Bullshit (and I say this as a compiler writer). Very few compilers do anything with data layout at all (some JVMs do, to a limited degree, because they live in a closed world) and none outside of a few research projects will replace one data structure with another. What compiler are you using that will replace and XOR linked list or a skip list with something more efficient?

The belief in the compiler as a magic box that can turn a crappy algorithm into a good one is one of the things that a computer science education is meant to disabuse students of.

What are you talking about?

No, it really doesn't. If anything it's more relevant in other languages. For example, the cost of moving values from integer to floating point register files is a significant determining factor in JavaScript compiler design. To take JavaScriptCore as an example, the typical instruction cache size was one of the key inputs into the design of the interpreter and baseline JIT - it's written in a portable macro assembly language with precisely two design goals: the interpreter must have precise control over stack layout (so that deoptimisation can work easily) and the interpreter must fit entirely in the instruction cache of a modern CPU. The baseline JIT works by constructing a sequence of (predictable, because they have static destinations) jumps to the relevant entry points into the interpreter for a bytecode sequence. Trying to do this without understanding a reasonable amount of computer architecture would lead to all sorts of issues.

Understanding how a transistor works requires quantum mechanics, but 'transistors are tiny magical switches' is enough to be able to understand how to build them up into gates, how to assemble gates into arithmetic, logic, and memory circuits, how to assemble those into pipelines, and so on.

Eventually you need quantum mechanics (or relativity, or both) to understand how anything works, but understanding the electron transfer involved in combustion is not essential to understanding how a car works. Computer science is all about building abstractions.

There are different degrees of knowledge. I don't think anyone can be a competent programmer without understanding things like caches, TLBs, and pipelines (and, in particular, branch prediction). These things have significant impacts on the performance of code - often a factor of ten. Trying to write software for some hypothetical abstract machine, rather than a real modern processor leaves you with something that has the CPU gently warming the room while it waits for data from RAM. For example, I've seen people who skipped that part of their education think that XOR linked lists and skip lists are still good data structures to use.

You're paraphrasing Dijkstra, but missing his point. Astronomers, in general, know a heck of a lot about optics. His point wasn't to excuse ignorance of how computers work (he worked on the design of the STANTEC ZEBRA and wrote an incredibly scathing review of the IBM1620, for example, so clearly knew his way around the design process), it was to point out that this is a building block.

I'd consider any computer science curriculum that doesn't cover logic gates up to building adders, the basics of pipelining, the memory hierarchy and virtual memory translation at a minimum to have seriously skimped over computer architecture. The better ones will include design and simulation (on FPGA if budgets permit) of a simple pipelined processor.

If you want to work on compilers or operating systems, to give just two examples, then you need a solid grasp of computer architecture.

Normally you have to pay back loans in the currency that they were given. If you only have another currency, then you have to find someone willing to exchange them. The exchange rate isn't likely to be very good for a little while. That said, it would probably be great for the Greek economy, as a very weak currency will make exports very easy for them.

The other problem that you're not mentioning is that, if Greece leaves the Euro then Putin will be very happy to extend trade deals to them to get around sanctions and to piss off Germany and no one wants Greece to become a satellite state of the new USSR.

The Eurozone might be able to absorb a Greek exit in purely financial terms, but in psychological terms it would be a disaster. If Greece leaves, then that provides a precedent for countries leaving. Investments in Euros are based on the premise that the Euro is backed by a large economic base and if countries can leave that base then there's a lot less of an incentive to use the Euro. That's likely to lead to a drop in liquidity in the rest of the Eurozone, which would make leaving an even more attractive bet for some of the weaker economies.

The current size of the human head is limited by the pelvic size of women (humans already have a fairly high rate of mortality among newborns and mothers for unaided births compared to other species as a result). A more interesting approach would be to delay the age at which the head stops growing, though that would also need extra skeletal scaffolding to carry the larger head around, improvements to the cardiopulmonary system to keep it supplied with blood, and so on. Basically, the human brain is about as big as you can get it with small incremental changes to a hominid - you're going to need more than a few tweaks to get it a lot bigger.

Oh, and this isn't Twitter. You don't need to say '@GloomE' - we can tell from the fact that you replied to his post that you replied to his post.

When TFT monitors were new, having a couple of stuck pixels was normal and manufacturers said up-front the number that they considered normal. Typically it was under five, but for some of the cheap panels it was more. Apple had a fairly complex policy (it depended where on the screen they were and stuck-on counted as worse than stuck-off). They would refuse to accept returns for stuck pixels if you had fewer than this number (although, in the UK at least, you could return without giving a reason under consumer protection laws and then get a new one).

By about 2005, dead pixels were mostly gone from things that made it to consumers, but 2000-2005 there was a gradually decreasing acceptance for them.

I don't think you're replying to anything that I actually wrote, you're just attaching two unrelated sentence fragments...

It is a bit surprising. I have an ASUS TransformerPad TF700, which is pretty close to the ultrabook market segment and came with a 1080p screen. I've had it for a few years and the newer model comes with an even better screen.

Be a bit careful about that. The process for manufacturing TFTs is subject to the same rules as other semiconductor fabrication processes. If you double the area then you double the probability of an impurity resulting in a stuck pixel (and, these days, consumers don't accept even a small number of stuck pixels). This lowers the yields. As the feature size (i.e. pixel size) goes down the size of impurity that won't cause damage goes down too. This is why we got 225dpi screens on phones cheaply back in 2005, when IBM was still selling a 23" 225dpi screen for around $10K - the yields once you scale them up get painfully low.

It's also the reason why printers and so on have much nicer screens than they used to: the ones that don't make spec for tablets and phones are sold to consumer electronics vendors very cheaply.

I wouldn't be surprised if you want a higher resolution for document editing and web surfing than for gaming. Having a good (and, importantly, consistent) frame rate for games matters a lot, but resolution quickly hits diminishing returns. The thing I notice most with the retina display on my laptop and the 4K display on my desk at work is that text is a lot crisper. When you're spending a lot of time reading text, the higher resolution make a big difference. For games, I usually set the resolution to a quarter of native res (half each dimension) and don't notice the linear upscaling.

I've never seen a journal require that I cite my funding sources, but most grants require that you put some boilerplate that they provide in the acknowledgements containing both the funding body and the grant number. And it's usually a good thing to do, because if you want to ask the funding body for more money in the future then being able to point at a number of papers that were funded by them in the past is helpful in showing that you spend the money well.

Not sure if this is true on the other side of the pond (though I'd be surprised if it isn't in the rough ballpark), It typically costs about double someone's salary to employ them in a university (office space / equipment, part salaries of admin staff, technicians, tax obligations and so on). That means that, assuming that the $120K/year is paid to him as research grants and not a gift, it allows him to pay a salary of about $60K/year. On our pay scales, that's near the top end of what we pay postdocs and the low end of what we pay lecturers (associate professors, I think, in US terminology). Any equipment that you might need for a particular project, plus travel expenses, are extra.

It's a nice amount to have, but it's not enough to fund a faculty member full time. If it's guaranteed funding over 10 years, then it's definitely worth chasing. If it's money that just turns up as a gift, that's great. If it's something that requires the normal grant application process, then it's probably less attractive than normal funding bodies.