The cluster you use doesn't have to be in the University the research group is housed in. Many clusters are available to researchers worldwide; you just upload your data/code to a processing queue and it gets run. You can remote-login to monitor the status, restart jobs, etc. You have quite a bit of control.

In fact, if the research in question is "high quality" and not proprietary then you can get access to various clusters for FREE. It's hard to beat free in terms of bang/buck. For instance, the US Department of Energy runs various computer clusters within "user facilities" (other funding agencies in US, Europe and elsewhere have similar programs). What this means is that you submit a proposal/request where you describe the research you're doing and what kind of resource you need (in this case, routine access to a computer cluster). If the proposal is highly rated (externally peer reviewed) then you're allocated access for free. In addition to getting access to the cluster itself, you get "access" to the experts who run the cluster--their expertise in optimizing and parallelizing code is extremely valuable.

I understand that having immediate access to computing power is useful. But if you're on a shoestring budget then something's gotta give, and using pre-existing clusters is a very efficient way to spend money. In the case of user clusters, if you can get free access then you can use a mixture of a smallish in-lab cluster and occasional access to the large-scale cluster. This is so easy to do (and did I mention free?), there's almost no reason not to try. (Yes, the DoE accepts proposals internationally, so there's no problem there.)

Disclosure: I work for the DoE, so I guess I'm biased. Here are some links that might help:
http://www.bnl.gov/cfn/facilities/Theory_and_Computation.asp
http://computing.ornl.gov/
http://www.alcf.anl.gov/

They are indeed mimicking Apple. And making the same mistakes, in my opinion.

"Every single pixel of your beautiful screen is for your app," said Harris. "You're just immersed in the content."

As I said when OS X Lion was released, I think this push towards full-screen apps is a move backwards. Yes, having the app fill the screen makes a lot of sense for smartphones and tablets, where screen/interface space is limited and you're typically focusing on a single task at a time. But on a desktop?

The whole point of a multi-purpose desktop computer is to be able to do a myriad of things, and more importantly to combine all the various resources/applications together in powerful ways. I want to be able to have a web-page reference document open while I code something, or copy-and-paste something from a spreadsheet into a text document. I want to be able to cross-compare multiple graphs/images/whatever at the same time. To do all this, I need to be able to tile, stack, and move windows on my screen. Endless alt-tabbing just doesn't cut it.

With desktop monitors getting bigger and bigger, fullscreen apps just don't make sense. Even maximized apps don't make sense: your mouse has to travel ridiculously far to get from content to controls if you make your app fullscreen on a 30-inch monitor. (There are of course times when you want a single app fullscreen; e.g. photo editing on a large monitor gives you a much better view of the content.) One of the main advantages of modern large monitors is the ability to have multiple apps open at once, without them blocking each other or being ridiculously constrained. Why are we throwing away these advantages?

I'm fully aware of the cognitive science research on multi-tasking (specifically, that people are bad at it and that focusing on a single task for a longer period of time has big advantages). What I'm questioning is whether any non-trivial task can really be accomplished using a single application. We should be optimizing our user interfaces to maximize the efficiency and focus on tasks and workflows: not boxing ourselves into stripped-down full-screen apps.

I'm not sure how rhetorical that was... but numerous alternatives to the current copyright scheme have been proposed. Obviously there is considerable debate and disagreement, but there are actually some (reasonably independent) studies which show that copyright terms of 7-12 years maximize social good (lower and the incentive isn't enough, higher and the returns are insufficient to justify protection, etc.). My point is that there are, certainly, alternatives. "No copyright" is also an alternative, though not necessarily the best one.

In my own opinion, I think the options are, going from best to worst:
1. Social contract copyright: the protection is commensurate to the social benefit offered by the copyright holder, and terms are reasonable. E.g. free & open-source material is protected for 15 years, partially open-source (can see all source material, which is eventually public domain, but derivative works are not allowed until then) is protected for 10 years, all-rights-reserved copyright is protected for only 5 years.
2. Commercial-only copyright: enforcement/lawsuits against companies and those who turn a profit from infringement are allowed; individual citizens are exempt.
3. Reasonable-term copyright: just like status quo but with shorter terms (7-10 years).
4. No copyright of any kind (alternative business models will arise, such as patronage and donation-based works).
5. Taxed copyright: to maintain a copyright, companies must pay taxes (a sort of 'intellectual property tax'). The tax rate can increase year by year. Thus companies will only keep paying as long as it is profitable, and works will naturally fall into the public domain at some point.
6. Subsidizing art/entertainment production through taxes.
7. Paying artists/entertainers through licensing fees and levies (on blank media, radio stations, venues that play music, etc.).
8. Status quo copyright (fairly long terms, considerable enforcement).
9. Zealous copyright (even longer terms, massive enforcement campaigns).
10. Infinite copyright.

My point is that alternatives exist, which are likely better for society overall (but not necessarily better for the select group of people profiting off of the status quo). Moreover it's entirely possible that "do nothing" is a better option than the current system. (Doesn't mean it's the best possible option, but far too often people don't even consider the possibility of simply having no low/rule/regulation in a given area.)

My main concern with implementing this, even if it always remains "opt-in", is that you're basically creating a API that censorship tools can exploit.

It would be possible for a proxy or national firewall to redirect all requests for Wikipedia through a particular Wikipedia account where they had set the "hide obscenity" flag. So all users within that country, by default, wouldn't see the "offensive" stuff. The hard work of tagging, categorizing images, and rewriting the HTML, would have been done by Wikipedia itself. While some may view this as a good thing, since each country gets to set its own decency standard (and 'partial access to Wikipedia is better than an outright ban'), I think it is very dangerous to provide tools that make national censorship easier than it currently is.

You could of course try to design the system so that users can easily click "show image" whenever they want, or try to design it in such a way that it is difficult for proxies and national firewalls to exploit... But I would think that rather than getting into such a messy arms race, Wikipedia should just stick to their current policy: which is to include images that are appropriate and relevant to the given article. If people are offended by a factual account of reality, then really that's their problem.

Okay. But then the question becomes: "Why is such a dumbed-down article being posted to a technology site?"

I read TFA. It is terrible. It includes gems such as:

... have succeeded in developing an 800Mbps (Megabits per second) capable wireless network (WLAN - Wireless Local Area Network) by using nothing more than normal red, blue, green and white LED (Light-Emitting Diode) light bulbs. It apparently only takes "a few additional components" to turn regular LEDs ... into an affordable ultrafast wireless network.

Emphasis is added, to emphasize how these two sentences directly contradict each other. Did the author even read what they wrote?

More importantly, the article nowhere actually mentions what is new or different about this technology/process (or what the "additional components" might be). There is no way to discern if this is a breakthrough/innovation, or simply a standard configuration of off-the-shelf components. No indication about why tech-geeks should care. They claim a world record on data transfer, but don't provide any explanation, graphs, technical details... or, you know, evidence.

I'm not normally the Slashdotter who complains about this site going down the tubes. As far as I'm concerned this site has always had a mix of good and bad posts, and continues to have both good and bad posts. This particular article is ... terrible.

Perhaps there's an analogy to being drunk. Some people claim that you only do while drunk things that you wanted to do anyways. But that's akin to saying that our personal inhibitions are not part of what define us. My ability to restrain and control my thoughts are a critical part of who I am; I have a myriad of thoughts, ideas, and opinions. Some of them are in conflict. What 'gets to the surface' and becomes expressed is based on a balance of emotion, restraint, rational thought, etc. That's what defines me. Tweak part of that equation and you'll get a slightly different "me".

When it comes to being anonymous online, the context once again reshapes who we are. It's still me, still my thoughts, but the balances have been tweaked and so the emergent personality is somewhat different.

What's my point? Only to remind us of the fairly obvious point that the external persona that we project is affected by many things (both internal and external). This doesn't absolve us from responsibility, e.g. excuse us being meaner online. Quite the opposite: we must be aware of how context affects our behaviors, and adapt accordingly. We are ultimately responsible for whatever effects we have on others.

What you're describing will not work. You're trying to violate the laws of physics, similar to proposals of perpetual motion machines. It's a neat thought experiment, in order to identify the problems, but it won't work in the real world.

From basic Newtonian mechanics, we know that for every force there will be an equal and opposite reactive force. A closed system will not be able to achieve motion without an external force: either a force applied to other objects (e.g. pushing against the ground, or pushing against (a.k.a. 'blowing') a fluid like air or water) or by ejecting matter (as in a rocket).

Specifically regarding your design: As I understand it, you basically want an object where internally forces are applied to inclined planes, in order to push the planes 'upwards'. You imagine that this can be done in a way where there is no corresponding opposing force also pushing the object downwards. You try to get around this problem by imagining a decoupling where internal masses are momentarily not touching the main mass: so you have one piece that fires a 'bullet' horizontally, which hits the inclined plane (pushing it upwards). You imply that this means there is no corresponding opposing force. However you mention offhand that you will recover the 'bullets' and reuse them. But if the bullet hits the inclined plane, and pushes it upwards, then the bullet will be correspondingly deflected downwards. When the bullet hits the recovery mechanism, it will impart to it a downward force equal and opposite to the upward force that the inclined plane felt. The two forces will cancel out: the plane is pushed up, the recover mechanism is pushed down.

You can imagine putting the recovery mechanism further away from the inclined plane. But, at best this just creates a time lag between when the inclined plane is pushed upwards, and the bullet-recovery mechanism is pushed downwards. So the vehicle will jolt up-down but on average will stay in the same place and thus will not hover against the constant force of gravity. This is inescapable since the planes and the recovery mechanism are mechanically coupled to one another. The only way to solve this is to remove the recovery mechanism, and let the bullets shoot out the bottom of the object, so that the planes are pushed upwards and the opposing force is carried away by the bullets, out of the object. Of course 'flying' by shooting a gun downwards is generally inefficient, which is why we've invented things like helicopters, which push air downwards instead. That way you don't have to carry around a bunch of bullets; you just use the mass and hydrodynamic properties of the fluid you're flying through.

It's that easy, is it?

Giving human rights to constructs that have some human DNA would needlessly complicate a whole slew of matters. It also doesn't resolve things as clearly as you think it might: e.g. comatose or terminal humans certainly have human DNA, yet there is lively debate about how far their rights extend and whether we should keep them alive at all costs or let them die. Unborn embryo and fetuses have human DNA; and again society hasn't decided how many human rights they ought to have. A severed human limb, or surgically removed organ have fully human DNA yet clearly don't have human rights. Why should a non-sentient mass of tissue with some (or all) human DNA have human rights? Primates already have significant genetic overlap with humans, and substantial intelligence. Should they be accorded full human rights? Trying to treat constructs/tissue/animals as human doesn't make things clear at all. It raises more questions than it answers.

Besides, your maximalist solution amounts to saying that we shouldn't do these things: if every creature with partial human DNA is considered a full human, then it would be presumptively unethical to create such creatures for the purposes of research or tissue harvesting (even if they are, e.g. no more thinking and feeling than the cattle we happily slaughter today). You've circumvented the debate and gone straight for the "we shouldn't be doing this" conclusion. If that's really the conclusion we want, then an outright legal ban would be simpler and clearer.

But really, we should probably pursue these technologies in an ethical way. This will require some deep thought about ethics. We will need to decide more clearly what gives people their rights. Is it just human lineage? Is it our unique mental capacities? Our sentience? Our ability to feel pain? Our ability to communicate? A combination of all of these? I believe there are ways to pursue these research topics ethically; and we need to figure out how. But I don't think simplistic blanket rules will get us very far.

That's a useful list. There certainly are a lot of improvements in that list. Two that sound bad, however:

Having a lock feature is nice. But auto-locking the document seems like a nuisance. There are lots of documents that I edit on-and-off on a monthly or yearly schedule. I don't want to have an extra click just because I haven't touched that file in awhile. In fact, since OS X is pushing more and more for auto-backups and auto-versioning, auto-locking seems unnecessary. If you can always revert changes, then there's no need to give the user an extra 'are you sure you want to change this document' roadblock. To me, it's inconsistent for them to be pushing auto-saving/backup/versioning but also have auto-locking.

Apple's push towards full-screen apps seems like a small step backwards. They are basically expanding on the successful UI principles from iPhone and iPad and seeing if they work on laptops and desktops. This might be useful for some users, so as an option I think it's fine. I do, in fact, go to full-screen mode in Firefox sometimes, and I can see the benefit for other applications to really 'take over', even replacing the taskbar/etc. But the thing is that it breaks consistency. On iPhone/iPad, all applications behave a certain way, so it all makes sense and you can get used to it. But Apple machines now have too many kinds of applications (widgets, normal applications, maximized applications, these new full-screen applications, plus older 'full-screen apps' like front-row). It's becoming inconsistent, with a mixture of behaviors and UI conventions. This is the opposite of what Apple's nominal interface guidelines recommend. A full-screen UI also seems very inefficient on larger-display computers (desktops). It seems that Apple is optimizing the GUI for small form-factor devices at the expense of full-size computers. Optimizing for consumption over production of content. I worry that this is part of a larger trend to over-simplify desktop computing, making it less open, flexible and powerful.

I don't think that's a step in the right direction. Those little 'fade-away lines' make sense on a mobile phone, where space is at a premium. But on a desktop or laptop, I'd rather see the scroll-bars. It gives you something to mouse towards and grab. More importantly, it gives you constant feedback about where you are within a document, as well as information about the size of the document. This is useful information that you intuitively get when reading a book (you can see the thickness of the book and how far into it you are). Removing these subtle clues from applications reduces context and leads to user errors (e.g. thinking you've reached the end of the document when you hit some whitespace). The above complaints may seem nitpicky. Clearly there is a long list of very cool improvements. (Auto-saving and auto-versioning should be standard in any modern OS!) But as with any software/OS 'updrade' there always seem to be some things that get... worse.

Taxation is in some sense unethical--it is a 'taking'. But that doesn't mean the alternative is perfectly ethical. And I'm not talking about the fact that non-taxation could lead to poverty, social distress, and other ills that would be 'worse' than taxation (both pragmatically and ethically). There are arguments to be made there, but at present I want to instead focus on another idea: money is non-linear.

Money is our book-keeping method to avoid tedious micro-bartering all the time. It is a way for us to assign value to all the things we care about (from land, to resources, to time, to talent, ...). To a first approximation, we can treat it as a linear object (work twice as long, get twice as much pay; want 2 apples instead of 1, you pay twice as much). And yet we all know that there are many non-linearities in the money system. Everything from economies of scale (actually it's cheaper to buy in bulk), to 'rich get richer' (concentrations of money can actually be more powerful than the sum of the individual quantities of money), to psychological factors (happiness as a function of wealth, .99 pricing, the low value assigned to loose change, ...), etc. The way people value money is also non-uniform (the 'value' of $100 to people living in different areas is not the same) and is non-linear: as you become richer the marginal value of each new dollar is in some sense lower (hence why a rich person can basically value a dollar less than a poor person). Money is also complicated in other ways, such as its value changing with time (inflation) and format (liquidity), money requiring stable infrastructure to retain value (hence its value is contingent), and so on. These are just example; I'm sure you can think of more.

The point is that the value(money) graph is not necessarily linear. It's not even necessarily monotonically increasing. It's complicated, and really we don't know what it 'truly' looks like. (Even defining the question is hard since it varies from person to person, from place to place, and whether we're talking about 'purchasing power', or 'utility', or 'value', or 'happiness', etc.) Money is a crude approximation for what we really care about: value. (Yes, part of my argument is that morality/fairness/etc. should really be concerned with value (how much things matter to us) and not the artificial construct of money.)

Imagine we knew what the 'right' value(money) curve really was. We could then imagine all transactions occurring in 'new money' which is a linearized form of 'current money' where a myriad of calculations go on in the background to make 'new money' linear in terms of value(money). This would be horrendously complicated (possibly impossible, due to various inhomogeneities), so it's obvious why we don't do that: we just stick to our simplistic and naive linear-money model.

This brings up two points:
1. In some sense, you can think of many aspects of our current money system as being crude attempts to tweak the current naive money model so that it makes more sense. Things like interest rates are tweaks that somewhat offset inflation. Taxation is a tweak that takes into account marginal utility and externalities that were required to give that money value in the first place. We can certainly debate about what tax system would provide the fairest 're-numbering' of the money system... but it's by no means obvious that no taxation would really be the most ethical numbering. (Again, imagine two identical worlds, one using current money, and one using the hypothetical complicated 'new money' presented above... A tax in one numbering system appears as no tax in the other. Which one is 'right'? Well, it depends whether you consider linear money 'right', or whether you think that money is non-linear.)
2. My main point, however, is that arguing that taxation is theft is unfair. Because money is so complicated (historical, contingent, non-linear, inhomogeneous, chaotic, prone to feedbacks and emergent phenomena, etc.), it's by no means obvious which model is really the most 'fair' or 'ethical'. Simply not auto-decrementing your account balance is not necessarily fair (e.g. the government could just 'tax' you by printing more money, thereby decreasing the value of what you have, which is indeed one aspect of how we're all taxed).

Again, to be crystal-clear: I'm not saying that the current tax system is necessarily optimal, fair, or ethical. I'm saying that, abstractly, non-taxation is not necessarily fair. To know whether (or how much) taxation is ethical, you need to look in detail at the monetary system in place. And non-linearities make this complicated. The naive assumption that 'no taxation == no injustice' and 'taxation == theft' relies on a naive assumption that our current linear money model is correct. I challenge that assumption.

Indeed. I run a Wordpress install that works like this. The site is password-protected and not indexed by search engines. The RSS feed uses an obfuscated URL and is also password-protected. I also added a public RSS feed where the posts/comments are essentially empty: it just acts as a 'ping' so people know there is new content. (This is for users who for some reason can't use the password-protected feed.) Everyone can posts text, links, images, videos, etc. It's all private and versatile and customizable. (Both in terms of allowing me/us to customize the site feel/content, and because users can access using whatever web-browser and/or RSS reader they like best.) Works great.

Is such a system scalable? Sorta. If you are a member of many such 'private blogs' you can certainly add them all to your RSS reader and follow them all conveniently. However this is somewhat more clunky than something like Facebook, where another entity takes care of all the admin details for you (both in terms of hosting the content and managing your social subscriptions).

I could easily imagine refining the distributed Web/RSS system... e.g. creating a standard Wordpress plugin that does all of the securing/accounts/RSS in a simple way, using crypto key-pairs instead of passwords (or OpenID or something) so that everyone can just use a single login for the whole social network, rather that creating and manging numerous accounts all over the place. You could even have some big players offer 'one-click' social-network-group-creation (Wordpress.com, etc.) while still inter-operating with those of us who want to run and manage and control our own social-network-group.

So, I think it can be done. However Facebook and now Google+ certainly provide a simpler and easier interface. This will thus attract users and a distributed model will fail to gain traction because the advantages (true privacy and control over your own content) are too nebulous for most users.

The article is wrong to suggest that the boundary layer disappears for moving surfaces. Fan blades do indeed get dust on them. However the actual work, described in the technical report, makes it clear that they are not claiming an elimination of boundary layer effects, merely a reduction of the boundary layer thickness:

This rotating heat exchanger geometry places the thermal boundary layer in an accelerating frame of reference. Placing the boundary layer in this non-inertial frame of reference adds a new force term to the Navier-Stokes equations, whose steady state solution governs the functional form of the heat-sink-impeller flow field [Schlichting, 1979]. At a rotation speed of several thousand rpm, the magnitude of this centrifugal (in the frame of reference of the boundary layer) force term is as such that as much as a factor of ten reduction in average boundary layer thickness is predicted [Cobb, 1956]. Unlike techniques such as air jet impingement cooling, the mechanism for boundary layer thinning in the air bearing heat exchanger does not rely on a process that entails dissipation of significant amounts of energy, nor is the boundary layer thinning effect localized in a small area. Rather, the centrifugal force generated by rotation acts on all surfaces simultaneously, and all portions of the finned heat sink are subject to the resulting boundary layer thinning effect. For the limiting case of flat rotating disk, an exact solution of the Navier-Stokes equation is possible and indicates that the magnitude of the boundary-layer thinning effect is constant as a function of radial position.

(Emphasis added.)

Again, the technical document makes it clear that the rotating heat sink is not coupled via a bearing to the surface it's cooling. Rather there is a very thin layer of air separating them. Naively one might think that this layer of air (generally a poor heat conductor) would become limiting, and there would be poor heat transfer from the hot plate to the rotating heat sink. However they address this:

Heat flows from the stationary aluminum base plate to the rotating heat-sink-impeller through this 0.03-mm-thick circular disk of air. As shown later in Figure 18, this air-filled thermal interface has very low thermal resistance and is in no way a limiting factor to device performance; its cross sectional area is large relative to its thickness, and because the air that occupies the gap region is violently sheared between the lower surface (stationary) and the upper surface (rotating at several thousand rpm). The convective mixing provided by this shearing effect provides a several-fold increase in thermal conductivity of the air in the gap region.

So, basically by keeping the air gap very thin (30 microns), and by substantially shearing/mixing this thin air disk, its thermal conductivity can be sufficient to transfer heat up into the rotating fins. Overall a rather clever design.

I agree a lot of junk gets posted to Slashdot. But in this case, a link was actually provided to a good technical document that answers many questions, provides schematics, and shows graphs of various performance measures.

He didn't steal the code from the other engines (it was open source). His biggest offense is denying the other coders credit.

Well, it seems that Fruit is open-source in the sense that people can look at the codebase, but it is not FOSS. The license text (see, e.g. readme in this tarfile) says:

All right reserved. Fruit and PolyGlot may not be distributed as part of any software package, service or web site without prior written permission from the author.

Indeed it looks like a commerical product that you are meant to pay for. Rajlich's engine is closed-source and also commercial. He is not making his code available, so even if Fruit were, say, released under the GPL, he would be in violation of the license. But in fact Fruit is "all rights reserved" so if Rajlich took code from it then he is blatantly violating copyright, and thus breaking the law.

I would think that the competition has a blanket ethics rule that says that you cannot win by breaking the law. So Rajlich, if he did indeed appropriate code, doesn't deserve the wins. (Yes, he obviously did ~something~ to improve upon Fruit, but he still cheated.)

TFA points out that Rajlich could exonerate himself by showing the source code, but then says that this isn't possible:

It’s a tricky situation, though: with Rybka now outlawed from the WCCC, and with the ICGA asking other tournaments to block its entry, the only real way Rajlich and the rest of the Rybka team can clear their names is to show their source code — a financially untenable move. In short, Rybka is stuck between a rock and a hard place.

This doesn't really hold up. Yes, Rajlich is trying to sell his software, so he can't open-source it to the world. But to exonerate himself he doesn't have to release the source-code to the world; he simply needs to arrange for the source code to be shown to the expert panel. As long as they can both confirm that: (1) the provided source compiles to the binary used in competition, and (2) there is no substantial overlap between the provided source and other known codebases, then he's in the clear. The expert panel doesn't have to retain copies of the source code beyond the review period (all copies could be destroyed).

So, really, it should be possible for Rajlich to demonstrate the originality of his code without releasing it or decreasing his commercial opportunities. The fact that he hasn't done this is strange. In that sense, it sounds to me like the ICGA made the right decision here.

As it so happens I played with one of these things for a few hours (friend got one for a music setup). When I first saw it, I was excited and was seriously considering buying one for myself, since I love having access to more ways to design shortcuts and streamline my computer work-flows. Alas, I was quite disappointed with it. The main problem are:
1. The buttons are flat and small (compared to a foot, I mean), making it hard to know which button you're actually touching. If you're shoe-less, then you can feel around with your toes, and figure out which button is which (though this is pretty slow). With shoes on you have no hope of knowing where you're hitting without actually looking. This rather defeats the idea of having a foot pedal for computer control. At best, you could program this to be perhaps 3 keys (by grouping the left keys, the middle keys, and the right-most 'direction-pad' to trigger three different functions).
2. The buttons have effectively no 'give' or feedback. The only way I knew I was actually pressing the buttons was seeing the triggers in the programming application. This has the unfortunate effect of causing you to press the keys really hard, so that you know you've activated them, which gets annoying and tiring very quickly.
3. No Linux support. It can probably be done with a bit of effort (I gave up getting it to work in Linux after some testing I did on OS X revealed the above flaws), but there is no official support.

Basically, the ergonomics of the device are terrible. I don't understand how this could be useful even for a musician, since they too would want something that they can find with their foot without looking, and know that they've triggered it (a traditional guitar foot pedal has nice feedback). I would not recommend this product.

Incidentally, I have the same complaint about the Fragpedal Deluxe: the buttons have essentially no give, and take too much force to activate. I'm still searching for a good USB foot pedal that has a satisfying key-like feedback and ergonomics properly designed for activation by, you know, a foot.