Pfft. It took them long enough to get a modern kernel. Even Microsoft had decent protected memory and preemptive multi-tasking since 1996 (WindnowsNT 3.51).
And OS X doesn't have a UNIX kernel. It uses Mach.
If you look at the Gates Foundation, you'll see that he's already donated about $30 billion dollars to its capital fund already. I'd say he's well on track towards his stated goal of giving away his entire net worth by the time he dies.
So he's not just giving the money out directly, but setting up an income investment fund to ensure that the charity can continue to operate for long, long after his death, bankrupcy, etc...
The developers of L4 are only releasing the specs to the next generation under NDA currently. The NDAs will be dropped, specificaitons and implementations will be available and Free in a couple of months.
Uh, it's more like: 'Yeah, yeah, we know the partition limit is annoying, but we only have so much time and there are more important things to work on.'
The Hurd will be switching to L4, when the NDAs on the latest version of L4 are dropped and there's a real implementation to work from.
That port is fairly extensive, so right now they're moving to Mach + OSKit, since OSKit supports the Linux 2.2 device drivers, something Hurd is sorely lacking.
Hmm. Quantum computers are often granted remarkable properties that actually go beyond what we know is actually possible. While a quantum computer is provably 'faster' than a classical computer, this doesn't mean that you can just take a normal program that would run on a classical computer, compile it with your quantum compiler and observe an exponential improvement in performance.
While it might be possible that a certain algorithm can be cracked exponentially faster with a quantum computer, you still have to find that algorithm. RSA is dependent on factoring large numbers, and is weak against quantum computers due to Shor's algorithm, but that doesn't imply all public key cryptography is.
As far as I'm aware nobody has conclusively determined whether or not quantum computers will be able to break elliptic curve cryptography any faster than classical computers, but there are algorithms (based on coding theory) which have been proven to be just as secure against quantum computers as they are against classical computers.
Apple has the eye candy, but they don't have the basic research into voice, handwriting recognition and other UI topics that others (such as Microsoft) have spent millions on over the years.
Handwriting recognition has been one of the big things holding tablets back. I've played with the TabletPCs a bit, and I think MS may have just got that problem solved, though... still, while they have the details right, they're still missing something in the form department. It should look more like a PocketPC than WinXP, IMHO.
Well, if you sign a contract to kill someone and the other party actually pays you... the contract isn't just null and void, you've both just committed conspiracy to commit murder.:)
If Hammer was already better, then why does the Itanic2 sit in the top spot in SPEC CPU rankings while AMD steadfastly refuses to release performance information?
AMD with its clawhammer is looking to be a better option anyway.
Christ no. Why do we have the people bringing up this same point every time?
x86 is old. It's nasty. It's full of kludges to work around its architectural flaws. And what does AMD want to do? Extend the lifespan another ten to fifteen years! For the sake of all that is sane in computing, don't buy this chip!
Mac users made the switch from their old m68k's to PowerPC, why can't PC users make the switch from x86 to something decent when the time comes? Maybe the Itanic isn't the solution to the ISA woes, but x86-64 definately isn't. In the long run, AMD is doing a disservice to the computing world.
*muttermutter* And of course, all this since a few zealots hate Intel and all it produces for some stupid reason.
Unlike scripting languages, or even C, C++ is hard to port between platforms. Implementations vary widely.
Not really. If you keep in mind portability, it's easy to design a C++ application which will take minimal rewriting, particularly if you make use of libraries like the stl, boost or the mtl. As for implementations... use gcc.:)
With a decent design, moving between Linux, FreeBSD and Solaris is practically trivial. And if you're aware of the pitfalls beforehand even moving from a UNIX to Win32 isn't particularly hard. The whole 'C++ isn't portable! Waah! You have to completely rewrite your application!' is a load of bs.
C++ is a language of great flexability and if used properly it can be anything you want it to. Oh yeah, and unlike Java or PHP you get the benefit of strong static typing. =)
You want RAD? *bam* use a RAD library. You want garbage collection? *bam* use a garbage collection library.
Let's see, NASA, you know the people who send space probes galavanting around the solar system slingshotting around the sun, planets, moons etc... to reach their final destination thinks that building a space station in an L1 point is a good idea. Obviously, you know better than NASA and don't try to figure out why they'd pick L1 over say L4 or L5. *sigh*
While maintaining position at L1 is technically more challenging than maintaining position at L4 or L5, it has a higher payoff. For one, you won't be trying to build your space station in a veritable gravel pit in space. Secondly, it's trivial to launch vehicles from the point - you just let them go and they'll drift off without active station keeping. And considering how the intended primary purpose would be as a place to launch other missions from, that's a slightly useful thing.
The scare did the job. After the widespread ban of CFCs and other ozone depleting chemicals, the ozone hole has begun to stabalise.
Of course, it's still there and as big as ever... it's just not newsworthy anymore.
It's the liberal's dilemma; once the scare they create has forced politicians into making policy changes, things begin to stabalize. Then five years later the conservatives point and hold it up as an example of liberal fallability.
So some guys with a website bought a bunch of large heatsinks from other companies, built a computer that uses almost entirely passive computing and we're expected to do what? Commend them on their innovative use of a credit card?;)
And it's not like super-quiet computers haven't been done before. Yawn. Boring.
Nice troll. I like how you conclude that since the smallest ELF executable is written in assembly, assembly coders are better at writing fast code than optimizing compilers.
Optimizing C compilers don't produce smaller code, and I don't know anybody who claims they do. They're optimizing for speed, size is rarely a significant concern. Writing fast assembly these days is a lot different from what it used to be, and a lot of speed penalties and gains are very non-obvious and requires a highly detailed understanding of the way the processor is implemented. The guys who write optimisers tend to know these, but in my experience most so called assembly programmers don't (although there are notable exceptions).
Your argument is moot- you're discussing material officially classified as Secret, Top Secret, or in the case of Alan Turing's work Ultra Secret.
The problem here is not that the government classifies material for national security, but the 'sensitive, but unclassified' categorization and attempts to browbeat the independent scientific community into not publishing results.
There is no such thing as a perfect mirror in real life, the silvering will always absorb some energy. A sufficiently powerful laser will destroy the silvering relatively quickly, then go about its business as usual.
You do realise there are more ways to get money to your shareholders than dividends, right? In fact, a lot of companies don't pay dividends, since it forces your shareholders to pay taxes on the income.
Now, having stock-splits, then using your profits to buy-back the stock and consequently maintain the stock price at its previous level is such a sweet way of paying profit to your investors. It's a wonder any companies even bother paying dividends at all.
This is, of course, why Nader was so cross about MS not paying dividends - in his view they're screwing the government out of tax revenue, which in a way they are. It's not that they're not paying out profit, it's that they're paying out profit in a very unusual way for tax reasons.
Not only will gigantahumongous hard drives be required to hold this data
Or perhaps the holographic storage that always seems to be just beyond the horizon?;)
Slashdot Mirror
Pfft. It took them long enough to get a modern kernel. Even Microsoft had decent protected memory and preemptive multi-tasking since 1996 (WindnowsNT 3.51). And OS X doesn't have a UNIX kernel. It uses Mach.
So he's not just giving the money out directly, but setting up an income investment fund to ensure that the charity can continue to operate for long, long after his death, bankrupcy, etc...
What have you done for charity lately?
The developers of L4 are only releasing the specs to the next generation under NDA currently. The NDAs will be dropped, specificaitons and implementations will be available and Free in a couple of months.
Uh, it's more like: 'Yeah, yeah, we know the partition limit is annoying, but we only have so much time and there are more important things to work on.'
That port is fairly extensive, so right now they're moving to Mach + OSKit, since OSKit supports the Linux 2.2 device drivers, something Hurd is sorely lacking.
We do have perfect (i.e., impossible to break) encryption. It's called the one time pad, and is a very old idea.
While it might be possible that a certain algorithm can be cracked exponentially faster with a quantum computer, you still have to find that algorithm. RSA is dependent on factoring large numbers, and is weak against quantum computers due to Shor's algorithm, but that doesn't imply all public key cryptography is.
As far as I'm aware nobody has conclusively determined whether or not quantum computers will be able to break elliptic curve cryptography any faster than classical computers, but there are algorithms (based on coding theory) which have been proven to be just as secure against quantum computers as they are against classical computers.
Handwriting recognition has been one of the big things holding tablets back. I've played with the TabletPCs a bit, and I think MS may have just got that problem solved, though... still, while they have the details right, they're still missing something in the form department. It should look more like a PocketPC than WinXP, IMHO.
Well, if you sign a contract to kill someone and the other party actually pays you... the contract isn't just null and void, you've both just committed conspiracy to commit murder. :)
Riiiiiiight.... Move along, nothing to see here.
Christ no. Why do we have the people bringing up this same point every time?
x86 is old. It's nasty. It's full of kludges to work around its architectural flaws. And what does AMD want to do? Extend the lifespan another ten to fifteen years! For the sake of all that is sane in computing, don't buy this chip!
Mac users made the switch from their old m68k's to PowerPC, why can't PC users make the switch from x86 to something decent when the time comes? Maybe the Itanic isn't the solution to the ISA woes, but x86-64 definately isn't. In the long run, AMD is doing a disservice to the computing world.
*muttermutter* And of course, all this since a few zealots hate Intel and all it produces for some stupid reason.
Not really. If you keep in mind portability, it's easy to design a C++ application which will take minimal rewriting, particularly if you make use of libraries like the stl, boost or the mtl. As for implementations... use gcc. :)
With a decent design, moving between Linux, FreeBSD and Solaris is practically trivial. And if you're aware of the pitfalls beforehand even moving from a UNIX to Win32 isn't particularly hard. The whole 'C++ isn't portable! Waah! You have to completely rewrite your application!' is a load of bs.
C++ is a language of great flexability and if used properly it can be anything you want it to. Oh yeah, and unlike Java or PHP you get the benefit of strong static typing. =)
You want RAD? *bam* use a RAD library. You want garbage collection? *bam* use a garbage collection library.
Linux? Christ. Linux's thread support is worse.
Correction: Microsoft is the most profitable company of all time.
While maintaining position at L1 is technically more challenging than maintaining position at L4 or L5, it has a higher payoff. For one, you won't be trying to build your space station in a veritable gravel pit in space. Secondly, it's trivial to launch vehicles from the point - you just let them go and they'll drift off without active station keeping. And considering how the intended primary purpose would be as a place to launch other missions from, that's a slightly useful thing.
And even if you can, they can probably weasel a settlement out of you anyway.
Of course, it's still there and as big as ever... it's just not newsworthy anymore.
It's the liberal's dilemma; once the scare they create has forced politicians into making policy changes, things begin to stabalize. Then five years later the conservatives point and hold it up as an example of liberal fallability.
And it's not like super-quiet computers haven't been done before. Yawn. Boring.
Optimizing C compilers don't produce smaller code, and I don't know anybody who claims they do. They're optimizing for speed, size is rarely a significant concern. Writing fast assembly these days is a lot different from what it used to be, and a lot of speed penalties and gains are very non-obvious and requires a highly detailed understanding of the way the processor is implemented. The guys who write optimisers tend to know these, but in my experience most so called assembly programmers don't (although there are notable exceptions).
Tell that to all the British farmers who were hit by foot and mouth last year... had only they known! Yeah. Right.
The problem here is not that the government classifies material for national security, but the 'sensitive, but unclassified' categorization and attempts to browbeat the independent scientific community into not publishing results.
There is no such thing as a perfect mirror in real life, the silvering will always absorb some energy. A sufficiently powerful laser will destroy the silvering relatively quickly, then go about its business as usual.
Kasparov was just too arrogant to consider Deep Blue a serious opponent.
You do realise there are more ways to get money to your shareholders than dividends, right? In fact, a lot of companies don't pay dividends, since it forces your shareholders to pay taxes on the income.
Now, having stock-splits, then using your profits to buy-back the stock and consequently maintain the stock price at its previous level is such a sweet way of paying profit to your investors. It's a wonder any companies even bother paying dividends at all.
This is, of course, why Nader was so cross about MS not paying dividends - in his view they're screwing the government out of tax revenue, which in a way they are. It's not that they're not paying out profit, it's that they're paying out profit in a very unusual way for tax reasons.
Not only will gigantahumongous hard drives be required to hold this data Or perhaps the holographic storage that always seems to be just beyond the horizon? ;)