian movie because the little robot he worked on Sojourner made an appearance o_0

Rodney believe's that the future is one of human-machine partnership; 600,000 manufacturing jobs in USA remain unfilled and are perfect for robots so they won't take away jobs (personally I wasn't convinced). His projections for the future are a bit conservative with regard to robot roll-out and capability; key problems that need solving are Mobility (difficult terrain), Messiness (object mapping and recognition), and Manipulation (better hands with Mechanics, Sensors, Algorithms, and Materials). Wasn't that enamoured with deep learning advances and thinks they are a long long way from human capabilities of generalisation.

My contribution: At the end of the talk someone asked about his thoughts on nanobots, Rodney started by saying "Ever since that guy . . . you know . . . engines of creation . . . " and I shouted out "Drexler" . . . "Yeah that's the guy, Eric Drexler . . ." Rodney is actually more optimistic for the next 20 years RE nanobots given recent demonstrations than he had been for the previous 20 years; science and tech advances appear to have won him over.
___

posted image

2016-04-12 13:52:17 (71 comments; 16 reshares; 70 +1s)Open 

Ultrasociety: How 10,000 Years of War Made Humans the Greatest Cooperators on Earth by Peter Turchin

This book changed my conception of war, culture, society, cooperation, competition, religion, and memes - to a lesser or greater extent.

The main thrust of Ultrasociety concerns support of new theories of evolutionary group selection to explain the rise and nature of the complex societies we currently find ourselves in, while challenging the standard criticisms of conventional evolutionary group selection. While genetics is of course mentioned, these advanced theories of group selection are based on Multilevel Cultural Selection, the selection of cultural traits, of memes or memeplexes, that lead to the evolution of ever-larger groups better able to compete, better able to cooperate, better able to produce more resources and more equitable societies. And this selection for the... more »

Ultrasociety: How 10,000 Years of War Made Humans the Greatest Cooperators on Earth by Peter Turchin

This book changed my conception of war, culture, society, cooperation, competition, religion, and memes - to a lesser or greater extent.

The main thrust of Ultrasociety concerns support of new theories of evolutionary group selection to explain the rise and nature of the complex societies we currently find ourselves in, while challenging the standard criticisms of conventional evolutionary group selection. While genetics is of course mentioned, these advanced theories of group selection are based on Multilevel Cultural Selection, the selection of cultural traits, of memes or memeplexes, that lead to the evolution of ever-larger groups better able to compete, better able to cooperate, better able to produce more resources and more equitable societies. And this selection for the large, complex, and equitable societies we now have was driven by war. The arguments, supporting historical data, and predictions are incredibly compelling and provide profound insights and a new and very useful way of looking at the world we live in and the direction we are headed.

I was incredibly fortunate to stumble across this book and have the opportunity to distill and consider its contents.

There are interesting and at first counterintuitive lessons here, such as the most important technology invented by our species after language was ranged, or thrown, weapons. This is because ranged weapons levelled physical ability and prowess, allowed dominant alpha males to be dispatched much more safely, and by doing so empowered the development of group egalitarianism and more cooperative groups of early humans.

The key and most important lesson however is that within-group competition destroys cooperation within the group, making it less cohesive and more likely to fall to another group, and between-group competition enhances cooperation within the group, boosting cohesion and making the group more likely to resist or destroy another group. Groups that develop rules, values, and institutions that boost cooperation within the group, that develop cooperative cultures, will overwhelmingly tend to outcompete those that do not or those that do so poorly. Cooperation is the key to success; a large group will always prevail over a small group due to sheer numbers and basic logistical superiority enabling terrifying staying-power.

Despite being against war, and making a case for war itself trending to extinction, Ultrasociety tends to suggest that an appropriate rephrasing of Bruce Springsteen’s classic lyrics would be: “War, what is it good for? Absolutely nothing everything we cherish and hold dear.”

Cultural selection as presented even makes a case for the value of organised religion. War and conflict between groups can drive the creation of large, complex states of millions of people but within-group cooperation at this scale is inherently fragile. Monotheistic religions arose because they served as a wonderful binder and glue to enforce and maintain cooperation during times of peace, helping to create institutions that facilitated trust between people. In small groups everyone knows and watches everyone else; in big groups with religion everyone is known by and watched by a God.

Looking forward, with religion having served its purpose and in the obvious absence of gods, I wonder what institutions we will create to fill this gap? Will we welcome pervasive transparent surveillance or will we be able to adequately implement blockchain technology to replace our ailing institutions and guarantee an even stronger level of trust between people? The lessons from Ultrasociety are also important for those considering or running a business, or marshalling any collective group towards a worthwhile goal.

Perhaps most provocatively, Ultrasociety makes a compelling, if unstated case against the philosophies of moral and cultural relativism. Not all cultures are equal; some cultures are better than others, and this can be measured empirically. Those cultures falling behind would be wise to adopt and improve upon the institutions of those more successful cultures if they are to avoid extinction.

Selected excerpts:

The theory of cultural multilevel selection says that this evolution is only possible when societies compete against each other, so that those lacking the right institutions fail. The costly institutions of complex societies manage to spread and propagate because the societies that possess them destroy those that don’t.

It was competition and conflict between human groups that drove the transformation of small bands of hunter-gatherers into huge nation-states. Not to put too fine a point on it, it was war that first created despotic, archaic states and then destroyed them, replacing them with better, more equal societies. War both destroys and creates. War is a force of destructive creation, a terrible means to a remarkable end.

Nevertheless, this brutal, murderous force can also be creative. By eliminating poorly coordinated, uncooperative, and dysfunctional states it creates more cooperative, more peaceful, and more affluent ones.

Here’s how war serves to weed out societies that “go bad.” When discipline, imposed by the need to survive conflict, gets relaxed, societies lose their ability to cooperate. A reactionary catchphrase of the 1970s used to go, “what this generation needs is a war,” a deplorable sentiment but one that in terms of cultural evolution might sometimes have a germ of cold logic. At any rate, there is a pattern that we see recurring throughout history, when a successful empire expands its borders so far that it becomes the biggest kid on the block. When survival is no longer at stake, selfish elites and other special interest groups capture the political agenda. The spirit that “we are all in the same boat” disappears and is replaced by a “winner take all” mentality. As the elites enrich themselves, the rest of the population is increasingly impoverished. Rampant inequality of wealth further corrodes cooperation. Beyond a certain point a formerly great empire becomes so dysfunctional that smaller, more cohesive neighbors begin tearing it apart. Eventually the capacity for cooperation declines to such a low level that barbarians can strike at the very heart of the empire without encountering significant resistance. But barbarians at the gate are not the real cause of imperial collapse. They are a consequence of the failure to sustain social cooperation. As the British historian Arnold Toynbee said, great civilizations are not murdered — they die by suicide.

The 30 years in America since about 1985 were a giant social experiment. What would happen if ideologies extolling extreme individualism and elevating self-interest as the sole basis on which to organize society were to gain the upper hand? The results are in: a decline of social cooperation at all levels of American society, resulting in a decreased ability to get the job done.

Complex human societies, including our own, are fragile. They are held together by an invisible web of mutual trust and social cooperation. This web can fray easily, resulting in growing social dysfunction. When cooperation is lost, a typical result is a wave of political instability and internal conflict and, in extreme cases, outright social collapse.

Social theorists have a name for smart people motivated solely by greed and fear — “rational agents.” It turns out that a group consisting entirely of rational agents is incapable of cooperation. In particular, such people will never manage to put together a fighting troop.

In fact, no matter what others do, a rational agent’s best course of action is always to defect. In a tribe of rational agents, all will feel this way, and therefore none will go out to meet the enemy. They’ll all pretend to be sick — until they are dragged out of bed and killed by their enemies.

Although societies differ in their tolerance of inequality (especially if it’s justified by high performance), there is always a point beyond which unequal division of rewards ceases to seem legitimate. When people feel that they are not getting their fair share, they begin to withdraw their cooperation. In a baseball team in which one player — the superstar — earns 10 times as much as his mates, the other players begin to slack off. As a result, baseball teams with highly unequal payrolls win fewer games than teams in which rewards are distributed more equitably. This is despite the fact that the more unequal teams have extremely strong players.

Teamwork pays. A single hunter can spend a lot of effort chasing down a rabbit, and get only a pound or two of meat to show for it at the end of the day. When a team of hunters brings down a buffalo, they will divide among themselves close to a thousand pounds of meat—perhaps a hundred pounds each. Economists call this kind of arithmetic “increasing returns to scale” — when a group working together can significantly increase each individual’s payoff, compared with what they would get working on their own.

In particular, high-trust societies tend to be more successful — better governed, more economically productive, simply nicer places to live and to visit.

Natural selection can simultaneously act on individuals within groups, and on whole groups. Within each tribe, cowards do better than brave men, on average increasing every generation. But at the same time, cowardly tribes are eliminated by courageous ones. Which of these processes will be stronger depends on many details: just how great is the cost of bravery?

Generally speaking, the capacity for culture should evolve (assuming that such pre-adaptations as sophisticated cognitive abilities are in place) when the environment changes too fast for genetic adaptation to work, but slowly enough for information accumulated by previous generations to be useful. If environmental change is faster than that, you are better off learning everything yourself, even though it’s risky and inefficient. Some aspects of the modern environment are changing very rapidly indeed.

As a corollary, while competition between teams creates cooperation, competition among players within a team destroys it. In other words, to succeed, cooperative groups must suppress internal competition. Equality of group members is, therefore, a very important factor in promoting group cohesion and cooperation, which translates into the capacity of the group to win against other groups.

Remember, the raw stuff of evolution is variation. When different teams, firms, ethnic groups, or whole societies are allowed (even encouraged) to experiment with different ways of doing things, it becomes possible to see what works best. Then best practices can be selected, either by the process of blind evolution, or by conscious choice. However, after that it would be a mistake to force everybody to do things the same way, because that would stop evolution in its tracks. You never know if there’s an even better solution just around the corner — or a danger that your training never prepared you for.

A coalition of punishers armed with stones (remember that stoning is one of the most ancient forms of capital punishment) or, better, with spears and bows and arrows, can easily dispatch an upstart with little risk to themselves. The killing power of projectile weapons is what made men equal, and drove the evolution of egalitarianism, thanks to our collective ability to control and subdue aggressive, physically powerful males.

This effect is known as Lanchester’s Square Law, because during each round of engagement, the proportion of casualties inflicted by an army on its adversary is the square of its numerical advantage. The moral of this mathematical digression is that, on flat plains, with warriors using projectile weapons, any numerical superiority that an army can achieve over its enemy is magnified out of all proportion. In other words, Lanchester’s Square Law yields an enormous return to social scale.

This means that under conditions of intense warfare and a real existential threat to groups that are defeated, we should expect a strong selection not only for larger size, but also for effective military hierarchies. However, the more effective a military hierarchy is, the more power it has. There is a principle in Sociology known as the Iron Law of Oligarchy. It says that all forms of organization, regardless of how democratic or autocratic they may be at the start, will eventually and inevitably develop into oligarchies. Power corrupts.

Why was cultural group selection the key to the transition from forager to farmer? Because you cannot switch to farming when everybody else in your community is foraging. The whole group needs to shift together. It requires a new set of cultural norms and institutions shared by all. The most important such institution would have been property rights over the food that you have grown.

The logic of cultural group selection also explains why agriculture was adopted in spite of its huge health costs. Groups of poorly nourished — perhaps even chronically sick — farmers were able to exterminate healthy and tall foragers simply by force of numbers. So individual fitness (both in the evolutionary sense and in the everyday sense of physical condition) declined, but evolutionary group fitness increased, and that is what drove the whole process.

Just as cows breed more cows, wealth breeds more wealth. The Matthew Principle means that economic inequality always increases. Short of a destructive war or a revolution that expropriates from the rich, economic inequality can only be kept in check by some kind of periodic redistribution, such as progressive taxes on wealth and inheritance.

Around 2,500 years ago, we see qualitatively new forms of social organization — the larger and more durable Axial mega-empires that employed new forms of legitimation of political power. The new sources of this legitimacy were the Axial religions, or more broadly ideologies, such as Zoroastrianism, Buddhism, and Confucianism (and later Christianity and Islam). During this time, gods evolved from capricious projections of human desire (who as often as not squabbled among themselves) into transcendental moralizers concerned above all with prosocial behavior by all, including the rulers. The Axial religions introduced several innovations that enabled post-Axial states to increase the scale of social cooperation. In other words, universal religions expand the circle of cooperation beyond the ethnolinguistic group; they work as a glue that holds together diverse groups in multiethnic empires.

So large groups in which belief in a moralistic, all-knowing punisher became rooted would be more cooperative than the atheistic ones. In small-scale societies, people behaved prosocially because they were being watched by acquaintances and neighbors. In large-scale anonymous societies they had to be good because gods watched them.

In Norenzayan’s words, “watched people are nice people.” It doesn’t matter whether the watchers are your friends and neighbors or supernatural beings (or even “Big Brother,” as in our modern societies). As long as people are watched, they behave nicely. And groups of people who behave nicely to each other win over groups that don’t. Sincere belief in supernatural moralistic punishers is particularly important because of the way it can restrain the powerful. A monarch may not care very much what peasants think of him, but he would think twice before crossing an all-knowing omnipotent god.

Available here: http://www.amazon.com/Ultrasociety-Years-Humans-Greatest-Cooperators-ebook/dp/B0185P69LU

The Secret of Our Success by Joseph Henrich also looks good and along similar lines to Ultrasociety.
___

posted image

2016-04-10 08:50:02 (11 comments; 19 reshares; 73 +1s)Open 

SciTech #ScienceSunday Digest - 15/2016.
Permalink here: http://www.scitechdigest.net/2016/04/gut-bacteria-therapeutics-large.html

Gut bacteria therapeutics, Large cortical maps, Metal foam applications, Plastic proteins, Stem cell regeneration, Nanocrystal ink transistors, Elegant brain navigation, Harmonic gear improvements, Smartphone blood tests, Carbon nanotube advances.

1. Reprogrammed Gut Bacteria Therapeutics
Synologic is seeking to launch a range of living therapeutics comprising engineered gut microbes http://news.mit.edu/2016/startup-synlogic-reprogramming-gut-bacteria-living-therapeutics-0405. The bacteria are intended to correct metabolic disorders that cause major diseases by providing augmented metabolic capabilities in the gut or otherwise complement functionality that may have been lost in other organs. These are metabolic... more »

SciTech #ScienceSunday Digest - 15/2016.
Permalink here: http://www.scitechdigest.net/2016/04/gut-bacteria-therapeutics-large.html

Gut bacteria therapeutics, Large cortical maps, Metal foam applications, Plastic proteins, Stem cell regeneration, Nanocrystal ink transistors, Elegant brain navigation, Harmonic gear improvements, Smartphone blood tests, Carbon nanotube advances.

1. Reprogrammed Gut Bacteria Therapeutics
Synologic is seeking to launch a range of living therapeutics comprising engineered gut microbes http://news.mit.edu/2016/startup-synlogic-reprogramming-gut-bacteria-living-therapeutics-0405. The bacteria are intended to correct metabolic disorders that cause major diseases by providing augmented metabolic capabilities in the gut or otherwise complement functionality that may have been lost in other organs. These are metabolic thermostats, regulating levels of ammonia or amino acids in the first examples, and seeking to treat things like inflammatory bowel disease in future, and possibly even things like lactose and gluten intolerance. Such a platform might be used not only for disease treatment but enhancement too. This is especially interesting given recent studies demonstrating gut bacteria regulating brain structure and function https://www.theguardian.com/science/neurophilosophy/2016/apr/05/gut-bacteria-brain-myelin and suggesting gut bacteria modification might actually impact intelligence.

2. Largest Network of Cortical Neurons Mapped
The largest network map of connections between neurons in the cortex has been published, another important milestone in the field of connectomics https://www.alleninstitute.org/what-we-do/brain-science/news-press/press-releases/research-largest-network-cortical-neurons-date-published-nature. This work is increasingly building tools to reverse engineer the brain and discover relationships between circuit wiring and neuronal and network computations, identifying modular architectures and functionally specific connectivity between neurons. This work comprised a 100 terabyte 3D data set, specifically analysing neurons in the visual cortex that responded to particular stimuli such as horizontal bars on a screen.

3. Composite Metal Foam Applications
The development of composite metal foams has developed in interesting directions, with the latest demonstration showing a 1-inch thick composite metal foam easily withstanding the impact from a high-power armour-piercing bullet, indeed the bullet turns into dust on impact - see the video https://news.ncsu.edu/2016/04/metal-foam-tough-2016/. In addition these metal foams are significantly lighter than normal metal plate, effectively shield x-rays, gamma rays, and neutron radiation, and handle fire and heat twice as well as bulk metals.

4. Versatile Plastic Proteins
Cyanuric chloride is a molecule being used as the base for a platform materials technology able to create a huge range of synthetic plastic proteins http://www.pnnl.gov/news/release.aspx?id=4266. This base molecule is used to produce a wide range of monomers with different side chains (functional groups) like amino acids, the monomers can be linked together in specific sequences as desired, and the chemistry facilitates predictable non-covalent bonds able to fold the polymer into desired shapes (like a protein). The benefits are the ability to create a much larger library of stable monomers, and potentially longer, larger polymer chains and more diverse folded shapes with novel functions - even replacing or mimicking protein drugs if needed. The benefits are these materials are much more stable, longer-lived, and resistant to protein-type degradation than normal proteins.

5. Next Generation Stem Cell Tissue Regeneration
A new stem cell technique reprograms bone and fat cells into induced multipotent stem cells (iMS) that have the ability to regenerate multiple tissue types in mice http://newsroom.unsw.edu.au/news/health/medical-scientists-develop-%E2%80%98game-changing%E2%80%99-stem-cell-repair-system. Once human iMS cells have been demonstrated as safe and effective in mice then human trials with the new cells may begin in 2017. The transformation procedure seems quite straightforward, and involves treating the cells with the compound 5-Azacytidine and platelet-derived growth factor. This looks fascinating for regenerating and repairing damaged or diseased tissues, but I wonder if it will also rejuvenate simply aged tissues? In related news the genes driving natural regeneration in animals have been mapped and counterparts have been found in humans https://today.duke.edu/2016/04/genetrees.

6. Nanocrystal Ink Transistors
Different spherical nanoparticles with the necessary electrical properties were dispersed in liquids to make inks, and a library of four of these inks can be used to lay down precise patterns to form transistors and other electrical components on flexible surfaces https://news.upenn.edu/news/penn-engineers-develop-first-transistors-made-entirely-nanocrystal-inks. This is a low-power, low-temperature fabrication technique and a good example of additive manufacturing process for creating logic gates, integrated circuits, and other electronics. It is also nicely modular as individual inks and nanocrystals can be independently developed and optimised to achieve better results. In related news precisely layered quantum dots enhance light-to-current energy conversion https://www.bnl.gov/newsroom/news.php?a=11829.

7. Brain Location Tracking with Natural Logarithms
It has been known that brains use grid cells to keep track of their physical location, but the mechanism by which grid cells encode and decode this information has now been deciphered for the first time http://www.upenn.edu/spotlights/penn-researchers-theory-brains-location-tracking-cells-use-transcendental-number-system. Grid cells use a transcendental number system based on the mathematical constant ‘e’ with different sized grids acting as the equivalent of tens, hundreds, and thousands in a decimal number, and with different sized grids found in ratios of ‘e’. To find ‘e’ lurking in the brain’s codes like this is pretty powerful and I wonder where else in the brain it might be found; it turns out that this is the most efficient way to encode the most amount of information with the least amount of cells and so it shouldn’t be too surprising the evolution figured this out a long time ago. This system works for animals navigating a predominantly 2D environment, but for those navigating 3D you only need a slightly different factor of ‘e’ to achieve the same.

8. Improvements for Harmonic Gears
Harmonic gears have the same size, weight, and form factor regardless of gear ratio, allowing flexible design revisions and reduction ratios of 30:1 to 160:1 being common; as a bonus developments in materials technology should allow harmonic gears to maintain their exceptional torque capacity while realising a 30% reduction in weight http://nextbigfuture.com/2016/04/stronger-metals-will-make-lighter-and.html. Prime benefits and applications here appear to be in robotics and prosthetics, for which you might want this analogue self-powered smart skin http://www.nanowerk.com/spotlight/spotid=43084.php.

9. Portable, Personal, Smartphone, Blood Testing
Cor is launching a product to bring blood chemistry testing via spectroscopy to the home http://www.gizmag.com/cor-blood-chemistry-health-tracker/42608/. The system uses (i) single-use cartridges that have a fine needle to take surface level blood samples, (ii) a reader unit then takes the cartridge and analyses the sample via vibrational spectroscopy, (iii) data is sent to the cloud for processing, (iv) results are returned to the person via a smartphone app. The first tests will include cholesterol, glucose, fibrinogen, and triglycerides. It is great to see this space increasingly heating up and I’m sorely tempted to back Cor’s Indiegogo campaign.

10. Latest Interesting Carbon Nanotube Tech
First, wafer-scale uniformly-aligned high-density carbon nanotube films are now being produced, which can be patterned by standard photolithography methods, and which have produced transistors, LEDs, photodetectors, and polarizers http://spectrum.ieee.org/nanoclast/semiconductors/materials/wafer-scale-nanotube-film-is-finally-here. Second, carbon nanotubes have been used as nanoreactors to produce long (micrometer), stable one-dimensional carbon chains known as carbyne within the nanotubes; carbyne’s mechanical properties exceed all known materials http://phys.org/news/2016-04-proof-stable-ultra-long-1d-carbon.html. Finally, carbon nanotubes have been confirmed as the fastest proton conductors ever discovered http://phys.org/news/2016-04-tiny-tubes-fast-lane.html.

SciTech Tip Jar: http://www.scitechdigest.net/p/donate.html___

posted image

2016-04-07 14:56:17 (16 comments; 21 reshares; 97 +1s)Open 

Paywalling Publicly Funded Knowledge Should be Illegal

Good coverage from The Washington Post on Alexandra Elbakyan's effort to liberate and make free 50 million scientific journal articles from their private vaults https://www.washingtonpost.com/local/this-student-put-50-million-stolen-research-articles-online-and-theyre-free/2016/03/30/7714ffb4-eaf7-11e5-b0fd-073d5930a7b7_story.html.

Like Aaron Swartz before her, Elbakyan is an ideological trail-blazer. Her website, https://sci-hub.io/ continues to add more scientific journal articles every day under the core mission of removing all barriers in the way of science. This should be permanently bookmarked by everyone.

There are issues anyway with the scientific peer-review process and this is complicated by the fact that legacy journals seem like dinosaurs in the modern age, at risk of hindering progress and the... more »

Paywalling Publicly Funded Knowledge Should be Illegal

Good coverage from The Washington Post on Alexandra Elbakyan's effort to liberate and make free 50 million scientific journal articles from their private vaults https://www.washingtonpost.com/local/this-student-put-50-million-stolen-research-articles-online-and-theyre-free/2016/03/30/7714ffb4-eaf7-11e5-b0fd-073d5930a7b7_story.html.

Like Aaron Swartz before her, Elbakyan is an ideological trail-blazer. Her website, https://sci-hub.io/ continues to add more scientific journal articles every day under the core mission of removing all barriers in the way of science. This should be permanently bookmarked by everyone.

There are issues anyway with the scientific peer-review process and this is complicated by the fact that legacy journals seem like dinosaurs in the modern age, at risk of hindering progress and the free dissemination of knowledge. Hopefully SciHub survives and evolves to better help with these issues.

Some of these issues have been covered by +Jesse Powell in a number of different posts, e.g.
(1) Editorial failures https://plus.google.com/u/0/+JessePowell_ad_astra/posts/QgNDLBrPX7w
(2) Unlocking papers https://plus.google.com/u/0/+JessePowell_ad_astra/posts/KeKVwLUNM2h
(3) Open peer review https://plus.google.com/u/0/+JessePowell_ad_astra/posts/AZ1gK5XbvyX___

posted image

2016-04-07 12:18:17 (15 comments; 43 reshares; 137 +1s)Open 

The Accelerating Feedback Loop Between Brain Mapping and Machine Learning

Quanta has yet another great article providing a relatively detailed overview of the current state of brain mapping advances, machine learning advances, and the accelerating feedback between the two helping to drive progress forward https://www.quantamagazine.org/20160406-brain-maps-micron-program-iarpa/.

It covers the Machine Intelligence from Cortical Networks program run by the Intelligence Advanced Research Projects Activity, one of whose goals is to Revolutionize machine learning by reverse engineering the algorithms and computations of the brain. Indeed, machine learning algorithms they will develop based on the neural connection diagrams they uncover will be tested against pattern recognition tasks.

One challenge will be dealing with the enormous amounts of data the research produces —1... more »

The Accelerating Feedback Loop Between Brain Mapping and Machine Learning

Quanta has yet another great article providing a relatively detailed overview of the current state of brain mapping advances, machine learning advances, and the accelerating feedback between the two helping to drive progress forward https://www.quantamagazine.org/20160406-brain-maps-micron-program-iarpa/.

It covers the Machine Intelligence from Cortical Networks program run by the Intelligence Advanced Research Projects Activity, one of whose goals is to Revolutionize machine learning by reverse engineering the algorithms and computations of the brain. Indeed, machine learning algorithms they will develop based on the neural connection diagrams they uncover will be tested against pattern recognition tasks.

One challenge will be dealing with the enormous amounts of data the research produces — 1 to 2 petabytes of data per millimeter cube of brain. The teams will likely need to develop new machine-learning tools to analyze all that data, a rather ironic feedback loop of its own.

"By building machines that think, these researchers hope to reveal the secrets of thought itself." and as the great Richard Feynman said. "What I cannot create, I cannot understand."___

posted image

2016-04-06 14:44:00 (12 comments; 7 reshares; 35 +1s)Open 

The Potential for Blockchains to Replace the State

Launching off from Thomas Hobbs' metaphor of the state and its interconnected system of institutions and social governance as a Leviathan, an artificial being whose capabilities far exceed those of any individual, we have a very interesting hypothesis for whether blockchain technology might eventually evolve into a Leviathan of its own, essentially exerting decentralised governance and force over a constituent population http://ieet.org/index.php/IEET/more/Danaher20160331.

The interesting hook here is the nature of trust. The state, with its institutions and social governance helps solve and coordinate (and enforce) problems of trust between individuals who wish to enter into contracts, and including the default social contract as well as more complex contracts.

Of course, one of - if not the - main reasons... more »

The Potential for Blockchains to Replace the State

Launching off from Thomas Hobbs' metaphor of the state and its interconnected system of institutions and social governance as a Leviathan, an artificial being whose capabilities far exceed those of any individual, we have a very interesting hypothesis for whether blockchain technology might eventually evolve into a Leviathan of its own, essentially exerting decentralised governance and force over a constituent population http://ieet.org/index.php/IEET/more/Danaher20160331.

The interesting hook here is the nature of trust. The state, with its institutions and social governance helps solve and coordinate (and enforce) problems of trust between individuals who wish to enter into contracts, and including the default social contract as well as more complex contracts.

Of course, one of - if not the - main reasons blockchain technology exists is to solve the very same issues of trust between two individuals that may be strangers. However it does this in a decentralised fashion rather than the centralised model of the state.

For this hypothesis to become reality it requires a mature Internet of Things technological ecosystem in which pretty much everything is connected to the IoT and can crunch / send / receive relevant blockchain code.

If everything becomes susceptible to this type of control, we have a technological platform for implementing Leviathan. We won’t need governments, laws and civil institutions anymore. Everything can be managed through the technological infrastructure.

Once this infrastructure exists then one can imagine setting up a Decentralised Autonomous Corporation running on the blockchain and enforcing its contracts, agreements, and control over any physical assets it has been granted, even perhaps those of an enforcement nature.

Fascinating.

I can't remember who first shared this with me; might have been +John Verdon? In any case, thanks!___

posted image

2016-04-06 14:00:15 (12 comments; 4 reshares; 29 +1s)Open 

Is Medical Tourism in Your Future?

FightAging has an interesting proposal on putting together a "Group Purchasing" organisation for medical tourism in order to realise economies of scale and reduce the price of a hypothetical rejuvenation therapy for individuals. See here: https://www.fightaging.org/archives/2016/03/developing-the-art-of-group-buy-medical-tourism-100-people-traveling-to-pay-10-20000-for-a-rejuvenation-therapy/#comment-23855

BioViva has already blazed a trail in this regard, albeit with a single individual, but the exact same therapies are essentially available for anyone who can afford them. While prices can be expected to decline, such therapies are affordable now for probably most in the first world with an interest in such things as part of a group purchasing strategy.

Of course the benefits here mainly to do with accessibility: getting... more »

Is Medical Tourism in Your Future?

FightAging has an interesting proposal on putting together a "Group Purchasing" organisation for medical tourism in order to realise economies of scale and reduce the price of a hypothetical rejuvenation therapy for individuals. See here: https://www.fightaging.org/archives/2016/03/developing-the-art-of-group-buy-medical-tourism-100-people-traveling-to-pay-10-20000-for-a-rejuvenation-therapy/#comment-23855

BioViva has already blazed a trail in this regard, albeit with a single individual, but the exact same therapies are essentially available for anyone who can afford them. While prices can be expected to decline, such therapies are affordable now for probably most in the first world with an interest in such things as part of a group purchasing strategy.

Of course the benefits here mainly to do with accessibility: getting the therapies now rather than a decade or more in future and exploiting one's informed consent to circumvent glacial, vested, precautionary dinosaurs known as regulatory agencies. And being content to travel to more liberal locations if necessary.

The things that might be available in the near future include better stem cell therapies, a range of gene therapies including follistatin and myostatin, version 1 senescent cell clearance and amyloid clearance, and others that are worth investigating further.

I agree with others in the comment thread that bio-hackers will be able to replicate and copy many therapies that are developed by larger biotechs and also will slowly but surely begin producing their own therapies. Indeed this is something that I would like to devote my time and energies to.

However I disagree that such "open source" or DIY efforts wouldn't be able to profit handsomely from this, and along similar lines to the main thrust of this current proposal; I do agree they might profit from books too, and perhaps "Dallas Buyers Clubs". Once you've copied or otherwise developed a therapy you could just set up a shell in some other country, producing and selling it from that other jurisdiction in which the proprietor does not have patent rights. The FDA only regulates sale of therapies into the USA for example.

And so people might travel to that jurisdiction to access it. Alternatively they might order it online and you ship it to them just like an online pharmacy today. The proprietor might try to sue the customers in countries in which they have patents, but this is unlikely given the difficulty in policing this. And would be made more difficult if paid for in some digital currency. This is like a Pirate Bay model for disease and enhancement therapeutics.___

posted image

2016-04-03 08:26:55 (10 comments; 26 reshares; 79 +1s)Open 

SciTech #ScienceSunday Digest - 14/2016.
Permalink here: http://www.scitechdigest.net/2016/04/cell-programming-language-mini-personal.html

Cell programming language, Mini personal pharmacy, Hereditary epigenetics mechanism, IBMs neuromorphic computer, MEMS gravity sensor, Repurposed glucose monitors, Bacteriophages vs amyloids, Automating patch clamping, Biological nanomotor structures, Vision processing units.

1. Programming Language for Cells
A new programming language allows people to program a desired cellular function such as detecting and responding to some environmental condition, and then have the program automatically generates or compiles the DNA sequence that can be inserted into the cell to achieve that function http://news.mit.edu/2016/programming-language-living-cells-bacteria-0331. The proof of concept generated bacteria with... more »

SciTech #ScienceSunday Digest - 14/2016.
Permalink here: http://www.scitechdigest.net/2016/04/cell-programming-language-mini-personal.html

Cell programming language, Mini personal pharmacy, Hereditary epigenetics mechanism, IBMs neuromorphic computer, MEMS gravity sensor, Repurposed glucose monitors, Bacteriophages vs amyloids, Automating patch clamping, Biological nanomotor structures, Vision processing units.

1. Programming Language for Cells
A new programming language allows people to program a desired cellular function such as detecting and responding to some environmental condition, and then have the program automatically generates or compiles the DNA sequence that can be inserted into the cell to achieve that function http://news.mit.edu/2016/programming-language-living-cells-bacteria-0331. The proof of concept generated bacteria with genetic circuits able to respond to three different inputs and respond in different ways and also the largest biological circuit ever created. The design interface will be made available on the web and work will continue to enable the program to compile for species of organisms beyond simple bacteria to include gut bacteria and yeast and potentially higher organisms.

2. Mini Personal Portable Pharmacy
A compact system the size of a refrigerator can be configured to locally produce a variety of drugs on demand including Benadryl, Lidocaine, Valium, and Prozac, and is itself a evolution of an earlier system that was much larger and more complex http://news.mit.edu/2016/portable-pharmacy-on-demand-0331. See image #1 of this post. The group plan to shrink the device by another 40% (another further reduction will get it to desktop scale), and also expand the range of drugs that can be synthesised to include more complex molecules. On demand decentralised production of drugs and other chemicals (like the chemputers I’ve covered previously) is a powerful paradigm with numerous benefits that I can’t wait to see mature and roll out.

3. Mechanism for Hereditary Epigenetic Changes
A mechanism for how epigenetic changes to DNA can be passed onto subsequent generations has been elucidated that depends on small RNA molecules and newly discovered genes that work to turn epigenetic transmissions on and off https://www.aftau.org/weblog-medicine--health?=&storyid4704=2261&ncs4704=3. These mechanisms dictate what epigenetic responses will be inherited and for how long, or how many generations they will be active for. This is very interesting support for many prior studies showing that environmental stress of certain forms can influence the activity of certain genes in subsequent generations. I’m wondering what novel uses this could be used for in synthetic biology applications?

4. IBM’s Latest Neuromorphic Computer
IBM announced the development of a new Scale-up Synaptic Supercomputer based on their TrueNorth architecture to deliver 16 million neurons and 256 million synapses, all for a cool $1 million and consuming just 2.5 watts http://nextbigfuture.com/2016/03/neuromorphic-supercomputer-has-16.html and there is more news on IBM’s resistive computing technology http://nextbigfuture.com/2016/04/bms-resistive-computing-could.html. The main link to Modha’s blog has a lot more detail for those interested. In related news new low-power chips running neural networks developed by large commercial interests are aiming to bring powerful pattern recognition and deep learning applications to mobile platforms (& see #10) http://spectrum.ieee.org/computing/embedded-systems/bringing-big-neural-networks-to-selfdriving-cars-smartphones-and-drones.

5. Tiny MEMS Gravity Sensor
A tiny MEMS gravity sensor has been developed that measured the Earth’s tides - the movement of the Earth’s crust due to the gravitational pull of the Moon and Sun - via detecting 16nm movements of its main silicon mass http://spectrum.ieee.org/energywise/energy/fossil-fuels/stampsized-gravity-meter-could-have-big-impact-on-oil-exploration. The group are working to make the device more sensitive and shrink it (and related equipment) further into something the size of a tennis ball. One might imagine higher resolution gravity maps performed by drones, as well as easier surveying of geological features and volcanoes for example.

6. Detecting Other Diseases with Glucose Monitors
Standard, cheap, portable glucose monitors are being repurposed to detect other diseases http://phys.org/news/2016-03-glucose-diseases.html. The latest approach works by using functionalised liposomes filled with enzymes that produce glucose, that are designed to burst open in the presence of a target molecule and so release the enzymes and cause an increase in glucose in the test solution. The proof of concept demonstrated accurate detection of the thrombin protein, which can indicate heart disease. Repurposing such a basic technology like a glucose monitor in this fashion makes for a pretty compelling platform.

7. Bacteriophage Proteins Treat Amyloid Diseases
The M13 bacteriophage is a type of virus that only infects bacteria and never human cells, but which turns out to have key proteins on its coat that tightly bind to and subsequently dissolve a wide range of amyloid proteins in the brain that are characteristic of a wide range of different diseases including Alzheimer’s, Parkinson’s, Huntington’s, and others like prion diseases http://nextbigfuture.com/2016/03/neurophage-pharma-may-be-able-to-treat.html. Simply delivering the virus nasally was enough to dissolve amyloid plaques in rodents, leading to an 80% reduction, and refining the proteins into a better drug has resulted in a candidate that should be more effective and safer in humans and should soon undergo human trials.

8. Automated Robots to Record Neuronal Activity in Brains
The technique known as neuronal patch-clamping, in which individual neurons are securely contacted in order to measure in fine detail the internal activity of an individual neuron, is known as the finest art in neuroscience. It is very labour intensive but the first automated system for whole-cell patch-clamping has now been developed, which should help to drastically speed up progress in this space http://www.nature.com/news/robots-record-brain-activity-inside-neurons-1.19675?WT.mc_id=GPL_NatureNews. Experienced humans average success rates of 20% - 60%, with the new device currently averaging 33%, but with additional developments, refinements, and improvements can be reasonably expected surpass humans at some point.

9. Biological Nano-motors and Nano-tubes
Electron cryotomography has allowed imaging at 2 - 5nm resolution to capture unprecedented structural details of bacterial flagella and pili motor cell-wall spanning protein complexes, considered to be the strongest molecular motors in existence https://www.caltech.edu/news/close-view-bacterial-motors-50189. This gives functional knowledge of how such motors apply additional torque propel the cell forward, or actively disgorge a protein rope that can motor can then haul the cell along and through viscous and dense environments; this should prove useful for efforts to build stronger and artificial versions. A new class of peptoid molecules have also been shown to reliably and robustly self assemble into long nanotubes of well defined “stepped” diameters http://newscenter.lbl.gov/2016/03/28/peptoid-nanotubes/.

10. Vision Processing Units for Mobile Chips
Much as a CPU can off-load heavy-duty graphical processing to an on-chip GPU to speed up the process and save power, dedicated and specialised VPUs or vision processing units are being developed and proposed to do the same thing for real-time image processing http://www.digit.in/general/the-rise-of-vpus-giving-eyes-to-machines-29561.html. See image #2 of this post. Early versions of VPUs have been used in DJI’s Phantom drones and Google’s Project Tango technology. We can expect our smartphones and other devices to make VPUs much more common in future as they help power a range of incredibly useful navigation, augmented reality, tracking, environment mapping, eye tracking, object classification, and machine learning applications, and also naturally enable performance boosts for computational photography applications. Regarding a type of computational photography, this method to convert black and white images to colour was too good not to share http://richzhang.github.io/colorization/.

SciTech Tip Jar: http://www.scitechdigest.net/p/donate.html
___

posted image

2016-04-02 14:45:47 (9 comments; 2 reshares; 49 +1s)Open 

Bees at 240fps

I captured some photos and videos while walking through the Wittunga Botanic Garden a short drive from home today, including a few clips shot at 240fps with my Nexus 6P of bees flying around this flower. The Google Photos App on Android has some pretty powerful and user-friendly create options for albums, edited images, stories, and movies and let me quickly create this short clip that combines the two best slow-mo segments I was able to capture, and the App also automatically added the music and filter.

At some points you can actually make out the individual wing flaps of the bees, something that I just find amazing being able to do on your smartphone. 

Bees at 240fps

I captured some photos and videos while walking through the Wittunga Botanic Garden a short drive from home today, including a few clips shot at 240fps with my Nexus 6P of bees flying around this flower. The Google Photos App on Android has some pretty powerful and user-friendly create options for albums, edited images, stories, and movies and let me quickly create this short clip that combines the two best slow-mo segments I was able to capture, and the App also automatically added the music and filter.

At some points you can actually make out the individual wing flaps of the bees, something that I just find amazing being able to do on your smartphone. ___

posted image

2016-03-27 09:16:52 (12 comments; 28 reshares; 79 +1s)Open 

SciTech #ScienceSunday Digest - 13/2016.
Permalink here: http://www.scitechdigest.net/2016/03/crispr-vs-hiv-transdermal-drug-implant.html

Living factories, CRISPR vs HIV, Transdermal drug implant, Tiny RFID chip, Smallest genomic organism, Perfect graphene nanoribbons, New neuroprosthetics, Machine learning developments, Smartphone sensor extensions, Plasma printing nanoparticles.

1. Living Factories via Synthetic Biology
Nature has a good review article concerning the rise of synthetic biology and its use in developing living factories able to produce a huge array of products and materials http://www.nature.com/nature/journal/v531/n7594/full/531401a.html. Examples include yeast that produces pain-killing pharmaceutical drugs, implanted encapsulated immune cells that produce signalling molecules for psoriasis, cellular sensors that detect and... more »

SciTech #ScienceSunday Digest - 13/2016.
Permalink here: http://www.scitechdigest.net/2016/03/crispr-vs-hiv-transdermal-drug-implant.html

Living factories, CRISPR vs HIV, Transdermal drug implant, Tiny RFID chip, Smallest genomic organism, Perfect graphene nanoribbons, New neuroprosthetics, Machine learning developments, Smartphone sensor extensions, Plasma printing nanoparticles.

1. Living Factories via Synthetic Biology
Nature has a good review article concerning the rise of synthetic biology and its use in developing living factories able to produce a huge array of products and materials http://www.nature.com/nature/journal/v531/n7594/full/531401a.html. Examples include yeast that produces pain-killing pharmaceutical drugs, implanted encapsulated immune cells that produce signalling molecules for psoriasis, cellular sensors that detect and respond to metabolites in the blood, bacteria that produce engineered spider silk for textiles, and others that produce biofuels and plastics. Also discusses rationale for moving away from model organisms to those optimised for production, and the exciting possibilities and power of moving to cell-free systems for bioproduction. The growing importance of standard software, standard bioparts, standard gene circuits to help with reliability and repeatability in the field is also outlined.

2. CRISPR Shuts Down HIV
CRISPR has been used to disable HIV DNA in human T-cell genomes, and these cells (i) didn’t produce virus to infect other T-cells and (ii) were protected from reinfection after subsequent exposure to HIV http://www.sciencealert.com/scientists-have-removed-hiv-dna-from-human-immune-cells-using-new-gene-editing-technique. The group hopes to improve accuracy before moving into human trials. But overall this is very promising as a platform demonstration for the ability of CRISPR to deactivate pathogenic DNA from cells and so cure chronic viral infections such as HIV, HPV, HSV, and HCV for example.

3. Living Transdermal Implant for Antibody Production
A small, thin implantable capsule has been developed that comprises two permeable membranes held by polypropylene frame and containing a hydrogel loaded with cells engineered to produce recombinant anti-amyloid antibodies http://www.kurzweilai.net/transdermal-implant-releases-antibodies-to-trigger-immune-system-to-clear-alzheimers-plaques. When implanted in mice with Alzheimer's the capsule prevented the immune system from attacking the cells or device, allowed passive exchange of nutrients and wastes, and successfully produced antibodies that dramatically reduced amyloid plaques in the brain. This is a very interesting platform for smart engineered cell therapies applicable across a wide range of applications.

4. Smallest RFID Chip
Hitachi has launched the smallest RFID chip ever produced, measuring just 0.15mm by 0.15mm and 7.5 microns thick http://thefutureofthings.com/3221-hitachi-develops-worlds-smallest-rfid-chip/. This is really getting towards smart dust. Each chip can store 128-bits of read-only memory, and might be useful in a range of logistics, tracking, counterfeiting, and other applications.

5. Minimal Organism syn3.0
Craig Venter’s group has created the smallest viable organism genome, building on work released in 2010 dubbed syn1.0 to create organism syn3.0 with just 473 genes https://www.quantamagazine.org/20160324-in-newly-created-life-form-a-major-mystery/. Painstakingly whittling down the genome of model organism Mycoplasma genitalium (517 genes / 580kilobases) to arrive at the minimal genome required to survive. The big surprise is that up to a third of the genes have unknown functions (the rest are for DNA maintenance, RNA production and translation to protein, and cell membrane maintenance). There appears to be different ways to have a core set of instructions. In related news excellent work has unveiled the structure and function of fundamental gene transcription machinery http://newscenter.lbl.gov/2016/03/23/unlocking-the-secrets-of-gene-expression/.

6. Fabricating Perfect Graphene Nanoribbons
By using specific precursor molecules, perfect graphene nanoribbons with straight defect-free zigzag edges have been fabricated for the first time http://www.mpip-mainz.mpg.de/4567880/PM2016_4. The narrower such perfect graphene nanoribbons are the bigger their electronic bandgap, and they come with the added bonus of each edge conducting electrons in perfect, opposite spins. In related work repeated crumpling and wrinkling of graphene sheets leads to not only workable bandgaps but compressed graphene materials that retain extreme flexibility and electrical conductivity for energy storage and computational applications http://spectrum.ieee.org/nanoclast/semiconductors/materials/crumpling-graphene-repeatedly-adds-a-new-wrinkle.

7. Trio of New Neuroprosthetics
First, a new chip has been implanted into mice brains where it directly senses levels of dopamine and is able to adjust these levels on the fly by electrically stimulating certain neurons to produce more http://spectrum.ieee.org/the-human-os/biomedical/devices/implantable-chip-measures-and-adjusts-dopamine-levels-in-mouse-brain, basically functioning like a thermostat for dopamine rather than temperature, and possibly being able to replace certain drugs for humans for disease or enhancement. Second, a new DARPA program called Targeted Neuroplasticity Training is seeking to use peripheral nerve stimulation in order to enhance general learning processes in the brain http://www.darpa.mil/news-events/2016-03-16. Finally, novel headphones containing twin electrode patches for targeted direct current stimulation of the brain are being used for boosting learning of motor skills https://www.technologyreview.com/s/601054/brain-zapping-headphones-could-make-you-a-better-athlete/.

8. Trio of Machine Learning Advances
First, Google made its Cloud Machine Learning service generally available to everyone as part of the Google Cloud Platform, and allowing people to use the same tools that power products like voice recognition, Google Now, and Google Photos https://cloudplatform.googleblog.com/2016/03/Google-takes-Cloud-Machine-Learning-service-mainstream.html, which will hopefully help accelerate the development of powerful new applications in this space. Second, machine learning systems are making progress on human lip reading and being able to infer speech in the absence of any audio data if the speaker’s lips are visible for a range of applications https://www.uea.ac.uk/about/-/read-my-lips-new-technology-spells-out-what-s-said-when-audio-fails. Finally, DARPA’s latest Grand Challenge is a machine learning competition for collaborative adaptation in optimising spectrum allocation and generally improving wireless communications http://www.darpa.mil/news-events/2016-03-23.

9. Trio of Smartphone Sensor Extensions
First, a new MEMS-based sensor platform plugs into the USB port of smartphones to provide highly sensitive gas sensors capable of exhaled breath analysis to detect volatile organic compounds indicative of certain diseases http://phys.org/news/2016-03-sensitive-electronic-biosniffers-diseases-biomarkers.html. Second, new company Clarius has launched a wireless portable ultrasound device for accurate ultrasound imaging via smartphones http://www.androidpolice.com/2016/03/21/clarius-introduces-a-wireless-portable-ultrasound-for-android-and-ios/. Finally, a $10 infrared laser can grant smartphones accurate depth sensing and range-finding, even outdoors http://news.mit.edu/2016/phone-based-laser-rangefinder-works-outdoors-0325.

10. Plasma Printing with Nanoparticles
Nanoparticles and related materials can now be printed onto arbitrary surfaces including crumbled paper and cloth via a new plasma printing process https://www.aip.org/publishing/journal-highlights/printing-nanomaterials-plasma. This removes the problems of inkjet printing of nanoparticles including needing a liquid form and unable to print on flexible 3D surfaces and also works at low temperatures. The proof of concept used carbon nanotubes, printed onto paper to form simple chemical and biological sensors. Multi-nozzle designs are planned, as well as large-area capabilities. I’m thinking it’d be interesting to see printed large-area quantum dot solar panels or LEDs.

SciTech Tip Jar: http://www.scitechdigest.net/p/donate.html___

posted image

2016-03-25 05:55:23 (4 comments; 4 reshares; 33 +1s)Open 

My Cryocrastination Needs to be Rectified

Tim Urban has a very good, and very detailed overview of cryonics and brain preservation http://waitbutwhy.com/2016/03/cryonics.html. Tim discusses the arguments against and rebuttals for cryonic preservation, the past, present, and future state of the cryonics industry and cryonics technology, and addresses common misconceptions and related fallacies.

Those of you familiar with cryonics will likely still find it worthwhile with several novel points / arguments. For example I learned the new term "cryocrastination", basically the shortsighted procrastination to sign up to a cryonic preservation plan "one day".

Do not go gentle into that good night. Rage, rage against the dying of the light. 

My Cryocrastination Needs to be Rectified

Tim Urban has a very good, and very detailed overview of cryonics and brain preservation http://waitbutwhy.com/2016/03/cryonics.html. Tim discusses the arguments against and rebuttals for cryonic preservation, the past, present, and future state of the cryonics industry and cryonics technology, and addresses common misconceptions and related fallacies.

Those of you familiar with cryonics will likely still find it worthwhile with several novel points / arguments. For example I learned the new term "cryocrastination", basically the shortsighted procrastination to sign up to a cryonic preservation plan "one day".

Do not go gentle into that good night. Rage, rage against the dying of the light. ___

posted image

2016-03-23 13:34:33 (11 comments; 18 reshares; 83 +1s)Open 

Ideas for These Amazing Polymagnets

Destin's wonderful video of these printable Polymagnets did the rounds this week and if you were living under a rock you should definitely give it a watch https://www.youtube.com/watch?v=IANBoybVApQ.

I went to buy a bunch of these magnets from Polymagnet but the ones I wanted were sold out. Their website provides a bunch of information in addition to purchase options, but this page in particular has some additional cool (and short) video animations that make it very clear what some of the benefits and capabilities are http://www.polymagnet.com/polymagnets/. In summary the benefits include:

(i) much stronger magnetic fields close to the magnet surface,
(ii) variable fields with custom repulsion and attraction distances from surface
(iii) variable fields with auto-alignment of magnet surfaces (surfaces “fall” andacc... more »

Ideas for These Amazing Polymagnets

Destin's wonderful video of these printable Polymagnets did the rounds this week and if you were living under a rock you should definitely give it a watch https://www.youtube.com/watch?v=IANBoybVApQ.

I went to buy a bunch of these magnets from Polymagnet but the ones I wanted were sold out. Their website provides a bunch of information in addition to purchase options, but this page in particular has some additional cool (and short) video animations that make it very clear what some of the benefits and capabilities are http://www.polymagnet.com/polymagnets/. In summary the benefits include:

(i) much stronger magnetic fields close to the magnet surface,
(ii) variable fields with custom repulsion and attraction distances from surface
(iii) variable fields with auto-alignment of magnet surfaces (surfaces “fall” and accelerate down a magnetic well
(iv) variable fields with orientation-dependent attraction and repulsion

Technium Pointers

Two of the defining features that are common across the entire Technium and the evolution of technology generally are (i) modularity and the ability to combine different technologies and the phenomena they control in myriad different ways, and (ii) geometry and the ability to form the same technology into different forms and structures in order to exploit the same phenomena differently.

Both are important but in this case I want to focus on geometry. A useful example technology to consider is a particle accelerator, which combines a number of different technologies in such a way to control electric and magnetic fields in a manner that allows subatomic particles to be accelerated to high velocity. The first particle accelerators were linear accelerators that were limited by distance: keep laying down accelerator pipe to get higher exit velocity. It didn't take long to come up with the idea to change the geometry of this design to loop the straight track in on itself to create a cyclotron (spiral) or a synchrotron (circle) like the LHC. Such designs are no longer limited by distance but by the power of the magnet technology and the precision of their control.

A Few Ideas

First off, please share your own ideas in the comments and please critique any errors you see in my own. So, with these novel magnet possibilities in mind and the benefits of novel geometries in technology enabling new applications let’s see what we come up with.

Cylinder
Imagine a cylinder in which the main tube is all one north pole and the ends both south poles, how does this behave? What about a hollow cylinder in which the inner surface is all north and the outer all south? What about a hollow cylinder in which the inner surface is patterned with a self-aligning surface along its length and into which you insert a smaller magnet patterned with a complimentary self aligning surface: would this be like a bullet, accelerating along the length of the tube before exiting ballistically? How strong would the magnetic field need to be in order to match conventional gun muzzle velocity?

Replace the “bullet” with a similarly patterned rod and could you position and anchor this to form a shock absorber for car tyres for which no surfaces touch? What about a cylinder patterned with spiraling (helix) north-south domains along its length, what function could that have when partnered with a similar cylinder to transmit rotational force?

What about megnetically patterned inner and outer cylinders that work together like a specific key and lock respectively; a lock that would be very difficult to pick given it might have the equivalent of over 1,000 “tumblers.”

What would be the value of a mechanical acoustic speaker in which a base magnet is rotated back and forth rapidly, causing an inner magnet to rise up and down vibrating a sheet, instead of the usual mechanism of electrical induction coil plus magnet?

Disk
Imagine a disk patterned along its edge with regular north-south domains; paired with another similar disk this could form a cog force transmission system. Although torque would be low; high torque would cause slippage I’m thinking. What about convenient and improved “air” bearings in which aligned disks and cylinders didn’t touch and could rotate rapidly without friction?

Sphere
Imagine a sphere whose entire surface is patterned in magnetic north poles; unless you’re very close to the surface, wouldn’t this be the equivalent of a magnetic monopole? What benefits or function would there be to playing with the dynamics of multiple magnetic monopoles?

Imagine such a perfect magnetic monopole sphere, placed inside a hollow magnetic sphere whose inner surface was also patterned with the same north pole such that the inner hollow surface and the smaller sphere within mutually repelled one another, forever suspending the smaller sphere in space, not touching any surface.

If the outer sphere had a hole in it, through which you could shine a laser onto the inner sphere and measure the reflected signal, and the entire apparatus was evacuated to vacuum, couldn’t this form a sensitive gravitational wave detector like LIGO?

With the air evacuated to vacuum of our floating inner sphere within a sphere setup, would rapid rotation of the outer sphere and similar laser measurements (say timed pulses through tiny outer sphere holes), constitute an interesting and novel apparatus by which to test the Mach Effect? How would you determine if any rotation of the inner sphere was caused by the magnetic field or just by mass?

Torus
Imagine a hollow donut torus (hollow cylinder curved until ends joined) the inner encircling tube of which is patterned with looped self-alignment domains, essentially setting up a continual / looping magnetic gradient, and into which is placed a small ball magnet patterned with opposite self-alignment domains that causes it to accelerate around and around the inner tube like a particle in a particle accelerator along the magnetic gradient. What velocity would it reach (in air or vacuum) and at what point would it overcome the repulsion to strike the outer surface? What if the inner portion of the torus was removed (hollow cylinder cut in half with semi-circle ends, then curved and ends joined) and the same ball magnet was allowed to accelerate but this time it was anchored via cable to a rotating rod at the centre of the torus to prevent it striking the torus surface?

Change the ball magnet for a slightly smaller torus that is patterned similarly on its outer edge and placed inside the larger hollow torus to rotate freely as per above. Would its final velocity be limited by its structural integrity / strength rather than overcoming repulsion and impacting the wall? What rotational velocity might the rotating ring / torus reach? Would this be useful as a flywheel? What other purposes would it have?

What Are Your Ideas?
___

posted image

2016-03-23 11:15:05 (22 comments; 4 reshares; 44 +1s)Open 

Favourite Web Comic?

After XKCD one of my favourite web comics is Oglaf http://oglaf.com/, which I find consistently humorous but I should warn of occasional frequent NSFW content. Share the love: what are some of your favourite web comics? 

Favourite Web Comic?

After XKCD one of my favourite web comics is Oglaf http://oglaf.com/, which I find consistently humorous but I should warn of occasional frequent NSFW content. Share the love: what are some of your favourite web comics? ___

posted image

2016-03-20 11:27:41 (7 comments; 25 reshares; 87 +1s)Open 

SciTech #ScienceSunday Digest - 12/2016.
Permalink here: http://www.scitechdigest.net/2016/03/low-energy-switches-chicken-dinosaur.html

Low energy switches, Chicken dinosaur legs, Neuroprosthetic hacks, Fullerene molecular surgery, MinION DNA sequencer, Drones with LIDAR, Working memory insights, Sonar phone input, Quantum causality, Osteoporosis stem cells.

1. Zeptojoule Switching
Tests of new nanomagnet arrays demonstrate switching or polarity flips consuming just 6 zeptojoules of energy, or 3x10^-21 joules http://spectrum.ieee.org/tech-talk/semiconductors/devices/zeptojoule-nanomagnetic-switch-supports-fundamental-limit-of-computing. This happens to just be double the Landauer Limit, considered the smallest amount of energy needed to store or reset a bit of information as part of an irreversible computation. Current computers aren’t anywheren... more »

SciTech #ScienceSunday Digest - 12/2016.
Permalink here: http://www.scitechdigest.net/2016/03/low-energy-switches-chicken-dinosaur.html

Low energy switches, Chicken dinosaur legs, Neuroprosthetic hacks, Fullerene molecular surgery, MinION DNA sequencer, Drones with LIDAR, Working memory insights, Sonar phone input, Quantum causality, Osteoporosis stem cells.

1. Zeptojoule Switching
Tests of new nanomagnet arrays demonstrate switching or polarity flips consuming just 6 zeptojoules of energy, or 3x10^-21 joules http://spectrum.ieee.org/tech-talk/semiconductors/devices/zeptojoule-nanomagnetic-switch-supports-fundamental-limit-of-computing. This happens to just be double the Landauer Limit, considered the smallest amount of energy needed to store or reset a bit of information as part of an irreversible computation. Current computers aren’t anywhere near this level of energy efficiency, but this is the first time this Landauer limit test has been performed in a real system that has computational relevance.

2. Growing Dinosaur Legs on Chickens
Genetic manipulation of chicken embryos to inhibit the expression of a single gene resulted in the birds (which did not reach hatching stage) growing leg bones that are anatomically similar to dinosaur leg bones http://www.sciencealert.com/scientists-have-grown-dinosaur-legs-on-a-chicken-for-the-first-time. This was done as a study in evolutionary mapping of changes to gene networks but it accompanies recent work that used similar techniques to grow dinosaur feet on chickens and also dinosaur beaks on chickens. I’d love to see a real, live chickenosaurus!

3. Neuroprosthetic Memory & Sensory Hacks
Rats with Alzheimer’s have difficulty remembering recent experiences, however memories of these recent experiences can be activated by stimulating certain brain circuits via optogenetics http://news.mit.edu/2016/retrieve-missing-memories-early-alzheimers-symptoms-0316. This evidence suggests that these memory problems are specifically a recall problem, and not a storage problem, and also suggesting ways to help develop treatments in future. Rats with electrodes from 4 infrared sensors inserted into brain regions that receive whisker input can adapt with 3 days to sensing and reacting to infrared light signals in the environment https://www.newscientist.com/article/2080671-rats-learn-to-sense-infrared-in-hours-thanks-to-brain-implants/. How long before the first human claims infrared sensing?

4. Fullerene Encapsulated Molecules via Molecular Surgery
Two water molecules have been placed inside a C70 fullerene cage for the first time, in a process dubbed molecular surgery in which precise chemistry is used to first open part of the fullerene shell, water is forced in, and then the process chemically reversed to restore the carbon bonds and reform the fullerene shell http://www.rsc.org/chemistryworld/2016/03/fullerene-traps-water-dimer-hydrogen-bond-study. Interesting technique with a range of new materials possibilities and opportunities to study quantum confinement, intermolecular interactions, and the use of these structures as “nanolaboratories.”

5. Latest on MinION DNA Sequencer
The MinION nanopore DNA sequencer has been in use for a year now, with a new device being shipped this month along with an accompanying DNA data analytics service called Metrichor to provide bioinformatics services http://spectrum.ieee.org/the-human-os/biomedical/devices/portable-dna-sequencer-minion-help-build-the-internet-of-living-things. The company behind MinION hopes to trigger the growth of an “Internet of Living Things” and ultimately to distribute these DNA sequencing units - that plug into computers via USB - to consumers everywhere; sample preparation is still complex however and to address this for people not trained in lab techniques they are also developing a product called VolTRAX as a portable sample prepping device.

6. Drones with LIDAR and Novel Architectures
The cost of LIDAR continues to drop with Scanse now offering a new compact LIDAR unit for use on drones and robots to enable cost-effective autonomy and 3D laser mapping http://www.gizmag.com/scanse-sweep-lidar/42328/. The device isn’t solid state (there are competing efforts towards this) and still relies on mechanical sweeping of beams to build up a local map, is plug-n-play, and can power object detection, tracking, environment mapping, and navigation, however the recently launched TeraRanger One may provide even more convenient distance and navigation tools http://www.teraranger.com/products/teraranger-one/. In related news the Aerotain Skye is one of the more novel and interesting drone designs I’ve seen with novel propellor placement and orientation for navigation, a genuine “eye” in the sky that has to be seen http://spectrum.ieee.org/tech-talk/robotics/drones/cebit-2016-the-aerotain-skye-could-be-your-friendly-floating-camera-drone.

7. A Better Understanding of Working Memory
The old model of short-term working memory, in which you might hold a number in mind, has been overturned with new evidence suggesting that sporadic coordinated bursts of neuronal firing are key, a finding made possible with more detailed and finer-grained measurements http://news.mit.edu/2016/bursts-neural-activity-brain-working-memory-0317. Different bursts at slightly different times allow different items to be held in working memory and separate from each other. These bursts create waves of coordinated activity in the gamma frequency, and the group suggests these hallmarks might also be important in related cognitive functions such as attention. It also sounds like they’d be important for consciousness generally.

8. Touch Input to Smartphone via Sonar
A new finger-based input method for smartphones and smartwatches utilises sonar; the speaker of the device emits inaudible sound waves, echoes from the surroundings are recorded by the device’s microphone and can can calculate the location and movement of a finger nearby to within 8mm and is accurate enough to interact with and control the device http://www.washington.edu/news/2016/03/15/smartwatches-can-now-track-your-finger-in-mid-air-using-sonar/. Seems like a very cool interface technology that probably has a range of applications beyond these obvious ones; I wonder how long until it is available in consumer devices?

9. Messing with Quantum Causality
Moving from using atoms as qubits that have two stable energy states, a recent experiment has examined qutrits as atoms that have three stable but different energy states http://www.sciencealert.com/new-quantum-computer-model-takes-advantage-of-a-loophole-in-causality. Qutrits are typically first given enough energy to get to the middle or second energy state before given a different packet of energy to reach the third or highest state; however the qutrits manage to achieve this even when the order of energy packets is reversed. While there are suggestions that this demonstrates violations of conventional causality, I do wonder if there are other explanations and avenues for exploring this. In related news another group claims to have demonstrated a classical version of information teleportation http://www.uni-jena.de/en/Research+News/FM160304_Teleportation_en.html.

10. Stem Cell Treatment for Osteoporosis
A simple, single injection of mesenchymal stem cells (MSCs) has proven remarkably effective in reversing and treating osteoporosis in mice https://www.fightaging.org/archives/2016/03/towards-a-stem-cell-treatment-for-osteoporosis.php. Mice with age-related osteoporosis had declining MSC levels, and 6 months (quarter lifespan) after the MSC injection all signs of osteoporosis were reversed and replaced with healthy functional bone structure. The group are hoping this can enter human clinical trials sooner rather than later.

SciTech Tip Jar: http://www.scitechdigest.net/p/donate.html___

posted image

2016-03-15 14:00:23 (0 comments; 2 reshares; 47 +1s)Open 

8x Telephoto Lens for Phone

Following the arrival and testing of the little macro (and other) clip-on lenses I had another unit arrive last week, the clip-on 8x optical zoom or telephoto lens. It was $8 from Ebay http://www.ebay.com.au/itm/271938747699.

Here's an example of the test shot I captured over the weekend. This is a tree in my garden and - very unusually - at this moment it was packed with native Australian Rosella birds squawking a racket.

Top photo is what the phone camera would normally capture; bottom left is the view with the clip-on 8x telephoto lens attached, standing from about the same spot.

You certainly see and capture detail you would always miss with a phone camera. But this big clip-on lens is a bit fiddly and needs a little more practice than the others, especially getting used to the twist-focus band to make sure the thing... more »

8x Telephoto Lens for Phone

Following the arrival and testing of the little macro (and other) clip-on lenses I had another unit arrive last week, the clip-on 8x optical zoom or telephoto lens. It was $8 from Ebay http://www.ebay.com.au/itm/271938747699.

Here's an example of the test shot I captured over the weekend. This is a tree in my garden and - very unusually - at this moment it was packed with native Australian Rosella birds squawking a racket.

Top photo is what the phone camera would normally capture; bottom left is the view with the clip-on 8x telephoto lens attached, standing from about the same spot.

You certainly see and capture detail you would always miss with a phone camera. But this big clip-on lens is a bit fiddly and needs a little more practice than the others, especially getting used to the twist-focus band to make sure the thing you're capturing is actually in focus.

Can't go wrong for $8 and you get what you pay for of course. ___

posted image

2016-03-14 04:30:25 (5 comments; 3 reshares; 33 +1s)Open 

Against Interminable Arguments

Another interesting post on SlateStarCodex http://slatestarcodex.com/2016/03/06/against-interminable-arguments/, covering rational norms, productive discussion vs interminable arguments, the need to respond, defending the self, new-comers vs those who've been there before, the lure of long-form discourse, and the benefits of niceness.

Also directly relevant to Google+, because I feel that a similar dynamic has grown and taken place here over the years - in a good sense, for me at least.


Against Interminable Arguments

Another interesting post on SlateStarCodex http://slatestarcodex.com/2016/03/06/against-interminable-arguments/, covering rational norms, productive discussion vs interminable arguments, the need to respond, defending the self, new-comers vs those who've been there before, the lure of long-form discourse, and the benefits of niceness.

Also directly relevant to Google+, because I feel that a similar dynamic has grown and taken place here over the years - in a good sense, for me at least.
___

2016-03-13 13:06:15 (17 comments; 6 reshares; 40 +1s)Open 

Fascinating Analysis of AlphaGo & Parallels with General AI Issues

Many thanks to +Matthew J Price for grabbing the original text below and to +Darius Gabriel Black for sourcing and sharing it in the first place. This makes for fascinating reading - for me at least - especially with the new and concrete examples for issues regarding hard take-off AI scenarios.

------------------------------------------------------------------

+Eliezer Yudkowsky​ on the book of faces:

//(Long.) As I post this, AlphaGo seems almost sure to win the third game and the match.

At this point it seems likely that Sedol is actually far outclassed by a superhuman player. The suspicion is that since AlphaGo plays purely for probability of long-term victory rather than playing for points, the fight against Sedol generates boards that can falsely appear to a human tob... more »

#AlphaGo  

https://m.facebook.com/yudkowsky/posts/10154018209759228___Fascinating Analysis of AlphaGo & Parallels with General AI Issues

Many thanks to +Matthew J Price for grabbing the original text below and to +Darius Gabriel Black for sourcing and sharing it in the first place. This makes for fascinating reading - for me at least - especially with the new and concrete examples for issues regarding hard take-off AI scenarios.

------------------------------------------------------------------

+Eliezer Yudkowsky​ on the book of faces:

//(Long.) As I post this, AlphaGo seems almost sure to win the third game and the match.

At this point it seems likely that Sedol is actually far outclassed by a superhuman player. The suspicion is that since AlphaGo plays purely for probability of long-term victory rather than playing for points, the fight against Sedol generates boards that can falsely appear to a human to be balanced even as Sedol's probability of victory diminishes. The 8p and 9p pros who analyzed games 1 and 2 and thought the flow of a seemingly Sedol-favoring game 'eventually' shifted to AlphaGo later, may simply have failed to read the board's true state. The reality may be a slow, steady diminishment of Sedol's win probability as the game goes on and Sedol makes subtly imperfect moves that humans think result in even-looking boards.

For all we know from what we've seen, AlphaGo could win even if Sedol were allowed a one-stone handicap. But AlphaGo's strength isn't visible to us - because human pros don't understand the meaning of AlphaGo's moves; and because AlphaGo doesn't care how many points it wins by, it just wants to be utterly certain of winning by at least 0.5 points.

If that's true, the case of AlphaGo is a helpful concrete illustration of these concepts:

*

- Edge instantiation. https://arbital.com/p/edge_instantiation/

Extremely optimized strategies often look to us like 'weird' edges of the possibility space, and may throw away what we think of as 'typical' features of a solution. In many different kinds of optimization problem, the maximizing solution will lie at a vertex of the possibility space (a corner, an edge-case).

In the case of AlphaGo, an extremely optimized strategy seems to have thrown away the 'typical' production of a visible point lead that characterizes human play. Maximizing win-probability in Go, at this level of play against a human 9p, is not strongly correlated with what a human can see as visible extra territory - so that gets thrown out even though it was previously associated with 'trying to win' in human play.

*

- Unforeseen maximum. https://arbital.com/p/unforeseen_maximum/

Humans thought that a strong opponent would have more visible territory earlier - building up a lead seemed like an obvious way to ensure a win. But 'gain more territory' wasn't explicitly encoded into AlphaGo's utility function, and turned out not to be a feature of the maximum of AlphaGo's actual utility function of 'win the game', contrary to human expectations of where that maximum would lie.

*

- Instrumental efficiency. https://arbital.com/p/efficient_agent/

The human pros thought AlphaGo was making mistakes. Ha ha.

AlphaGo doesn't actually play God's Hand. Similarly, liquid stock prices sometimes make big moves. But human pros can't detect AlphaGo's departures from God's Hand, and you can't personally predict the net direction of stock price moves.

If you think the best move is X and AlphaGo plays Y, we conclude that X had lower expected winningness than you thought, or that Y had higher expected winningness than you thought. We don't conclude that AlphaGo made an inferior move.

Thinking you can spot AlphaGo's mistakes is like thinking your naked eye can see an exploitable pattern in S&P 500 price moves - we start out with a very strong suspicion that you're mistaken, overriding the surface appearance of reasonable arguments.

*

- Convergence to apparent consequentialism / explanation by final causes.

Early chess-playing programs would do things that humans could interpret in terms of the chess-playing program having particular preferences or weaknesses, like "The program doesn't understand center strategy very well" or (much earlier) "The program has a tendency to move its queen a lot."

This ability to explain computer moves in 'psychological' terms vanished as computer chess improved. For a human master looking at a modern chess program, their immediate probability distribution on what move the chess algorithm outputs, should be the same as their probability distribution on the question "Which move will in fact lead to a future win?" That is, if there were a time machine that checked the (conditional) future and output a move such that it would in fact lead to a win for the chess program, then your probability distribution on the time machine's next immediate move, and your probability distribution on the chess program's next immediate move, would be the same.

Of course chess programs aren't actually as powerful as time machines that output a Path to Victory; the actual moves output aren't the same. But from a human perspective, there's no difference in how we predict the next move, at least if we have to do it using our own intelligence without computer help. At this point in computer chess, a human might as well give up on every part of the psychological explanation for any move a chess program makes, like "It has trouble understanding the center" or "It likes moving its queen", leaving only, "It output that move because that is the move that leads to a future win."

This is particularly striking in the case of AlphaGo because of the stark degree to which "AlphaGo output that move because the board will later be in a winning state" sometimes doesn't correlate with conventional Go goals like taking territory or being up on points. The meaning of AlphaGo's moves - at least some of the moves - often only becomes apparent later in the game. We can best understand AlphaGo's output in terms of the later futures to which it leads, treating it like a time machine that follows a Path to Victory.

Of course, in real life, there's a way AlphaGo's move was computed and teleological retrocausation was not involved. But you can't relate the style of AlphaGo's computation to the style of AlphaGo's move in any way that systematically departs from just reiterating "that output happened because it will lead to a winning board later". If you could forecast a systematic departure between what those two explanations predict in terms of immediate next moves, you would know an instrumental inefficiency in AlphaGo.

This is why the best way to think about a smart paperclip maximizer is to imagine a time machine whose output always happens to lead to the greatest number of paperclips. A real-world paperclip maximizer wouldn't actually have that exactly optimal output, and you can expect that in the long run the real-world paperclip maximizer will get less paperclips than an actual time machine would get. But you can never forecast a systematic difference between your model of the paperclip maximizer's strategy, and what you imagine the time machine would do - that's postulating a publicly knowable instrumental inefficiency. So if we're trying to imagine whether a smart paperclip maximizer would do X, we ask "Does X lead to the greatest possible number of expected paperclips, without there being any alternative Y that leads to more paperclips?" rather than imagining the paperclip maximizer as having a psychology.

And even then your expectation of the paperclip maximizer actually doing X should be no stronger than your belief that you can forecast AlphaGo's exact next move, which by Vingean uncertainty cannot be very high. If you knew exactly where AlphaGo would move, you'd be that smart yourself. You should, however, expect the paperclip maximizer to get at least as many paperclips as you think could be gained from X, unless there's some unknown-to-you flaw in X and there's no better alternative.

*

- Cognitive uncontainability. https://arbital.com/p/uncontainability/

Human pros can't predict where AlphaGo will move because AlphaGo searches more possibilities than human pros have time to consider. It's not just that AlphaGo estimates value differently, but that the solution AlphaGo finds that maximizes AlphaGo's estimated value, is often outside the set of moves whose value you were calculating.

*

- Strong cognitive uncontainability. https://arbital.com/p/strong_uncontainability/

Even after the human pros saw AlphaGo's exact moves, the humans couldn't see those moves as powerful strategies, not in advance and sometimes not even after the fact, because the humans lacked the knowledge to forecast the move's consequences.

Imagine someone in the 11th century trying to figure out how people in the 21st century might cool their houses. Suppose that they had enough computing power to search lots and lots of possible proposals, but had to use only their own 11th-century knowledge of how the universe worked to evaluate those proposals. Suppose they had so much computing power that at some point they randomly considered a proposal to construct an air conditioner. If instead they considered routing water through a home and evaporating the water, that might strike them as something that could possibly make the house cooler, if they saw the analogy to sweat. But if they randomly consider the mechanical diagram of an air conditioner as a possible solution, they'll toss it off as a randomly generated arcane diagram. They can't understand why this would be an effective strategy for cooling their house, because they don't know enough about thermodynamics and the pressure-heat relation.

The gap between the 11th century and the 21st century isn't just the computing power to consider more alternatives. Even if the 11th century saw the solutions we used, they wouldn't understand why they'd work - lacking other reasons to trust us, they'd look at the air conditioner diagram and say "Well that looks stupid."

Similarly, it's not just that humans lack the computing power to search as many moves as AlphaGo, but that even after AlphaGo plays the move, we don't understand its consequences. Sometimes later in the game we see the consequences of a good move earlier, but that's only one possible way that Sedol played out the game, so we don't understand the value of many other moves. We don't realize how much expected utility is available to AlphaGo, not just because AlphaGo searches a wider space of possibilities, but because we lack the knowledge needed to understand what AlphaGo's moves will do.

This is the kind of cognitive uncontainability that would apply if the 11th century was trying to forecast how much cooling would be produced by the best 21st-century solution for cooling a house. From an 11th-century perspective, the 21st century has 'magic' solutions that do better than their best imaginable solutions and that they wouldn't understand even if they had enough computing power to consider them as possible actions.

Go is a domain much less rich than the real world, and it has rigid laws we understand in toto. So superhuman Go moves don't contain the same level of sheer, qualitative magic that the 21st century has from the perspective of the 19th century. But Go is rich enough to demonstrate strong cognitive uncontainability on a small scale. In a rich and complicated domain whose rules aren't fully known, we should expect even more magic from superhuman reasoning - solutions that are better than the best solution we could imagine, operating by causal pathways we wouldn't be able to foresee even if we were told the AI's exact actions.

For an example of an ultra-complicated poorly understood domain where we should reasonably expect that a smarter intelligence can deploy 'magic' in this sense, consider, say, the brain of a human gatekeeper trying to keep an AI in a box. Brains are very complicated, and we don't understand them very well. So superhuman moves on that gameboard will look to us like magic to a much greater extent than AlphaGo's superhuman Go moves.

*

- Patience.

A paperclip maximizer doesn't bother to make paperclips until it's finished doing all the technology research and has gained control of all matter in its vicinity, and only then does it switch to an exploitation strategy. Similarly, AlphaGo has no need to be "up on (visible) points" early. It simply sets up the thing it wants, win probability, to be gained at the time it wants it.

*

- Context change and sudden turns. https://arbital.com/p/context_change/

By sheer accident of the structure of Go and the way human 9ps play against superior opponents - namely, giving away probability margins they don't understand while preserving their apparent territory - we've ended up with an AI that is apparently not being superhumanly dangerous until, you know, it just happens to win at the end.

Now in this case, that's happening because of a coincidence of the game structure, not because AlphaGo models human minds and hides how far it's ahead. I mean, maybe Deepmind deliberately built this version of AlphaGo to exploit human opponents, or a similar pattern emerged from trial-and-error uncovering systems that fought particularly well against human players. But if the architecture is still basically like the October AlphaGo architecture, which seems more probable, then AlphaGo acts as if it's playing another AlphaGo; that's how all of the internal training worked and how all of its future forecasts worked in the October version. AlphaGo probably has no model of humans and no comprehension that this time it's fighting Sedol instead of another computer. So AlphaGo's underplayed strength isn't deliberate... probably.

So this is not the same phenomenon as the expected convergent incentive, following a sufficiently cognitively powerful AI noticing a divergence between what it wants and what the programmers want, for that AI to deceive the programmers about how smart it is. Or the convergent instrumental incentive for that AI to not strike out, or even give any sign whatsoever that anything is wrong, until it's ready to win with near certainty.

But AlphaGo is still a nice accidental illustration that when you've been placed in an adversarial relation to something smarter than you, you don't always know that you've lost, or that anything is even wrong, until the end.

*

- Rapid capability gain and upward-breaking curves. https://intelligence.org/files/IEM.pdf

"Oh, look," I tweeted, "it only took 5 months to go from landing one person on Mars to Mars being overpopulated."

(In reference to Andrew Ng's claim that worrying about AGI outcomes is like worrying about overpopulation on Mars.)

The case of AlphaGo serves as a possible rough illustration of what might happen later. Later on, there's an exciting result in a more interesting algorithm that operates on a more general level (I'm not being very specific here, for the same reason I don't talk about my ideas for building really great bioweapons). The company dumps in a ton of research effort and computing power. 5 months later, a more interesting outcome occurs.

Martian population growth doesn't always work on smooth, predictable curves that everyone can see coming in advance. The more powerful the AI technology, the more it makes big jumps driven by big insights. As hardware progress goes on, those big insights can be applied over more existing hardware to produce bigger impacts. We're not even in the recursive regime yet, and we're still starting to enter the jumpy unpredictable phase where people are like "What just happened?"

*

- Local capability gain. https://wiki.lesswrong.com/wiki/The_Hanson-Yudkowsky_AI-Foom_Debate

So far as I can tell, if you look at everything that Robin Hanson said about distributed FOOM and everything I said about local FOOM in the Hanson-Yudkowsky FOOM debate, everything about AlphaGo worked out in a way that matches the "local" model of how things go.

One company with a big insight jumped way ahead of everyone else. This is true even though, since the world wasn't at stake this time, Deepmind actually published their recipe for the October version of their AI.

AlphaGo's core is built around a similar machine learning technology to Deepmind's Atari-playing system - the single, untweaked program that was able to learn superhuman play on dozens of different Atari games just by looking at the pixels, without specialization for each particular game. In the Atari case, we didn't see a bunch of different companies producing gameplayers for all the different varieties of game. The Atari case was an example of an event that Robin Hanson called "architecture" and doubted, and that I called "insight". Because of their big architectural insight, Deepmind didn't need to bring in lots of different human experts at all the different Atari games to train their universal Atari player. Deepmind just tossed all pre-existing expertise because it wasn't formatted in a way their insightful AI system could absorb, and besides, it was a lot easier to just recreate all the expertise from scratch using their universal Atari-learning architecture.

The October version of AlphaGo did initially seed one of the key components by training it to predict a big human database of games. But Demis Hassabis has suggested that next up after this competition will be getting Deepmind to train itself in Go entirely from scratch, tossing the 2500-year human tradition right out the window.

More importantly, so far as I know, AlphaGo wasn't built in collaboration with any of the commercial companies that built their own Go-playing programs for sale. The October architecture was simple and, so far as I know, incorporated very little in the way of all the particular tweaks that had built up the power of the best open-source Go programs of the time. Judging by the October architecture, after their big architectural insight, Deepmind mostly started over in the details (though they did reuse the widely known core insight of Monte Carlo Tree Search). Deepmind didn't need to trade with any other Go companies or be part of an economy that traded polished cognitive modules, because Deepmind's big insight let them leapfrog over all the detail work of their competitors.

Frankly, this is just how things have always worked in the AI field and I'm not sure anyone except Hanson expects this to change. But it's worth noting because Hanson's original reply, when I pointed out that no modern AI companies were trading modules as of 2008, was "That's because current AIs are terrible and we'll see that changing as AI technology improves." Deepmind's current AI technology is less terrible. The relevant dynamics haven't changed at all. This is worth observing.

*

- Human-equivalent competence is a small and undistinguished region in possibility-space.

As I tweeted early on when the first game still seemed in doubt, "Thing that would surprise me most about #alphago vs. #sedol: for either player to win by three games instead of four or five."

Since Deepmind picked a particular challenge time in advance, rather than challenging at a point where their AI seemed just barely good enough, it was improbable that they'd make exactly enough progress to give Sedol a nearly even fight.

AI is either overwhelmingly stupider or overwhelmingly smarter than you. The more other AI progress and the greater the hardware overhang, the less time you spend in the narrow space between these regions. There was a time when AIs were roughly as good as the best human Go-players, and it was a week in late January.

posted image

2016-03-13 07:33:55 (7 comments; 27 reshares; 98 +1s)Open 

SciTech #ScienceSunday Digest - 11/2016.
Permalink here: http://www.scitechdigest.net/2016/03/advanced-protein-nanomaterials-powerful.html

Advanced protein nanomaterials, CRISPR enhancements, Milligram gravity measurements, Powerful magnetogenetics, Busting bacterial drug resistance, Texture sensing bionics, Deep learning grasping, Reconfigurable nanomaterials, 2D material applications, Regenerating eye lenses.

1. Improved Nanotechnology with Advanced Protein Design
A new platform of combinatorial protein evolution can quickly produce protein pairs from billions that very tightly bind each other, and these can subsequently be used either on their own or bound to nanoparticles (such as gold) to enable and direct the self-assembly of ordered, robust materials with novel properties http://www.nanowerk.com/spotlight/spotid=42804.php. In the case of... more »

SciTech #ScienceSunday Digest - 11/2016.
Permalink here: http://www.scitechdigest.net/2016/03/advanced-protein-nanomaterials-powerful.html

Advanced protein nanomaterials, CRISPR enhancements, Milligram gravity measurements, Powerful magnetogenetics, Busting bacterial drug resistance, Texture sensing bionics, Deep learning grasping, Reconfigurable nanomaterials, 2D material applications, Regenerating eye lenses.

1. Improved Nanotechnology with Advanced Protein Design
A new platform of combinatorial protein evolution can quickly produce protein pairs from billions that very tightly bind each other, and these can subsequently be used either on their own or bound to nanoparticles (such as gold) to enable and direct the self-assembly of ordered, robust materials with novel properties http://www.nanowerk.com/spotlight/spotid=42804.php. In the case of nanoparticle-protein structures, adding an excess of one of the proteins to the solution can cause the spontaneous disassembly of the material. This is a promising platform for atomically precise fabrication, and potentially more powerful than DNA origami, with the group moving onto multiple-component assemblies and control of 3D orientation of large assemblies. Applications across materials science, from catalysts to LEDs. In related news, new and better models of DNA folding should help with the design of DNA origami structures http://engineering.tamu.edu/news/2016/03/07/dna-flexibility.

2. Further Enhancements to CRISPR Technology
A new modification to CRISPR involves small changes to the guide RNA component of the system, a simple extension by 5 basepairs, which resulted in much greater efficiency for gene targeting and knockouts http://www.eurekalert.org/pub_releases/2016-03/ttuh-rec030116.php. This further reduces error rates and off-target effects, and when combined with other approaches should help make the system more specific and safer for use in humans.

3. Measuring the Gravity Between Milligram Masses
It might surprise some but the lightest masses ever to have their gravitational force of attraction between them measured is 90 grams. A new proposal seeks to use modern MEMS springboards to measure the attractive gravitational force between two test masses weighing only milligrams, offering an improvement of three orders of magnitude https://www.technologyreview.com/s/600932/how-to-measure-the-gravitational-field-of-a-quantum-object/. Isolating internal and external vibrations will be crucial, difficult, but achievable, and an important advance for probing ever smaller gravitational fields.

4. Advances in Magnetogenetics
Optogenetics involves transferring genes for light sensitive channel proteins into neurons and then controlling those neurons with pulses of light. In a similar way magnetogenetics involves transferring genes for magnetic proteins conjugated to proteins able to activate the cell https://news.virginia.edu/content/uva-scientists-use-synthetic-gene-and-magnets-alter-behavior-mice-fish. The recent advance was used to turn on neurons in mice and zebrafish via controlling an external magnetic field; for example mice with the gene installed in the pleasure centers of their brains preferentially visited a part of their cage where a magnetic field was located and switched on. Unlike light for optogenetics, magnetic fields can easily penetrate the brain.

5. Busting Bacterial Drug Resistance
Tarocin A and Tarocin B are compounds that have been found to target a different component of bacterial cell walls that, on their own, don’t kill bacteria; however when bound to conventional antibiotics the combination kills bacteria, even those that are resistant to those antibiotics, and so far in clinical samples and infected mice (humans soon hopefully) https://www.newscientist.com/article/2080180-mrsa-superbugs-resistance-to-antibiotics-is-broken/. This is a nice, elegant approach to potentially resurrect many antibiotics for which pathogenic bacteria such as MRSA have developed resistance to. To prevent the development of subsequent resistance to these new combinations the group hope to introduce a third molecular component. In related news some bacteria appear to exhibit a type of collective group memory http://www.eawag.ch/en/news-agenda/news-portal/news-detail/news/bakterien-koennen-kollektives-gedaechtnis-entwickeln/.

6. Bionic Finger Provides Tactile Feeling to Amputee
An amputee with electrodes wired into nerves in his arm has been able to feel surface textures via an artificial bionic finger https://actu.epfl.ch/news/amputee-feels-texture-with-a-bionic-fingertip/. The sensors on the finger translated differences in texture into pulsed signals similar to those that would be normally delivered to the nervous system, with the amputee able to distinguish between rough and smooth surfaces 96% of the time, and quoted as providing a similar sensation to that of the normal hand. The test was repeated with non-amputees and temporary (less invasive) electrodes implanted in the skin to successfully convey tactile information at least 77% of the time.

7. Deep Learning Grasping for Robotics
Google’s new deep learning system for training robotic grasping objects provides continuous feedback as a type of “hand-eye coordination”, getting progressively better at observing its own gripper and correcting / adjusting gripper motions in real-time to increase the chances of a successful grasp on an arbitrary object http://googleresearch.blogspot.com.au/2016/03/deep-learning-for-robots-learning-from.html. And there are other related research systems using deep learning techniques to boost robot performance in complex object grasping and manipulation tasks http://vcresearch.berkeley.edu/bakarfellows/profile/pieter_abbeel. In related deep learning news and following the preliminary announcement a couple months ago, AlphaGo officially defeated the reigning world Go champion this week https://deepmind.com/alpha-go.html.

8. Reconfigurable Nanomaterials
First, the orientation of magnetism in a new class of materials comprised of thin layers of perovskites can be precisely controlled at will, and offering interesting spintronics and other applications https://www.utwente.nl/en/news/!/2016/3/477452/nanotechnologists-at-ut-make-orientation-of-magnetism-adjustable-in-new-materials. Second, a different technique based on thermal scanning probe lithography uses a hot nanotip to heat and cool thin-films of ferromagnetic and antiferromagnetic materials and offering a different way to reconfigure magnetic nanopatterns at will http://asrc.cuny.edu/2016/03/09/riedo-magnetic-nanopatterns/. Finally, in related news a flexible metamaterial can be stretched and tuned to reduce the reflection of a wide range of radar frequencies http://www.news.iastate.edu/news/2016/03/04/meta-skin.

9. Interesting 2D Materials Advances
First, a new graphene water filter removes anything larger than 1nm and is prepared by a novel and apparently scalable process http://www.eng.monash.edu.au/news/shownews.php?nid=11&year=2016. Second, the company Graphenano claims to have developed a graphene polymer battery that achieves 1,000 Wh / kg that would boost a the range of a Tesla Model S from 334 km to >1,000 km http://nextbigfuture.com/2016/03/spanish-company-graphenano-claims.html. Finally, a novel lens 6.3nm thick made from nine layers of molybdenum disulphide and shaped into a lens with a focused ion beam possesses exceptional optical properties http://www.anu.edu.au/news/all-news/worlds-thinnest-lens-to-revolutionise-cameras.

10. Stem Cells Regenerate Human Eye Lens
A human clinical trial with 12 patients successfully regenerated eye lenses damaged by congenital cataracts in all cases http://ucsdnews.ucsd.edu/pressrelease/stem_cells_regenerate_human_lens_after_cataract_surgery_restoring_vision. The procedure involved (i) new minimally invasive surgery to remove the damaged lens while leaving the lens capsule intact, and (ii) methods to stimulate latent lens epithelial cells to regrow the a healthy lens able to restore vision. Great demonstration of using latent stem cells in the body to heal; we’ll hopefully see more of this in future.

SciTech Tip Jar: http://www.scitechdigest.net/p/donate.html___

posted image

2016-03-10 12:58:08 (1 comments; 4 reshares; 23 +1s)Open 

Ethereum and Bringing Blockchains into the Physical World

Here's a good updated overview post of Ethereum from yesterday https://coincenter.org/2016/03/what-is-ethereum/ that is really useful to help explain blockchains and Ethereum to others who have little familiarity with distributed ledgers and bitcoin generally.

And in exploring Ethereum in more detail lately I've acquired a bunch of ETH (Ethereum "currency") too.

That Ethereum post includes a link to Slock.it, via https://slock.it, which I find utterly fascinating and the reason for sharing this post. I've attached their intro ~3min video below that is worth a watch. Slock.it are starting with basic physical locks to connect the Ethereum blockchain to the physical world, but with a broader vision to build this out as a generic Internet of Things platform. The applications and... more »

Ethereum and Bringing Blockchains into the Physical World

Here's a good updated overview post of Ethereum from yesterday https://coincenter.org/2016/03/what-is-ethereum/ that is really useful to help explain blockchains and Ethereum to others who have little familiarity with distributed ledgers and bitcoin generally.

And in exploring Ethereum in more detail lately I've acquired a bunch of ETH (Ethereum "currency") too.

That Ethereum post includes a link to Slock.it, via https://slock.it, which I find utterly fascinating and the reason for sharing this post. I've attached their intro ~3min video below that is worth a watch. Slock.it are starting with basic physical locks to connect the Ethereum blockchain to the physical world, but with a broader vision to build this out as a generic Internet of Things platform. The applications and possibilities here sound pretty exciting.

Finally, thanks to +Rick Heil for sharing this Ethererum resource https://www.reddit.com/r/ethtrader/comments/45shqc/boneheads_ethereum_news_to_come_updated/ summarising some of the many projects, tools, services, and corporations currently being built and tested on Ethereum.
___

posted image

2016-03-06 13:49:23 (5 comments; 31 reshares; 83 +1s)Open 

SciTech #ScienceSunday Digest - 10/2016.
Permalink here: http://www.scitechdigest.net/2016/03/biomolecule-computers-finding-dna.html

Biomolecule computers, Finding DNA modifications, Enhanced learning device, Sophisticated quantum dots, Molecular legos, Synthetic biology containment, Quantum factoring, Sperm in a dish, Deep learning materials, Gravity wave transportation. 

1. Computers Made of Biomolecules
A prototype hybrid biological computer has been demonstrated in which a grid of etched channels is coated with myosin (or kinesin) proteins that use ATP to shuttle actin (or microtubule) protein cylinders around the grid and through different gates http://www.eurekalert.org/pub_releases/2016-02/mu-blb022616.php. Using these biomolecules or nanostructures to shuttle information around the chip is an incredibly energy efficient means to perform... more »

SciTech #ScienceSunday Digest - 10/2016.
Permalink here: http://www.scitechdigest.net/2016/03/biomolecule-computers-finding-dna.html

Biomolecule computers, Finding DNA modifications, Enhanced learning device, Sophisticated quantum dots, Molecular legos, Synthetic biology containment, Quantum factoring, Sperm in a dish, Deep learning materials, Gravity wave transportation. 

1. Computers Made of Biomolecules
A prototype hybrid biological computer has been demonstrated in which a grid of etched channels is coated with myosin (or kinesin) proteins that use ATP to shuttle actin (or microtubule) protein cylinders around the grid and through different gates http://www.eurekalert.org/pub_releases/2016-02/mu-blb022616.php. Using these biomolecules or nanostructures to shuttle information around the chip is an incredibly energy efficient means to perform computations; having demonstrated simple computations with the prototype the group hope to scale-up the architecture to create full-scale functional general computer. 

2. Discovering New DNA Modifications
A new DNA sequencing and detection method is able to identify novel DNA modifications such as epigenetic tags and other chemical alterations quickly and accurately for the first time instead of chance as has usually been the case, and throws up interesting new methods of DNA regulation and control that we may not have been aware of http://news.mit.edu/2016/regulate-protect-dna-modifications-0229. In related news there is renewed discussion of the role of an epigenetic “clock” in organism aging http://nextbigfuture.com/2016/03/epigenetic-clock-controls-aging.html. 

3. Enhanced Learning via tDCS
Low current electrical brain stimulation via transcranial direct current stimulation (tDCS) can improve the learning and retention of complex real-world skills http://www.hrl.com/news/2016/0210/. In this study commercial and military pilots had their brain activity measured, and this pattern was transmitted to student pilots in a flight simulator who subsequently demonstrated improved piloting abilities. It’ll be interesting to see whether this is generally applicable and can be expanded to far broader areas. In related news a new wireless brain computer interface allows monkeys to navigate a wheelchair around an environment with their thoughts http://spectrum.ieee.org/tech-talk/robotics/medical-robots/monkeys-navigate-wheelchair-with-their-thoughts. 

4. Growing Sophistication of Quantum Dots and Nanolights
Nature has a good feature / overview article on the state of development of various quantum dot technologies and their applications http://www.nature.com/news/the-nanolight-revolution-is-coming-1.19482. There’s the typical crystalline semiconductor quantum dots of course and efforts to resolve toxicity concerns, as well as semiconducting polymer p-dots that emit light when they aggregate, AIE-dots comprised of molecules that only emit light when tightly packed, and finally lanthanide upconversion particles that accumulate low energy light and emit high energy light. A range of applications in displays, biosensors, photovoltaics, general sensors, and others are discussed. 

5. Self Assembled Molecular Legos
Using electrospray ion beam deposition, spraying a solution of peptides onto a surface over 1,000s of volts results in ultrapure uncontaminated surfaces on which the peptides self assemble to form regular patterned, atomically precise 2D monolayers http://www.mpg.de/10315424/nanotechnology-selforganized-molecule-lego. Different peptides allow different patterns and chemical properties to be engineered into the surface; such an array might serve as an atomically precise grid for attaching other (e.g. DNA origami) components. In related news organic dye molecules are being made to self assemble atomically precise 2D monolayer arrays of photovoltaic sheets https://www.tum.de/en/about-tum/news/press-releases/short/article/32977/. 

6. Controlling Cells via Peers
A new method for controlling the spread and containment of engineered bacteria and other microbes involves engineering the cells to be dependent on collective survival http://pratt.duke.edu/news/engineered-swarmbots-rely-peers-survival. In this case the cells are engineered to (i) produce a molecule and (ii) when that molecule is in high enough concentrations also produce an antibiotic resistance protein; so long as the bacteria stick together in a high-antibiotic environment they survive. This is demonstrated as a platform that could use any number of signals to control the spread and survival of engineered bacteria. 

7. Proof of Concept Quantum Factoring
Five atoms confined in an ion trap comprise a simple quantum computer able to implement Shor’s Algorithm to find the prime factors of a number, in this case the number 15 http://news.mit.edu/2016/quantum-computer-end-encryption-schemes-0303. The system created by the group is designed to be scalable and extensible and so should enable the building of much larger systems able to factor much larger and more interesting numbers. If this is accomplished then modern cryptographic systems and security will be at risk as large factors form the basis of encryption. 

8. Creating Sperm in a Dish
The latest work directed towards creating artificial sperm and eggs has resulted the the production of rudimentary mouse sperm and the subsequent use of these sperm to successfully fertilise mouse eggs and create healthy offspring http://www.nature.com/news/researchers-claim-to-have-made-artificial-mouse-sperm-in-a-dish-1.19453. This involved reprogramming mouse embryonic stem cells to create primordial germ cells and then incubating these with testicular tissue and other factors to turn them into rudimentary sperm that were directly injected into eggs. The result is controversial and still to be replicated. 

9. Creating New Materials with Machine Learning
A new machine learning system can automatically scan millions of theoretical compounds for novel properties and useful qualities that might be synthesised for a beneficial purpose http://today.uconn.edu/2016/03/building-a-better-mousetrap-from-the-atoms-up/. The group concentrated on certain polymers and created a complex sample of 238 different polymers, calculating their bulk structures and electrical properties to use as training data for the machine learning system, that was then able to accurately predict and calculate the properties of novel polymer combinations, optimising structures for certain desired properties. In related news new company Nervana launched a deep learning cloud service that anyone can use to build computer models that learn http://fortune.com/2016/02/29/nervana-deep-learning/. 

10. FTL Spacetime Routes via Gravity Waves
Here is an interesting theoretical proposal to use gravity waves to create a spacetime geometry between two locations that allows faster-than-light traversal between the two locations compared to or as observed by distant observers http://nextbigfuture.com/2016/02/creating-spacetime-shortcuts-with.html. This attempts to provide a third theoretical option to fast-than-light travel that is consistent with General Relativity, in addition to the more commonly known Alcubierre Drives and Lorentzian Wormholes. Of course, the proposal suffers along with its peers the requirement for truly astronomical sources of energy and no known means by which to achieve this. Interesting nonetheless. 

SciTech Tip Jar: http://www.scitechdigest.net/p/donate.html ___

posted image

2016-03-03 13:39:28 (3 comments; 26 reshares; 69 +1s)Open 

Startup Tips & Quotes for Work and Life

I was going through some old archives the other day and came upon this list of quotes and excerpts I'd collected from an old blog called The Startup Daily that I used to follow but which closed many years ago. I thought I'd share my clippings here because I found them interesting again, still relevant after so many years, and applicable to life and work in general outside of startups and business.

It’s important to regularly ask yourself why you do what you do.
When you don’t have a clear answer, your work is not at its best and your focus is easily diverted from what matters most. Reflecting on Your Purpose Regularly Keeps You Motivated and Protects Against Distractions. A clearly written statement of purpose can act as your compass. It encourages and motivates you to push yourself, and gives you a reason tosay... more »

Startup Tips & Quotes for Work and Life

I was going through some old archives the other day and came upon this list of quotes and excerpts I'd collected from an old blog called The Startup Daily that I used to follow but which closed many years ago. I thought I'd share my clippings here because I found them interesting again, still relevant after so many years, and applicable to life and work in general outside of startups and business.

It’s important to regularly ask yourself why you do what you do.
When you don’t have a clear answer, your work is not at its best and your focus is easily diverted from what matters most. Reflecting on Your Purpose Regularly Keeps You Motivated and Protects Against Distractions. A clearly written statement of purpose can act as your compass. It encourages and motivates you to push yourself, and gives you a reason to say no to distractions that threaten to steal your focus. When the work you are doing is uncomfortable or unnatural, referring to your statement helps to remind you why you should push through the discomfort.

Study Everyday Situations with an Eye Towards Improving Them
Don’t wait for somebody else to tell you what products to develop and what problems to solve. To find a challenge worthy of your time, look no further than the points of friction in your own life. Become a nitpicker. Identify all the little things that are inconvenient or broken in the world around you. Then, instead of just accepting them, pick one and commit to doing something about it. Life’s little problems are an entrepreneur’s opportunities.

Most Successful Products First Languish in the Marketplace, Missing One Crucial Ingredient
“The greatest opportunities around new technologies are not in making things possible, but in solving the myriad little problems that prevent widespread adoption.” The Sony ebook reader was first released in the U.S. over a year before the Kindle. It was lightweight, used the same E Ink display technology, and was similar to the Kindle in almost every way. But it was missing wireless connectivity and the backend that enabled one-click access to the impressive Amazon catalog. Buying and loading new content was a major hassle, and the product floundered. While Sony was in the business of making e-book readers, Amazon was in the business of solving the customer’s problem of buying and reading books, even when the solution meant going outside the device. To find the missing element, try creating a hassle map. List every step, decision, hassle, and potential for making mistakes involved in using your product or service. The key to unlocking the market is probably on that list. The hassles most likely to be overlooked are outside of the obvious functionality of the product, but inside the experience of using the product.

The Stuff You Learn Along the Way Is Ten Times More Valuable than What You Could Learn Beforehand
Many people postpone their plans because they don’t yet know the details of what they will create, how to get where they are going, or how the industry they will be entering works. A little preparation is helpful, but don’t overdo it. You don’t need to be an expert in everything that you might face in the future. You don’t need a plan for every contingency. And you don’t need to gain the approval or permission of industry insiders. Everything that you really need to know can be learned along the way.

Start Before You’re Ready
When launching a new business, project, or initiative, be careful about spending too much time preparing. Don’t fall into the trap of analysis paralysis. The hard part of creating something new is not getting adequately prepared, the hardest part is getting started.

Improve Your Business by Just One Percent Each Week
Most of us have a vision for our business that is far grander than the current reality. But the scope of work required to realize that vision can be immobilizing. It’s hard to know where to start and what to do next. Employees and customers are often threatened by drastic changes. And if you take on too much, too fast and fail to deliver, then your efforts may backfire. Small changes are more manageable, give you time to perfect each step before moving on, and allow you to stay flexible. Your customer’s needs will change and your vision for the business will change, so don’t plan your one percent too far in advance. Reevaluate the next improvement each week. By taking one step at a time, you are able to respond to changing conditions and you avoid being locked into a stale vision. If you improve by just one percent each week, your business will be 50% better after a year.

Don’t Judge the Significance of a New Innovation by the Quality of Early Implementations
The first automobiles were noisy, expensive, and required constant maintenance. The early Internet was not very useful. The first mobile computing devices were extremely limited and barely mobile.To properly gauge the potential of a new technology, you must see past the flaws in early versions. Ask yourself what this experience might look like if it was 10x faster, smaller, cheaper, or better.

You Can Encourage a Desired Behavior by Giving Feedback as if They Were Already Behaving that Way.
People feel compelled to act in ways that are consistent with how others see them. Providing feedback—even false feedback—modifies someone’s self-image, making them more likely to confirm the perception they believe other people already have of them.

Whatever your idea is, someone else has probably already thought of it.
Anyone Can Steal Your Idea, But They Can’t Steal Your Passion or Execution. When you keep your ideas secret to protect them, you are hurting your chances of connecting with people who can help you make them a reality. Experienced investors know this. It’s why they don’t back ideas, they back teams they believe in. Everyone has ideas. It’s the ability to execute that’s priceless.

Your Real Competition Is the Extreme Clutter of the Marketplace
Your competition is no longer just the other offerings in your category. In today’s environment, there is a bigger concern. Every day our choices are increasing in complexity. The number and scope of the products, services, features, advertising messages, and delivery channels that we are subjected to are rapidly growing. Ironically, most companies react to this by adding more clutter. You cannot fight clutter with more clutter. You must find or create your own “white space” to stand out.

Mainstream Customers Will Not Feel Comfortable Unless They Can Compare Your Offering With Something Else
Unlike early adopters, mainstream customers need to feel that they are buying into a mature market, and competition validates that market. Yet in a new market that competition isn’t likely to exist. 
If you don’t have any competition, create some. Look outside of your category to find alternatives that the customer can compare your offering with. By defining the competition you can select for the criteria you want to win by.

For something to go viral there should be a built-in incentive for each user to invite new users.
Small adjustments to the sharing mechanism can have a big impact on how quickly something spreads. If each user brings a single new user you will have linear growth; if each user brings two new users you will have exponential growth. Of course the first step is to have a great product that is worth sharing, bad products do not go viral.

Start Your Presentation by Talking About the Audience
Telling a story is a great way to start. Just make sure the story is about something the audience cares about—like themselves or people just like them. Asking the audience a question is another good way to start. As long as it’s relevant to the audience, and not a meaningless question like “Who wants to be more successful?” Whatever you do, don’t start a presentation by talking about yourself. ___

posted image

2016-03-02 12:14:34 (2 comments; 13 reshares; 61 +1s)Open 

Summary of Recent Rejuvenation & Longevity Science Updates

H+ Magazine has a good summary update on a wide range of recent research projects advancing rejuvenation biotechnologies and longevity interventions. This covers everything from blunt tools like metformin, rapamycin, and resveratrol, to intermediates like immune-boosters and parabiosis-mimics, and also more advanced interventions such as senolytics, organ regenerators, thymic anti-involutants, telomere extenders, mitochondrial reformatting, and cross-link breakers.

Here: http://hplusmagazine.com/2016/03/01/29485/

Worth checking out if only for the spike of optimism from all of this activity, but also for the references and links to more detailed information on each of the programs.

Also, slightly (un)related, those interested in this stuff will probably also be interested in nootropics... more »

Summary of Recent Rejuvenation & Longevity Science Updates

H+ Magazine has a good summary update on a wide range of recent research projects advancing rejuvenation biotechnologies and longevity interventions. This covers everything from blunt tools like metformin, rapamycin, and resveratrol, to intermediates like immune-boosters and parabiosis-mimics, and also more advanced interventions such as senolytics, organ regenerators, thymic anti-involutants, telomere extenders, mitochondrial reformatting, and cross-link breakers.

Here: http://hplusmagazine.com/2016/03/01/29485/

Worth checking out if only for the spike of optimism from all of this activity, but also for the references and links to more detailed information on each of the programs.

Also, slightly (un)related, those interested in this stuff will probably also be interested in nootropics generally, and Scott Alexander has an interesting post and survey results on nootropics here: http://slatestarcodex.com/2016/03/01/2016-nootropics-survey-results/ ___

2016-03-01 09:12:38 (18 comments; 2 reshares; 24 +1s)Open 

A Line Between Your Identity & Another's Memory?

I was pondering the future possibility of directly uploading skills and memories the other day while cleaning the pool. As you do.

I didn't always know how to take care of a pool of course. It was something I had to learn to do; the process, the tools, the methods, the testing, the chemistry, the little things to look out for and the little things to fix.

This goes for any new skill or knowledge of course.

I wondered about the alternative to encountering such a novel problem and dealing with it by tapping into the network and immediately having that memory or skill integrated into my brain. Not just the sense of knowing this knew knowledge and skill as an external thing to be known and acted upon. But integrated to such an extent that I knew it intuitively, knew it without thinking about it,... more »

A Line Between Your Identity & Another's Memory?

I was pondering the future possibility of directly uploading skills and memories the other day while cleaning the pool. As you do.

I didn't always know how to take care of a pool of course. It was something I had to learn to do; the process, the tools, the methods, the testing, the chemistry, the little things to look out for and the little things to fix.

This goes for any new skill or knowledge of course.

I wondered about the alternative to encountering such a novel problem and dealing with it by tapping into the network and immediately having that memory or skill integrated into my brain. Not just the sense of knowing this knew knowledge and skill as an external thing to be known and acted upon. But integrated to such an extent that I knew it intuitively, knew it without thinking about it, essentially performing someone else's knowledge, someone else's skill . . . without knowing how or why I knew it.

This would of course be very quick and extraordinarily convenient. It would be powerful; seductively tempting to do the same with every new skill and piece of knowledge.

But would we lose something in the process? Would this lessen or demean our sense of Self? What would it say about our identity? Especially if you take it to extremes in which every skill and every piece of knowledge about the world in your head was uploaded there, like Trinity's helicopter pilot knowledge in The Matrix, copied from someone else's network, someone else's identity? Who are you in this scenario?

When all you have, uniquely, are your memory of life events? And even these might be added and deleted and altered and swapped, possibly with, and possibly without your knowledge. Does you and your lose all meaning in this situation? Does the collective then reign supreme? Can our memes, which we gave birth to, then be said to have succeeded in engineering their environment to their "purposes" as homo sapiens has its own environment?___

posted image

2016-02-29 13:46:45 (4 comments; 16 reshares; 58 +1s)Open 

How To Grow Almost Anything

HTGAA http://bio.academany.org/ is a synthetic biology program directed by George Church and is part of the growing "academany" and the distributed network of FabLabs around the world. Like the 3D printing FabLabs and the associated courses for How to Make Almost Anything, the idea is to provide (i) an engaging curriculum for people to quickly pick up knowledge and skills, and most importantly (ii) access to a laboratory for people to put those skills and new knowledge to work in making - or in this case growing - something new.

There are a bunch of participating laboratories around the world and I am very interested in setting up an Australian node for this program. Which makes a lot of sense because it is something I've been tinkering with and thinking about doing in my garage anyway.

Classes

I mean, with... more »

How To Grow Almost Anything

HTGAA http://bio.academany.org/ is a synthetic biology program directed by George Church and is part of the growing "academany" and the distributed network of FabLabs around the world. Like the 3D printing FabLabs and the associated courses for How to Make Almost Anything, the idea is to provide (i) an engaging curriculum for people to quickly pick up knowledge and skills, and most importantly (ii) access to a laboratory for people to put those skills and new knowledge to work in making - or in this case growing - something new.

There are a bunch of participating laboratories around the world and I am very interested in setting up an Australian node for this program. Which makes a lot of sense because it is something I've been tinkering with and thinking about doing in my garage anyway.

Classes

I mean, with free classes, guidelines, and course materials on topics including DNA Nanostructures, Biofabrication & Additive Manufacturing, Genome Engineering, Gene Drives & Synthetic Ecosystems,_ and Engineering the Human Gut Microbiome, who wouldn't want to get stuck in and do this stuff!?

See: http://bio.academany.org/doc/classes

Inventory

HTGAA also provides a very helpful resource for sourcing cheap, accessible equipment for putting a laboratory together that will allow you to do these things, for example (i) a DIY generic lab equipment guide, and (ii) How to set up a molecular laboratory for $1,000 from Ebay.

See: http://bio.academany.org/doc/inventory

EDIT:
Easily Accessible Genetic Parts

AddGene provides easily accessible genetic parts for people, including CRISPR / Cas9 constructs as needed. See: https://www.addgene.org/ (via https://www.technologyreview.com/s/601156/the-scientific-swap-meet-behind-the-gene-editing-boom/)___

posted image

2016-02-29 07:56:47 (3 comments; 3 reshares; 18 +1s)Open 

Joy Flight & Fleur de Peau Light Show

Elise and I had a big day on Saturday and covered a lot of ground around Adelaide, enough for me to make this little 3 minute highlight clip :)

We started the day with a low altitude joy flight on a little 4-seater plane that proved to be much bumpier than either of us expected, but we were rewarded with some nice views over Adelaide that we'd never seen nor appreciated before. This included the mangroves where we'd gone fishing and kayaking previously, as well as Adelaide Oval where we were headed later that night. Funnily enough on approach to land, listening to the air traffic control chatter we heard the pilot say "Nah, not going to be able to make it from here." and then he did a loop of the air field before coming in a second time to land.

We ended the day at Adelaide Oval to watch the launch of the... more »

Joy Flight & Fleur de Peau Light Show

Elise and I had a big day on Saturday and covered a lot of ground around Adelaide, enough for me to make this little 3 minute highlight clip :)

We started the day with a low altitude joy flight on a little 4-seater plane that proved to be much bumpier than either of us expected, but we were rewarded with some nice views over Adelaide that we'd never seen nor appreciated before. This included the mangroves where we'd gone fishing and kayaking previously, as well as Adelaide Oval where we were headed later that night. Funnily enough on approach to land, listening to the air traffic control chatter we heard the pilot say "Nah, not going to be able to make it from here." and then he did a loop of the air field before coming in a second time to land.

We ended the day at Adelaide Oval to watch the launch of the Adelaide Festival of the Arts (run in parallel with the Fringe Festival and incidentally every other festival. For those that don't know, this time of year is referred to as "Mad March" in Adelaide because for some reason my city decides to host every event for the year at once). The launch was performance Group F and their show A Fleur De Peau, a light, audio, projection, and fireworks show. Aside from the fireworks I felt they were a little drowned out in the stadium as we were too far removed from the stage; otherwise the glow-suits were pretty cool, the beats and bass were impressive, the synchronised flames and lights were well done and the fireworks were some of the best I've seen in a long time.

The video is ~90 seconds flight + ~90 seconds light show and fireworks.

___

posted image

2016-02-28 12:44:47 (4 comments; 3 reshares; 59 +1s)Open 

Support for Bohmian Mechanics & Quantum Determinism

A recent experiment presents evidence as additional support for Bohmian Mechanics and similar hidden variables interpretations of quantum mechanics. This doesn't rule out other interpretations of course, but the new work does overturn a 1992 experiment that aimed to refute Bohmian Mechanics.

Popular overview of the work: https://www.newscientist.com/article/2078251-quantum-weirdness-may-hide-an-orderly-reality-after-all/

The new work shows that the act of sending one of an entangled pair of photons through a double-slit apparatus causes the other photon to change its polarisation mid-flight, to the point where you can no longer predict with certainty the polarisation fate of one with regard to the other.

[1] Paper: http://advances.sciencemag.org/content/2/2/e1501466
[2] Wikipedia:... more »

Support for Bohmian Mechanics & Quantum Determinism

A recent experiment presents evidence as additional support for Bohmian Mechanics and similar hidden variables interpretations of quantum mechanics. This doesn't rule out other interpretations of course, but the new work does overturn a 1992 experiment that aimed to refute Bohmian Mechanics.

Popular overview of the work: https://www.newscientist.com/article/2078251-quantum-weirdness-may-hide-an-orderly-reality-after-all/

The new work shows that the act of sending one of an entangled pair of photons through a double-slit apparatus causes the other photon to change its polarisation mid-flight, to the point where you can no longer predict with certainty the polarisation fate of one with regard to the other.

[1] Paper: http://advances.sciencemag.org/content/2/2/e1501466
[2] Wikipedia: https://en.wikipedia.org/wiki/De_Broglie%E2%80%93Bohm_theory

Update: Quanta also discusses the same work https://www.quantamagazine.org/20160517-pilot-wave-theory-gains-experimental-support/ ___

posted image

2016-02-28 12:21:44 (3 comments; 2 reshares; 38 +1s)Open 

Very Cool Android Experiments

The Android Experiments page provides a bunch of cool and innovative applications for you to check out and play with https://www.androidexperiments.com/.

The image below is from the Selfie x Selfie app, and I've also played with Tilt, Elements, Material Life, and Spheretones, on my phone and also the very cool Time Mesh on my watch. I'd also love to build and play with the IOIO Rover, which looks like a much improved version of a remote telepresence robot I built 3 or 4 years ago now. Although mine did have a cannon!


Very Cool Android Experiments

The Android Experiments page provides a bunch of cool and innovative applications for you to check out and play with https://www.androidexperiments.com/.

The image below is from the Selfie x Selfie app, and I've also played with Tilt, Elements, Material Life, and Spheretones, on my phone and also the very cool Time Mesh on my watch. I'd also love to build and play with the IOIO Rover, which looks like a much improved version of a remote telepresence robot I built 3 or 4 years ago now. Although mine did have a cannon!
___

Buttons

A special service of CircleCount.com is the following button.

The button shows the number of followers you have directly on a small button. You can add this button to your website, like the +1-Button of Google or the Like-Button of Facebook.






You can add this button directly in your website. For more information about the CircleCount Buttons and the description how to add them to another page click here.

Mark BruceTwitterCircloscope