Fuck you, asshole. Here's what Weird Al actually said:
Tim Sloane of Ijamsville, MD asks: Al, which of these purchasing methods should I use in order to make sure the most profit gets to you: Buying one of your albums on CD, or buying one of your albums on iTunes?
I am extremely grateful for your support, no matter which format you choose to legally obtain my music in, so you should do whatever makes the most sense for you personally. But since you ASKED... I actually do get significantly more money from CD sales, as opposed to downloads. This is the one thing about my renegotiated record contract that never made much sense to me. It costs the label NOTHING for somebody to download an album (no manufacturing costs, shipping, or really any overhead of any kind) and yet the artist (me) winds up making less from it. Go figure.
Interesting historical tidbit: back in the old days, it was actually routine for nursing students to use amphetamines to stay awake longer. Nearly everybody did it in some places. You could just go get them out of the medicine cabinets. And those things were powerful.
Since the PICs have interrupts and several timers, I doubt he was talking about that.
On the PIC series of microcontrollers, you can time any code simply by adding up the clock cycles taken by each instruction and figuring in your clock rate. There's even a nice tool to do this for you. This is often handy for simple delays; sometimes you're using all the timers or you don't want to stick stuff into a bunch of configuration registers just to slow down a loop. I don't see this sort of timing being as easy when there's no such thing as a clock cycle.
There are actually processors out there with compilers which can compile a few bottleneck C/C++ functions into hardware on an integrated FPGA. This expands the CPU instruction set in application-specific ways and can, in some cases, give absolutely enormous speedups.
In other words, they're working on processors which are programmed in general-purpose languages, but which adapt their hardware to the specific program.
We stress it because it's annoying to see people get it wrong over and over again. It starts to grate on you after a while. And it just goes on and on! "If we evolved from monkeys, why are there still monkeys?", endless voices repeat.
Think about it as a bunch of programs which just have abnormally large bundles of compatibility libraries distributed with them. This means that the people who write the programs don't have to worry as much about distribution hassles, and everything is nicely self-contained in its own folder.
If you had a porta-nuke powerful enough to blow up a nuclear plant, why would you waste it on a nuke plant? Why not just attack people directly with your nuclear might?
Over here in the US, decommissioning costs are part of the cost of building a power plant; they must be paid up front, and not supplied by the government. The entire industry is wrapped up in so much safety red tape that everything costs several times more than it should; a valve for a nuke plant often costs three times more than the exact same valve for a coal plant. Nuclear power could be much cheaper than it is---and yet it's still slightly cheaper than coal, making it the cheapest form of power generation in the country.
Yo do realize that nuclear plants can't explode like nuclear bombs, right? On top of that, it's insanely difficult to get a modern nuclear plant to even melt down, let alone do anything that's actually destructive to more than the plant itself. There are all sorts of safety systems, active and passive, that Chernobyl didn't have. Not to mention the simple fact that nuke plants in the USA have several-foot-thick concrete containment structures meant to withstand a collision from a fighter jet. Chernobyl didn't have one of those.
If all of America was powered by breeder reactors, we could fulfill current energy demands for over a hundred years by running them off the nuclear waste we have in storage right now. Isn't nuclear power cool?
Our nuclear fuel reserves can last a very long time with proper reprocessing, and even longer if we use breeder reactors. Fuel for nuclear reactors is finite, yes---but so is the sun's energy. They're both practically infinite well into the future.
Also, nuclear plants to not produce pollution comparable to coal power. Nuke plants take in relatively small amounts of fuel and produce a relatively small amount of contained waste. Coal plants take in a huge amount of coal and produce a huge amount of waste, some of which is contained and some of which is vented into the atmosphere.
I agree that there aren't a whole lot of people out there who can maintain Haskell code, but your comment about the lack of standardization is misleading. There is a stable Haskell 98 standard which specifies the core of the language; the problem is that Haskell 98 is conservative, and some very useful extensions have come into common use. If you just use the GHC compiler and forget that nhc and hugs exist, lack of standardization shouldn't be a big problem.
Iteration is not inherently more efficient than recursion. Tail calls can sometimes be changed into simple gotos by the compiler (and there are some compilers which guarantee this), making recursion and iteration exactly the same in the compiled output.
Some compilers actually transform iteration to recursion and then back again.
If they're letting you use assignment (set! and friends), you could (I don't recommend it) create Scheme versions of the usual looping constructs. For example, might it be nice to be able to write
(for (set! x 1) (< x 25) (set! x (+ x 1))
(print x))
Of course, this example requires that you use some form of macros to avoid needing to write lambda everywhere, but this sort of thing can be done. Even Haskell, which doesn't have assignment, has a way to fake it.
On a more practical note, you want to become fuckbuddies with the MAP, FOLDL, FOLDR, and REDUCE functions. They encapsulate a lot of unpleasant recursive machinery in a very pleasant way.
Yes, I have read RMS talking about the "GNU/Linux" thing. You should too, so that you'll know what he's actually saying. He's being silly and embarassing about the whole thing, but he's not as psychotic about it as many slashdotters seem to think. There is some reasoning behind it, and in order to make fun of him properly, you must understand the reasoning and what the phrase "GNU/Linux" is meant to denote.
Slashdot Mirror
He's not the greedy one here.
Interesting historical tidbit: back in the old days, it was actually routine for nursing students to use amphetamines to stay awake longer. Nearly everybody did it in some places. You could just go get them out of the medicine cabinets. And those things were powerful.
Nanotubes have a very high tensile strength. When it comes to other directions of loading, they don't fare nearly as well.
The plates are needed to hold the nanotubes in place and connect the nanotubes to the outside world.
Ease up on the bashing. This is a good thing, remember?
Have it your way, coward.
Honestly, you'd probably get a much more helpful reply on the mailing list.
Compressive strength != tensile strength, as any Roman engineer could tell you.
On the PIC series of microcontrollers, you can time any code simply by adding up the clock cycles taken by each instruction and figuring in your clock rate. There's even a nice tool to do this for you. This is often handy for simple delays; sometimes you're using all the timers or you don't want to stick stuff into a bunch of configuration registers just to slow down a loop. I don't see this sort of timing being as easy when there's no such thing as a clock cycle.
In other words, they're working on processors which are programmed in general-purpose languages, but which adapt their hardware to the specific program.
*shudder*
Think about it as a bunch of programs which just have abnormally large bundles of compatibility libraries distributed with them. This means that the people who write the programs don't have to worry as much about distribution hassles, and everything is nicely self-contained in its own folder.
Simple:
If you had a porta-nuke powerful enough to blow up a nuclear plant, why would you waste it on a nuke plant? Why not just attack people directly with your nuclear might?
Over here in the US, decommissioning costs are part of the cost of building a power plant; they must be paid up front, and not supplied by the government. The entire industry is wrapped up in so much safety red tape that everything costs several times more than it should; a valve for a nuke plant often costs three times more than the exact same valve for a coal plant. Nuclear power could be much cheaper than it is---and yet it's still slightly cheaper than coal, making it the cheapest form of power generation in the country.
Please, do some research before spouting off.
We haven't had any ice ages in the past 100 years. Now, WHAT DOES THAT HAVE TO DO WITH ANYTHING? You aren't making any sense.
If all of America was powered by breeder reactors, we could fulfill current energy demands for over a hundred years by running them off the nuclear waste we have in storage right now. Isn't nuclear power cool?
Also, nuclear plants to not produce pollution comparable to coal power. Nuke plants take in relatively small amounts of fuel and produce a relatively small amount of contained waste. Coal plants take in a huge amount of coal and produce a huge amount of waste, some of which is contained and some of which is vented into the atmosphere.
I agree that there aren't a whole lot of people out there who can maintain Haskell code, but your comment about the lack of standardization is misleading. There is a stable Haskell 98 standard which specifies the core of the language; the problem is that Haskell 98 is conservative, and some very useful extensions have come into common use. If you just use the GHC compiler and forget that nhc and hugs exist, lack of standardization shouldn't be a big problem.
Some compilers actually transform iteration to recursion and then back again.
(for (set! x 1) (< x 25) (set! x (+ x 1)) (print x))
Of course, this example requires that you use some form of macros to avoid needing to write lambda everywhere, but this sort of thing can be done. Even Haskell, which doesn't have assignment, has a way to fake it.
On a more practical note, you want to become fuckbuddies with the MAP, FOLDL, FOLDR, and REDUCE functions. They encapsulate a lot of unpleasant recursive machinery in a very pleasant way.
The .edu, .gov, and .mil TLDs are pretty strict. Anything with .org is likely to be a non-profit organization. So yes, TLDs do have a point.
You could try, you know, ignoring them.
Yes, I have read RMS talking about the "GNU/Linux" thing. You should too, so that you'll know what he's actually saying. He's being silly and embarassing about the whole thing, but he's not as psychotic about it as many slashdotters seem to think. There is some reasoning behind it, and in order to make fun of him properly, you must understand the reasoning and what the phrase "GNU/Linux" is meant to denote.