Slashdot Mirror

I've just looked it up and the human visual system recognises objects in 100-200ms (see http://www.scholarpedia.org/ar...).
Looking at this article (https://www.independent.co.uk/news/uk/home-news/highway-code-car-stopping-distances-wrong-drivers-thinking-time-brake-rac-a7859061.html) they use a 1.5 second reaction time.
Maybe you take 5-10 seconds because you're a racist piece of shit.

Speaking of aging, time could also be described as a function of entropy?
http://www.scholarpedia.org/article/Time%27s_arrow_and_Boltzmann%27s_entropy
https://en.wikipedia.org/wiki/Entropy_(arrow_of_time)

1) The speed of light is constant, everywhere and everywhen.

Variable Speed of Light theories, around for 20-30 years. Investigated in cds.cern.ch/record/618057/files/0305457.pdf
Current observations put very very very tight bounds on dc/dt.

2) The gravitational constant is the same, everywhere and everywhen.

Jordan-Brans-Dicke theories. Hundreds of papers, dating back to Dirac's large number hypothesis. See http://www.scholarpedia.org/ar...

3) The shape of space is uniformly flat, everywhere and everywhen.

No. Space-time is a curved manifold. Not only in the cosmological (isotropic+homogeneous limit) with limiting spaces of three-spheres or hyperbolic spaces, but also across huge perturbations on them. See https://arxiv.org/abs/1501.038... for tests of homogeneity and isotropy, for example.

4) Please don't get me started about standard candles.

OK, I won't. They conform with observations, match the fine-stucture constant and Lyman forest predictions incredibly well, vast literature that exists on these matches cosmic helium and hydrogen observations, matches with galactic rotations etc.

5) Or cosmological inflation.

Producing effects as predicted. See https://arxiv.org/abs/1311.165... for an in depth probability test of cosmological parameters including the spectral tilt and scalar to tensor ratio of perturbations (CMB) predicted by inflation.

6) Or the (luminiferous) aether. Sorry, the Higgs field/particle/whatever.

Predicted, observed, behaving exactly as predicted at 5 sigma significance in the mass.

Just because YOU don't know about it, doesn't mean that we haven't investigated it. But of course, feel free to keep having a reckon without looking stuff up first.

Sort of. If it took 5 years and actually worked they were clearly becoming very, very well acquainted with a heart of the problem. (I do mean "a" heart: there can be more than one way of dealing with a complex problem.)

Whether they were becoming very, very well acquainted with disciplines at the heart of the problem is another question.

For example in perturbation theory (examining what happens to a system if you change a parameter by a small amount) there are objects called Canards. These were first discovered by non-standard analysts who applied infinitesimal perturbations. (Which incidentally acts as a ripost to someone recently claiming in another thread that the concept of infinity was practically useless.) These analysts were French, and graphs of the objects had curves which looked a bit like ducks. Hence Canards.

The mathematician Ian Stewart writes about this in "From Here to Infinity" (Oxford University Press 1996), one of his books on problems in contemporary mathematics. He relates how some non-standard analysts found the canards, adds that the reaction of a conventional perturbation theorist was to find a rather complicated and delicate normal perturbation theory proof while expressing (unjustified) doubts about the non-standard proof. Stewart then comments that it's easy to discover things with hindsight, and that he'd be more impressed by the perturbation theory proof if it had been found before the non-standard analysts had done the donkey work.

And don't forget we've got two brains. There is also a new current in Cognitive Science rapidly gaining ground - Enactivisim - which rejects the brain-is-everything paradigm common in the Computationalist approaches. Brains are definitely necessary but definitely only part of understanding what goes on with humans, or any other animals for that matter.

The linked article was horribly written. I'll give a shot at trying to explain it (or rather, a really, really simplified version).

Two of the fundamental problems that neural circuits must solve are the noise-saturation dilemma and the stability-plasticity dilemma. The first is best explained in the context of vision. Our visual system is capable of detecting contrast (ie. edges) over a massive range of brightness, spanning a space of about 10^10. Given that neurons have limited firing rates (typically between 0 and 200hz), there needs to be some normalization criteria that allows useful contrast processing over massive variations in absolute input (more on this later). The stability-plasticity dilemma is that the brain needs to be sufficiently flexible to learn based on a single event (let's say, touching a hot stove is a bad idea), but once learned memories have to be sufficiently stable to last the rest of a creatures' life span.

The stability-plasticity dilemma implies that neural circuits must operate in at least two (as I said, very simplified) distinct states, a "resting" or "maintenance" state, and a "learning" state, and that there is a phase-transition point in between them. Furthermore, these states need to have the following properties regarding stability:
1) the learning state must collapse into the maintenance state in the absence of input (otherwise you get epilepsy).
2) reasonable stimulation (input) during the resting state must be able to trigger a phase change into the learning state (or you become catatonic).

Many circuits/mechanisms have been proposed to explain how the brain solves these dilemmas. Most of them involve the definition of a recurrent neural network using some combination of gated-diffusion and oscillatory dynamics to fit well known oscillatory and wave-based dynamics that have been recorded in neural circuits. Some of these models employ intrinsic learning using a learning-rule (ie. self-organized maps) while others are fit by the researcher. One key point about this class of models (as opposed to the TFA approach) is that they have a macro-circuit architecture specified by the modeler. Typically these models are at least somewhat sensitive to parametric perturbation.

TFA describes another approach, which comes out of research on cellular automata done by Ulam, von Neumann, Conway and Wolfram. This approach posits that parametric stability and macro-circuit organization is only loosely important so long as the system obeys a certain set of rules regarding local interaction (could also be through of as micro-circuit) because it will self-organize to a point of 'critical stability'. In the the two-state model described above, this approach predicts that neural circuits are always at a state of 'critical stability' where maintenance occurs through frequent small perturbations or avalanches, and any new input will trigger a large avalanche, causing learning. Bak has proposed this as a general model of neural circuit organization. One trademark of these type of models is that they show 'scale free' or 'power law' behavior, where the size of an event is inversely proportional to its frequency by some exponential function. Some recent data has shown power-law dynamics in neural populations (a lot of other data doesn't show power-law dynamics).

One big problem with the critical stability hypothesis is that it doesn't deal well with the noise-saturation dilemma: it needs to cause the same general size of avalanche whether it's hit by one grain of sand, or 10^10 grains of sand.

None of this is particularly new, neural-avalanches (albeit in a different context) were postulated in the early 70s. Could some systems in the brain exploit self-organized criticality? Sure, but there is a lot of data out there that's inconsistent with it being the primary method of neural organization.

Don't confuse the Wiki as a tool with Wikipedia.

A lot of the discussion here doesn't seem to have much to do with TFA. (Surprise, surprise...)

People seem to be missing the importance of "scholarly canons" in the summary. TFA is NOT about open-access publishing (except indirectly). This is NOT about Wikipedia (except perhaps as a model of how to do certain aspects of a scholarly encyclopedia better than Wikipedia).

TFA holds up Scholarpedia as its main exemplar of a better kind of scholar online encyclopedia of canonic knowledge in a particular field.

That's not the only one out there, and Scholarpedia does have its issues. Personally, I think if our goal is to produce a standard scholarly encyclopedia for a particular discipline (or for many disciplines), we could also take the Stanford Encyclopedia of Philosophy as a great example of a model project of collaboration by scholars to produce a summary of research and ideas in a discipline...

... and it's been around for nearly 20 years already. Long before most of these other things have existed.

good vision is about one arc minute.

Good vision "Snellen" Acuity is about one arc minute. However, visual hyperacuity, which is a sub-Nyquist analysis of the retina, can be down to a few seconds of arc for "vernier" tasks such as adjusting two lines to be directly on top of one another.

Furthermore, 1.5m is going to be small. The TV of the future (10-20 years from now) takes up your entire wall, and your field of view of it may be 60 degrees rather than 30 degrees. We may need 8K resolution to make the most of that

Plus you will not just be watching "TV" on your wall, but you will have Twitter feeds, Slashdot headlines, baby monitors, weather info, etc. tiled across it.

And then I read http://www.scholarpedia.org/article/Galactic_magnetic_fields and sure enough in the Milky Way "outside the central region, the large-scale [magnetic] field is mostly parallel to the plane of the Galactic disk." The central region is the exception, yet it still shows nebula alignment.. something still doesn't mesh.

Is quantum entanglement the only physical resource that allows for such strong encryption?

I.e. does exploiting thermodynamic properties already suffice as claimed in the Kish cypher?

This planet was discovered by Lorentz boosting, the theory of which predates Einstein. Meanwhile, 20 exoplanets have been discovered to date using gravitational lensing, an application of General Relativity (a theory created by Einstein ) that was itself first predicted by Einstein. Somehow, the press release (and thus all the subsequent press) failed to mention these "Einstein planets."

A lot of vague marketing-speak in this article. "Deep learning"? The article basically talks about neural networks, just one of the techniques in machine learning.

It's hard to tell from the article, but they probably are trying to refer to Deep Belief Networks, which are a more recent and advanced type of Neural Network, which incorporates many layers:

Deep belief nets are probabilistic generative models that are composed of multiple layers of stochastic, latent variables. The latent variables typically have binary values and are often called hidden units or feature detectors. The top two layers have undirected, symmetric connections between them and form an associative memory. The lower layers receive top-down, directed connections from the layer above. The states of the units in the lowest layer represent a data vector.

The sorting paper: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.52.7331&rep=rep1&type=pdf

Trilobites: http://www.scholarpedia.org/article/Punctuated_equilibria (see references)

Even the AI Mind in JavaScript uses inhibition for electronic brain function, not for enforcing a sparse olfacotry environment, but rather to suppress just-past thoughts in favor of yet-to-emerge thoughts.

>I've got a PhD in cosmology

Speaking of which, could you have a look here and here, and comment and pros and cons of both semi-vetted pages regarding astrophysics?

If you could remove ego from it, I could see a system in which someone like you could write the condensed matter physics page as a "placeholder" until someone in the field was able to come by and replace it. Unfortunately, the human ego is both a vanishingly small and vastly huge thing at the same time, and it throws a monkey wrench into that plan. Scientists don't get a whole lot of money, all they have to go on is recognition, hence accounting for the huge egos in science.

Or, something like Scholarpedia, perhaps?

There is http://www.scholarpedia.org/ , a wiki with peer-review that is for now limited to a few domains of physics and neuroscience. It is on invitation only, however.

The brain is an inexplicable thing

Bullshit. The brain is a computer. Sure, it's a strange architecture: it's made of billions of impressively energy-efficient gates each operating at the order of tens of hertz. Fan-out is huge --- a gate on a microprocessor might be connected to 50 others, but a neuron can have tens of thousands of connections. A CPU has one fast, global clock, while the brain has overlapping and distributed clock signals for synchronizing neuron firing. The short term memory system uses the equivalent of old-fashioned delay lines, while long-term storage is implemented with redundant, distributed rewiring. It's content-addressable and has a storage capacity in the terabyte range, though it has really lousy indexing. Input and output are essentially memory-mapped, with lots of special purpose hardware acceleration.

There are a lot of similarities too: both our computers and our brains run software, with only a few basic features baked into the hardware. Both parse raw environmental input and parse it into abstractions that can be manipulated symbolically according to software-defined rules. Both can evaluate the lambda calculus and run a universal Turing machine. Neither can solve the halting problem in all cases. Both have large data stores. Both have networked inputs. Both crash. Both employ algorithms and data structures to process information. Both eventually fall apart.

Our brains are not magical devices somehow above scientific inquiry. They are ordinary, pedestrian objects in that obey the same laws of physics that govern baseballs and light switches. That we don't completely understand all the brain's mechanisms is no reason to believe it's qualitatively different from any other computer. Have you read every line of code in the web browser you're staring at?

Good reference. That's also how our brains create consciousness and what we call "selves" (which are actually self-models) if Metzinger is right. This could be a primitive start toward Earth itself becoming a conscious entity.

No, that would be Jacob Bekenstein. Hawking and Bekenstein collaborated to precisely fix the ratio of entropy to surface area, but the original idea wasn't Hawking's.

There have been several other attempts in to setup similar wikis. For example, Scholarpedia is exactly this model of a peer-reviewed topical encyclopaedia, but for mathematical sciences. There were two comments from other Slashdotters, complaining that a group of academics, or any group of people will often struggle to reach consensus. But I think that there are qualitatively different types of disagreements. Some are about writing or presentation style ("where the place the word 'the'"). But, some are more substantive, especially in topics that are not entirely resolved. For example, there is little disagreement that Newton's laws are wrong, but nearly exact for certain spatial and time scales. But, if you were to write an article on information coding in neurons, there are probably as many opinions as there are labs working in that area!

If only Wikipedia became more widely used than it is presently, especially in academic circles, then more groups will be interested in having articles reflect debates. To reflect different opinions is particularly important in fields involving subjectivity (pretty much every thing other than Mathematics). If there is enough interest among academics in Wikipedia, then the current state of debates on various topics is bound to be reflected in the articles.

Given that Stanford's plato website is simply a fledgling effort, I do not see why it is newsworthy. If for example, someone cited an article from the plato website in a peer-reviewed journal article (and reviewers accepted it), that would be newsworthy. Short of that, it is simply yet another effort at collaborative information sharing. It cannot be newsworthy simply because it is from a well known university.

The answer depends on the type of code, process used, and history of the people involved. I don't doubt that a PM at Microsoft believes that every bug is simple to find if just enough eyes look at it. No doubt at all.

I've worked on real-time space vehicle GN&C code where a slow answer is a wrong answer. We've had a few really complex bugs and a large number of "duh" bugs that the review team studied, but then were convinced it was fine. We've also had bugs where 20 seasoned professionals missed the bug and someone with less than a month on the job caught it because he (actually, it was me) didn't assume something operator order of the compiler that everyone else had assumed. I looked up what the order of operations was and found something very non-standard.

I've seen many array bounds bugs, string handling bugs, pointer miss management bugs, RTTC bugs and library bugs. Most common bugs can be avoided by how you write your code, IME.

1) Always set variables to known values at instantiation and when you are completed with them.
2) Always perform tests with the constant on the left side of the comparison operator.
3) Always set pointers to non-allocated memory to NULL before and after use. It is easy to continue using a pointer that happens to work even when it points to a freed memory block. Better to get a null pointer access error during development than for anyone to find it during runtime.
4) Run all code through an indentation tool to correct any user specific styles.

Oh, I've seen rendezvous code fail due to only using a single precision floating point variable in the calculations. The fix was to use double precision floats AND to initialize the variable to ZERO at the end of the calculation, so when another rendezvous 3 body problem calc http://www.scholarpedia.org/article/Three_body_problem was requested, errors didn't add up over time. Actually, this issue was discovered during a flight with many multiple rendezvous guidance calculations. Here's a mention of GN&C issues http://www.spaceref.com/news/viewsr.html?pid=26977 in a NASA release.

Did I mention - I am a rocket scientist.

Right, I suppose "shuns experts" was a bit vague. Let me clarify. Wikipedia is the place where an expert's credentials and experience are no match for an unknown conspiracy theorist who has decided an article must include certain content _he_ believes is perfectly valid and useful to mankind.

And the conspiracy theorist has reliable 3rd party sources for his claims, and the so-called expert can't find anything to back his argument up? If you say so.

The last time I edited a Wikipedia article in 2006 my changes were reverted by one of those zealous article owners (which I'm told by people like you are not supposed to exist), and I was later banned from editing for three days by one of his administrator buddies. Not by him you understand, by his buddy. I was given the choice to "file a content dispute" or something like that. All over a paragraph added to the article about an 80s rock band from Argentina. With a perfectly acceptable backing source, by the way.

Link please?

And then we have the usual "all the important people don't contribute to Wikipedia, waah". Well, there are other encyclopedias that work via different means - perhaps you should look into http://en.citizendium.org/wiki/Welcome_to_Citizendium or http://www.scholarpedia.org/ where brilliant experts such as yourself will be welcomed with open arms, and no other expert will ever possibly disagree with what you have to say about an 80s rock band from Argentina, honest.

Now, remind me how well those sites are doing compared to Wikipedia again? Wikipedia is the one struggling to get people to contribute, you say?

There is a very successful peer reviewed (scientific) website called Scholarpedia (http://www.scholarpedia.org/) which I tend to trust over wikipedia.

Heat difference provides also is quantified by entropy. So obviously the heat produced by the rabbit can be converted into libraries of congress.
The applicable equation is the first one in http://www.scholarpedia.org/article/Bekenstein-Hawking_entropy which relates entropy to the planck length (to bring it back on topic of TFA). The actual conversion factor is left as an exercise to the reader ....

Yes, of course informational entropy vs thermodynamic entropy as in http://en.wikipedia.org/wiki/Introduction_to_entropy, but the one is a function of the other per black hole theory. (http://www.scholarpedia.org/article/Bekenstein-Hawking_entropy)
This is also clearly the reason why we perceive intelligence as hot .... http://www.mediamarksurveys.com/playboy/

So .... how hot is the Library of Congress anyway ?

Scholarpedia looks set to address this difference, it is already quite good in its early stages. Essentially wikipedia where only scholars can edit.

Britannica is now out of date. The FLASH ADS on their site are abrasive and annoying; I will refuse to visit there site anymore due to this behaviour alone.

http://www.amazon.com/Physiology-Behavior-MyPsychKit-Neil-Carlson/dp/0205593895/ref=pd_bbs_sr_1?ie=UTF8&s=books&qid=1230187013&sr=8-1
Chapter 1 has a short item on blindsight and it's relation to consciousness. You should really read chapter 1 of this book.

IMHO, sight doesn't happen as much in the eyes as it does in the brain.

Well, that's how the brain does its thing. Your eyes, ears, nose, skin, are instruments that extend from the brain. Data that flows from your senses to your brain lacks meaning until the brain processes it.

A few posts earlier you say:
( http://science.slashdot.org/comments.pl?sid=1072951&cid=26228425 )

Without a working visual cortex, nothing from the eyes enters the brain. At all.

Blindsight does not imply that the visual cortex does not "work". Have a look at http://en.wikipedia.org/wiki/Primary_visual_cortex

Data still flows from the eyes over the optical chiasm to the primary visual cortex (the cortex around the calcarine fissure in the occipital lobe). From there, there are multiple "streams" of visual data. One of those streams is the one that "enters" consciousness.

It is the absolute certainty with which you refute the previous, and the postulation of the following near-superhuman senses that make you appear rather uninformed and quite arrogant.

Most likely, he is using sound or air pressure. Blind people can often maneuver by hearing things like subtle changes in sound of footprints, etc., echoing off of or being aborbed by walls, etc. There are also subtle changes in air pressure as you approach obstacles, and that can often be 'felt' by blind people.

Take a biological psychology course or your own medicine (your signature)

http://www.scholarpedia.org/article/Blindsight
http://en.wikipedia.org/wiki/Blindsight

This exact idea is being developed right now. Take a look at Scholarpedia. Its a direct extension of scientific publishing in its traditional sense to the wiki world.

Scholarpedia is a much better site than Citizendium. Some of the articles there are very impressive. Very much an academic exercise so you won't find a page on warhammer. It does not really do drive by editing either.

Like I said, it doesn't have a lot of content but much of the content that it does have, like the two articles I noted, is very high quality and authoritative, by experts in the field (i.e., far superior to Wikipedia).

The fields they currently list are astrophysics, computational neuroscience, computational intelligence, dynamical systems, and physics (it looks there is stuff planned for quantum field theory and related: http://www.scholarpedia.org/article/Category:Quantum_and_Statistical_Field_Theory).

So yeah, if you want to use it as an encyclopedia to search for random things, it is nearly useless. If you want to browse it for high-quality, authoritative articles by experts, it is a useful resource.

Have you seen the list of recent changes? That's the surest sign of a dead wiki.

On the topic of Wikipedia-like sites, I recently found Scholarpedia, which I imagine a lot of slashdotters might like. They don't have that much content yet, and they are currently focusing on a few fields (science- and tech-related), but I have found some really high-quality articles by experts in the field, like:

Neural Correlates of Consciousness, by Christof Koch.

Algorithmic Information Theory, my Marcus Hutter.

On the topic of Wikipedia-like sites, I recently found Scholarpedia, which I imagine a lot of slashdotters might like. They don't have that much content yet, and they are currently focusing on a few fields (science- and tech-related), but I have found some really high-quality articles by experts in the field, like:

Neural Correlates of Consciousness, by Christof Koch.

Algorithmic Information Theory, my Marcus Hutter.

You've spelled out the difference between the two paths brilliantly.

It's being done: see scholarpedia, a peer-reviewed version, if you will, of wikipedia. Never heard of it? I wonder why not! :)

nobody has ever come up with anything that works better

I think it depends how you define "better." Wikipedia has a lot of advantages, with the primary one being that it covers such a wide range of topics. However, for more specialized topics, other models may work better. Scholarpedia is a good example of this. It's based on wikipedia, except with named curators who are experts in their field assigned to control individual pages. It doesn't have the breadth of wikipedia, and it doesn't claim to, but for what it does cover it's a superior resource.

> human intelligence does depend on belief systems and knowledge,
> and those continually grow as we mature from infancy

I would not be so sure. Changes continuously - yes, to a degree. Grows continuously - no.

> But to create the equivalent of an 18 year old, you have to have
> what amounts to 18 years of accumulation of knowledge about the world

"What amounts to" is the key here. It does not need to be 18 years of learning in physical time.

I am pretty sure sleep is essential to learning. You got to have some theory for what brain does during sleep. What is it?

> and have explained my technology to my patent attorney

Patents are evil. If you are successful, you are going to get hit with patents from others, like Herbert Jaeger http://www.scholarpedia.org/article/Echo_state_network

And what will they have over Scholarpedia [scholarpedia.org]?

* Some articles for a start ;) Well okay, Citizendium doesn't have anything yet, but as you say Scholarpedia is very slow moving - only a few articles seem to have been written.
* Scholarpedia seems to mainly be "each article written by one person", where as my impression of Citizendium was that it was still a Wiki where articles are edited by a number of people - but it's a Wiki based on people who give their "credentials". This kind of ties in with the previous statement - it's hard to see how Scholarpedia will ever get anywhere all the while it just as a few academics writing articles (isn't this just Nupedia?) Citizendium can at least attempt to gain interest from a large number of contributors, which is what helped Wikipedia.
* Scholarpedia isn't free ( http://www.scholarpedia.org/article/Scholarpedia:G eneral_disclaimer - "You may display, print or download content on Scholarpedia only for academic, non-commercial use, provided that you cite Scholarpedia. You may not publish, distribute, retransmit, sell or provide access to the content of Scholarpedia, except as permitted under applicable law and as described here. You may not engage in systematic retrieval of content from Scholarpedia to create or compile, directly or indirectly, a collection, compilation, database, or directory without written permission from Scholarpedia. Nor may you mirror on your own site the home page or results pages of Scholarpedia.").

And what will they have over Scholarpedia? At least that site has some articles, and the way they produce content seems like it would be quite reliable, though slow.

I admit I haven't used that site very much, and it seems a crapshoot if a given subject will have articles right now, but will Citizendium even be worth looking at? Wikipedia already has tons of info on just about everything, though it could be inacurate. Scholarpedia will have highly accurate articles, but probably not on every possible thing. Where does this new site fit? I guess we will see...

Scholarpedia isn't really that wiki. Sure, it's based on Mediawiki wikitext, but each article has a single author, and usually a single maintainer. The information is available under standard copyright only, so it can't be integrated or improved by other sites using different processes like the Wikipedia/Citizendium symbiosis.

Maybe you could call it Scholarpedia?