"That is not a business decision, that is a technical decision where you try to come up with the most universal and correct to spec answer you can. You are not shaping the business with this decision, you are trying to shape your solution to the business. "

I'm sorry sir. I'd love to do that for you, but the computer won't let me.

IOW: Your belief that software design doesn't shape and make business decisions stems from a lack of forward-sightedness.

Of course they do. Real Engineers design up front, before implementing. We understand the implications of our decisions. We optimize. We know that there are many orthogonal factors to consider in doing this. Shoud we optimize with an emphasis size or speed? If we optimize for size, how will that decision effect scalability and the ability to add functionality we may not have originally considered, or that the original design specification didn't call for?

Anybody who thinks that Engineers don't have a major impact on the entire business model have never worked in the real world, or have no idea the impact we have. "Why do we do thing X even though it no longer makes sense? ... because they system won't work if we don't, and it would cost too much and be too risky too change it!.

Documented by whom? Plenty of people claim "documented" when they're really just pulling stuff out of their ass.

Does not matter what you did or didn't do- for all you know, the guy he really wants is standing behind you. By hitting the ground, you clear his field of fire.

OpenCL is highly specific in application. Likewise, RDMA and Ethernet Offloading are highly specific for networking, SCSI is highly specific for disks, and so on.

Well, since the CPU already specializes in general-purpose serial computation, other nodes in a heterogenous environment must logically specialize for either generic parallel computation or specific applications, otherwise you have just plain old SMP.

But it's all utterly absurd. As soon as you stop thinking in terms of hierarchies and start thinking in terms of heterogeneous networks of specialized nodes, you soon realize that each node probably wants a highly specialized environment tailored to what it does best, but that for the rest, it's just message passing. You don't need masters, you don't need slaves. You need bus switches with a bit more oomph (they'd need to be bidirectional, support windowing and handle multipath routing where shortest route may be congested).

That describes neither how this heterogenous network of equal nodes would function (how do you dispatch tasks to nodes without the dispatching node becoming de facto master) nor what advantage it would have over current model (heterogenous network of nodes with some specializing in overall control). In fact it sounds a lot like buzzword bingo.

Above all, you need message passing that is wholly target-independent since you've no friggin' clue what the target will actually be in a heterogeneous environment.

You mean like the extension card mechanism PCs have had from the very beginning? Also, SATA seems to be remarkably uncaring of whether the device on the other end stores information on spinning disks or in electric capacitors.

Fuck the muslims! Seriously. FUCK THE MUSLIMS! They want global domination, from the Middle East, to Europe, the Americas, Russia and yes, China too.

How is that any different from, say, evangelical christians? Stop exporting your own brand of religious evil before you start casting stones on other people.

Anti-american muslims? That sir is a badge of honor!

Depends. Is it because they "hate your freedom"? Or is it because you keep propping up dictatorships and meddling in bloody wars in Middle-East? Which one do you think is more likely?

I hate to breal it to you, but students who choose their college based on complaints that they can't use the Facebookz to get help with their incompetence are not the kind of students any self respecting university wants.

Yes. I missed the point on that one :-( Thanks for pointing it out!

"Don't kid yourself, it is their equipment, everything on it belongs to them."

I can't speak for you, but I have root on all my systems. I have to have it to do my job. You may not have control over your system, but I certainly do.

If we have this much surplus, clearly we're buying too much.

Not really. US's tactic is limiting casualties through high-tech warfare, and technology marches on. If you want to stay on the cutting edge, you'll constantly be replacing still-functional hardware with newer. This isn't limited to the military, of course, but is something all too familiar from the PC world.

A bigger problem is that giving military hardware to the police will eventually make the police into a domestic army. Is this desirable?

When they say "stop, I just want to talk" they mean "lay down face down on the ground with your hands behind your back"

Come to think of it, that's damn good advice no matter what color you are.

Haven't we heard this story line a few times before?

Finally, if we dump the cpu-centric view of computers that became obsolete the day the 8087 arrived (if not before), we can restructure the entire PC architecture to something rational. That will redistribute demand for capacity, to the point where we can actually beat Moore's Law on aggregate for maybe another 20 years.

Please explain how your vision is different from, say, OpenCL?

He also showed that one particular thing was absolute, if you recall.

Nope. Einstein showed consequences of the speed of light being a constant of nature. He didn't show or even predict that it was one, that was done by Maxwell's equations and various attempts to measure Earth's velocity relative to luminous aether (which turned out to be "zero").

And as it happens, one of those consequences is that timewise and spacewise distance are relative.

You never have 50,000 death per year in the US to coal.
Perhaps 5 to 10 in the long time average due to mining accidents. I really doubt the total number of workers mining coal is close to that number.

As you surely know, coal plants are huge polluters and pollution causes health issues, which in turn add up to early deaths, even if we ignore damage done to environment.

But then again, opposing nuclear power is not really about protecting humans or nature, now is it? It has long since turned into politics, where opposition is based more on identity than rational calculation of risks and rewards of various options. And who knows, perhaps being hit by the double-whammy of full-power climate change and energy crisis simultaneously will finally teach humanity to not treat important decisions as tribal identifiers. It's something we must learn before we venture beyond this planet, since the cost of irrational stupidity will continue getting higher. But I fear the lesson will be exremely painful, even by the scale of these things.

And: fix your damn mining safety issues instead of blaming it to 'coal', mining of uranium is only marginally more safe.

Thousandfold decrease in mining causes a thousandfold decrease in mining-related deaths, even before factoring in such details as coal being highly flammable and uranium being not. Also, unlike coal, uranium can be extracted from seawater, so with it we could theoretically eliminate mining altogether.