The thing about the tax (and likely why it remains unknown) is that by paying the tax, that legitimizes my using it for piracy; I've already paid the RIAA. Why should I pay the tax to compensate for copying, and then be told that I can't copy! If I pay for something, I want it, and I have right to it.
This sort of tax is bad PR. Unfortunately, I guess it is cheaper to use the money collected in this way to pay for some glitzy ad campaign to offset that PR.
The guy was pretty clear in recognising that he was generalising and that some people without degrees were likely smarter than many people with.
However, I think we can all agree that level of education is probably a good predictor of intelligence. One that will be wrong at times, but more often right.
What is the escape velocity of the moon? 1/6 of the earth? which would be ~ 2 km/s IIRC.
I'm thinking telepresence fabrication plants and a railgun. What is the roundtrip time to the moon? A couple of seconds? That's a bit much for telepresence, but fine for semi-autonomous ops.
Since you seem to have some figures at your fingertips, what is the cost breakdown of a mission? For X million dollars total cost, how much is payload manufacturing and how much is deliviery? half and half?
Would it make sense to put a big ass computer/communications relay in perpetual orbit around mars? One that could be reused from mission to mission? The idea is that you mount one big mission to deliver the relay with smarts into a near marssynchronous orbit. Then you just start trowing cheapo hardware at the surface. The hardware is cheap cause it has to be controlled remotely from the relay.
This would only work if delivery was a lesser part of the cost of a mission, so it made sense to cheapen the payload.
The thing is that to do VLIW well, you need long snippets of straight line code. You get these either by unrolling loops, inlining/duplicating code, and generally transforming the shit out of your program, or from dynamic execution traces.
Once you have long unbranching intruction sequences, you can then start reordering and peepholing your mini-instructions into groups that can execute simultaneously. These are the VLIW instructions.
This is why crusuoe has some level of ILP ('cause it has the runtime information to make long blocks) and why GCC could have a hard time (I don't know that it does, taking your word on that) generating good VLIW code ('cause the language it compiles is generally a bitch to transform safely -- this applies both to C/C++ and the internal representation, I'd think).
You know every so often, a technology comes along that just blows me away. I apply it everywhere, it becomes my favorite hammer (in that everything starts looking like a nail).
Dynamic software translation is such a technology. Yes I know that it completely failed to do anything for the Alpha's NT penetration (remember the FX!32 dual mode chip?) but we'll put
that down to poor marketing (did you ever see one system with it?).
This is how the revolution will be telecast. Apple did it. win/tel will do it too. While I love AMD for giving intel credible competetion, I do have to admit that I think sacrifices to binary compatibility are looking like a worse and worse strategy.
This is the optimal time to introduce a new architecture. When apple changed to PPC, they were switching to a much faster chip, so were able to claim significant speedups vs the old architecture. Win/tel doesn't really have that option (requiring longer lead times to see significant speed iumportovements over the current arch), but we finally have a plateau in speed requirements.
Games are bottlenecking on 3d hardware,
office suites are fast enough,
our UI is fast enough and there are no computationally expensive metaphors on the horizon
This plateau means that a new architecture can be introduced on obsolesence alone -- the first couple of generations don't need to compete on speed. As more and more applications get ported to native code, consumers will again see speedups.
Of course, the new architecure would offer speedups to their big-money early adopters -- java backends are already architecture independent. Just introduce a native java VM along with the chip.
So if there ever was a time to throw out cruft it is now, before the next wave of innovation starts. Go to a faster architecture now, get back-compat by emulation/translation.
Everyone wins; server users dump money into native JVMs and get speedups soon. Consumers wait and by the time the next wave of power hungry apps comes along, everyone has "gone native".
Notice that this entire argument is based on the assumtion that the plateau will last long enough to give most of the industry a chance to go native. So I ask you; what technologies are out
there waiting to use the power? Natural language speech recognition? Good AI in games? Immersive 3d environments?
picking nits, cause I'm hungry and feeling cantankerous. also, I'm making this up, so add salt to taste.
VHS was cheaper (licensing, consortium) I think, but the real reason it won out was... pr0n. Sony had a really tight grip on the betamax licence, and wouldn't allow questionable movie producers licence to use it. Or something like that.
Python is truly a wonderful prototyping language, but as soon as any project I use it for (tend to be home-sys admin things) I always get bitten by the global/local scoping rules (*).
I would cry tears of happiness if it only had lexical scoping. My "I'll just fix it" threshold is pretty high, but I've just about to reached it.
Johan
(*) variables in methods are global if you only read from them in the body, but if you assign to them in the body, they become global. Scoping is dynamicish, allowing you to declare the global afterr you have declared it in the method, as long as it occurs first in the runtime. The semantics are way to ad-hoc, no?
'nother thought. The argument for space borne mirrors is that the atomosphere gets in the way; even w/ adaptive optics to correct for effects, I'd imagine this get alot harder to do for a large mirror.
Now, a large mirror is heavy. You can't really take it appart to send it up into space, 'cause assembly would be a hassle. A liquid mercury mirror wouldn't have that problem. You can assemble the bucket in space, fill it with mercury carted up during a number of missions. Add some heating elements and you're all set.
or at least you would be if the whole point of the excersise wasn't to use the interplay between gravity and centripital force to form the mirror.
When you're weightless in orbit, there's not much gravity to play with. (well, there is gravity, it's just counterbalanced... you know what I mean)
Bummer!
Constant linear acceleration might work. Or just put it on the moon. Thoughts.
I was just about to suggest freezing. Condensation would be a bitch. You'd need an encolosed system with nitrogen. which incidentally would solve the mecury-will-kill-you problem.
I had thought that they were suing for lost income. When people saw how cute the cube was, they completetly stopped buying the old style computers, three days before apple was going to disclose the cube.
Thus the preemptive disclosure could have cost apple 3/365 ~ 1% of their annual sales. Now it wasn't that bad of course, not everyone wanted a cube and not everyone who would have been swayed by the leak saw it "in time".
I think most small guns (ie the targets of malicious lawsuits, or "slap suits as they're called) would agree, but they aren't the ones lobbying the decision makers.
It's just that there is something inherently "wrong" about a system administered by the same people who stand to gain from it being inefficient and slow.
I mean what if all the programmers in the world decided to start writing buggy software so that they could be indefinitley employed fixing it? That would be completely insane, no?
Speaking of plumes, does anyone have numbers on the average propagation speed of odors in still room temperature air at sea level?
It would involve mean free paths and random walks, prolly. It just seems to me that it is _really_ fast, and I've always wondered how fast exactly. This seems like the place to ask, eh?
Slashdot Mirror
I would be scared sh*tless to go up in that thing. I want one! I'll commute to work in it.
I would love to see someone frame-capture starwars and then run an ascii-art converter on it. Full length! gzip as a movie compressor!
The SW asciiart movie is hand animated, IIRC.
What about a ramjet? No moving parts at all. But you'd need someone to tow it up and let it glide/fall until it reaches sufficient velocity.
<grin>
The thing about the tax (and likely why it remains unknown) is that by paying the tax, that legitimizes my using it for piracy; I've already paid the RIAA. Why should I pay the tax to compensate for copying, and then be told that I can't copy! If I pay for something, I want it, and I have right to it.
This sort of tax is bad PR. Unfortunately, I guess it is cheaper to use the money collected in this way to pay for some glitzy ad campaign to offset that PR.
but why don't you just recalibrate the moderation system so that 5 becomes the really super highest, and 4 the normal highest score?
The guy was pretty clear in recognising that he was generalising and that some people without degrees were likely smarter than many people with.
However, I think we can all agree that level of education is probably a good predictor of intelligence. One that will be wrong at times, but more often right.
Lighten Up!
What is the escape velocity of the moon? 1/6 of the earth? which would be ~ 2 km/s IIRC.
I'm thinking telepresence fabrication plants and a railgun. What is the roundtrip time to the moon? A couple of seconds? That's a bit much for telepresence, but fine for semi-autonomous ops.
Since you seem to have some figures at your fingertips, what is the cost breakdown of a mission? For X million dollars total cost, how much is payload manufacturing and how much is deliviery? half and half?
Would it make sense to put a big ass computer/communications relay in perpetual orbit around mars? One that could be reused from mission to mission? The idea is that you mount one big mission to deliver the relay with smarts into a near marssynchronous orbit. Then you just start trowing cheapo hardware at the surface. The hardware is cheap cause it has to be controlled remotely from the relay.
This would only work if delivery was a lesser part of the cost of a mission, so it made sense to cheapen the payload.
Expand the roman numerals to get
1337 5kr1p7 k1dd13 est
Inscription found at Pompeii: Nero MCCCXXXVII VkrIpVII kIddXIII est.
now that is genius.
(also response to #157)
The thing is that to do VLIW well, you need long snippets of straight line code. You get these either by unrolling loops, inlining/duplicating code, and generally transforming the shit out of your program, or from dynamic execution traces.
Once you have long unbranching intruction sequences, you can then start reordering and peepholing your mini-instructions into groups that can execute simultaneously. These are the VLIW instructions.
This is why crusuoe has some level of ILP ('cause it has the runtime information to make long blocks) and why GCC could have a hard time (I don't know that it does, taking your word on that) generating good VLIW code ('cause the language it compiles is generally a bitch to transform safely -- this applies both to C/C++ and the internal representation, I'd think).
Johan
Dynamic software translation is such a technology. Yes I know that it completely failed to do anything for the Alpha's NT penetration (remember the FX!32 dual mode chip?) but we'll put
that down to poor marketing (did you ever see one system with it?).
This is how the revolution will be telecast. Apple did it. win/tel will do it too. While I love AMD for giving intel credible competetion, I do have to admit that I think sacrifices to binary compatibility are looking like a worse and worse strategy.
This is the optimal time to introduce a new architecture. When apple changed to PPC, they were switching to a much faster chip, so were able to claim significant speedups vs the old architecture. Win/tel doesn't really have that option (requiring longer lead times to see significant speed iumportovements over the current arch), but we finally have a plateau in speed requirements.
This plateau means that a new architecture can be introduced on obsolesence alone -- the first couple of generations don't need to compete on speed. As more and more applications get ported to native code, consumers will again see speedups.
Of course, the new architecure would offer speedups to their big-money early adopters -- java backends are already architecture independent. Just introduce a native java VM along with the chip.
So if there ever was a time to throw out cruft it is now, before the next wave of innovation starts. Go to a faster architecture now, get back-compat by emulation/translation.
Everyone wins; server users dump money into native JVMs and get speedups soon. Consumers wait and by the time the next wave of power hungry apps comes along, everyone has "gone native".
Notice that this entire argument is based on the assumtion that the plateau will last long enough to give most of the industry a chance to go native. So I ask you; what technologies are out
there waiting to use the power? Natural language speech recognition? Good AI in games? Immersive 3d environments?
picking nits, cause I'm hungry and feeling cantankerous. also, I'm making this up, so add salt to taste.
VHS was cheaper (licensing, consortium) I think, but the real reason it won out was... pr0n. Sony had a really tight grip on the betamax licence, and wouldn't allow questionable movie producers licence to use it. Or something like that.
anyway, betamax carts are smaller, IIRC
they become global
local.
preview, what's that?
White space schmight place.
Python is truly a wonderful prototyping language, but as soon as any project I use it for (tend to be home-sys admin things) I always get bitten by the global/local scoping rules (*).
I would cry tears of happiness if it only had lexical scoping. My "I'll just fix it" threshold is pretty high, but I've just about to reached it.
Johan
(*) variables in methods are global if you only read from them in the body, but if you assign to them in the body, they become global. Scoping is dynamicish, allowing you to declare the global afterr you have declared it in the method, as long as it occurs first in the runtime. The semantics are way to ad-hoc, no?
I'm thinking of getting a non-working cube on the cheap to reuse the case. Anyone know which mobo size it accepts?
'nother thought. The argument for space borne mirrors is that the atomosphere gets in the way; even w/ adaptive optics to correct for effects, I'd imagine this get alot harder to do for a large mirror.
... you know what I mean)
Now, a large mirror is heavy. You can't really take it appart to send it up into space, 'cause assembly would be a hassle. A liquid mercury mirror wouldn't have that problem. You can assemble the bucket in space, fill it with mercury carted up during a number of missions. Add some heating elements and you're all set.
or at least you would be if the whole point of the excersise wasn't to use the interplay between gravity and centripital force to form the mirror.
When you're weightless in orbit, there's not much gravity to play with. (well, there is gravity, it's just counterbalanced
Bummer!
Constant linear acceleration might work. Or just put it on the moon. Thoughts.
I was just about to suggest freezing. Condensation would be a bitch. You'd need an encolosed system with nitrogen. which incidentally would solve the mecury-will-kill-you problem.
Johan
erm.
litteracy in the asian continent is usually quite high.
and for n=2, you just use XOR. Not much redundancy in that tho.
I thought that GETs were supposed to be idempotent? Then they should be indexable.
Additionally you might need is (probably already exists, there is nothing new under the sun) epiration dates for GETted pages.
Johan
I had thought that they were suing for lost income. When people saw how cute the cube was, they completetly stopped buying the old style computers, three days before apple was going to disclose the cube.
Thus the preemptive disclosure could have cost apple 3/365 ~ 1% of their annual sales. Now it wasn't that bad of course, not everyone wanted a cube and not everyone who would have been swayed by the leak saw it "in time".
but still
I think most small guns (ie the targets of malicious lawsuits, or "slap suits as they're called) would agree, but they aren't the ones lobbying the decision makers.
paranoia. paranoia
It's just that there is something inherently "wrong" about a system administered by the same people who stand to gain from it being inefficient and slow.
I mean what if all the programmers in the world decided to start writing buggy software so that they could be indefinitley employed fixing it? That would be completely insane, no?
Oh.
I'll shut up now
Speaking of plumes, does anyone have numbers on the average propagation speed of odors in still room temperature air at sea level?
It would involve mean free paths and random walks, prolly. It just seems to me that it is _really_ fast, and I've always wondered how fast exactly. This seems like the place to ask, eh?