What a minute, I've looked through all of the 4 mod level posts here, and I see defenses and attacks on apple, but has anyone actually bothered to try this?
It asks for my username, password, again, in either case. When I decline to enter it, it tells me "unauthorized" and sends me on my merry way.
So... what the hell? Of course, what is a little more serious is that this data is all being sent plaintext, but the story as posted doesn't seem to be true, at least based on my casual test.
Also, isn't it considered good form to bother providing a link to the story we're summarizing? I know this is slashdot and no one bothers to read the text anyway, but for those that do, having to copy/paste URLs and browsing the site for the story being discussed is kind of stupid.
Perhaps, but even if you're right, it might suffice if there is a nucleus of high intelligence is preserved even as the rest of humanity goes to rot. In fact, I'd say this is precisely what is happening. While some intelligent people are left to mix with the rest of the population, if they are educated, highly intelligent people do tend to cloister themselves with other intelligent people.
Consider yourself for a moment: how much time do you actually spend mixing with people vastly stupider than yourself in any sort of meaningful interaction? When you first meet a woman, even if she's very pretty, will anything kill your initial animal attraction to her faster than if she opens her mouth and says something stupid? Also, consider the brilliant people you know. Though they are perhaps having fewer children than the remainder of the population, what few children they do have are with other brilliant people. How many theoretical mathematicians do you know married to idiot slobs that spend their days on the couch watching NASCAR or football or whatever shouting "DEFENSE" every 10 minutes or so? It just doesn't happen. Though geographic isolation is breaking down, social and professional isolation is strengthening.
Raskin has been suggesting for years now that the MacOS has failed the interface test. My impression is that he would prefer an entirely different machine that may perhaps be radically different than what we have now. If this is so, Raskin should go out and create his OS of choice.
He did, though it was a long time ago. See information on Raskin's Canon Cat. It would be interesting to see him make a more modern computer interface, but he seems content to just make vague complaints nowadays.
Joe Beigebox wants to render a stunning 3-D scene, but lacks the processing power to do it before the heat death of the Universe. He can buy time on the grid by donating his computer's copious downtime to the Grid, earns 10,000 Whuffie, and renders the picture overnight.
Let's suppose this hypothetical economy exists. So, needing resources, Joe decides to sell his single computer's downtime for whuffie. He then turns around, takes the wuffie he earned, and with it buys power that is many times greater than the power of his computer. So donating 10 gigaflops buys, what, a teraflop? 10 teraflops?
Let's not allow the whuffie word to confuse us. Not even a cute word can introduce energy into a closed system. Whuffie is no different than any other currency. The value of a single whuffie will, of course, be correlated positively with how many resources are available and negatively with how much whuffie people are spending -- clever people could probably take advantage of whatever fluctuations arise -- but the profit will not be enormous. Anyway, whuffie will tend to hover around the same value from day to day. 1 flop donated will tend to buy 1 flop of computing time (and on average probably less).
In this way whuffie is really useful only as storage of the unused computational potential of your computer. So, Joe decides he lacks the processing power to do it before the heat death of the universe, and suddenly decides to sell his computer's power. He is still screwed: he would have to have that power available anyway to buy equal power.
The idea of an economy based partially on buying and selling computational resources is very interesting, but it will not be magical.
Eh? What sort of dumbass admin would be relying on the stock configurations of openssh and sendmail anyway, and moreover relying on Apple to decide when and how they get upgraded or patched?
Nope. The hardware has room inside--physical slots, I mean--for up to 8 GB of RAM.
Hi there. In cropping off the end of my sentence, you rather missed the point. I was not saying that it was capable of addressing only 8 GB, merely that implicit in its ability to address 8 GB is the ability to address more than 32 bits. In other words, we're not actually disagreeing about anything.:D
Instead, the next major release of the Mac operating system will be a hybrid, much like version 10.2.7
So what? This is already known. We know the OS itself will be capable of addressing 8 GB, that is, more than 32 bits. We also already know it will run on 32 bit processors since Panther made its debut on a G4.
I don't mean to be snarky, and I know that post keynote is always slow, but it seems the./ Apple news moderation hasn't been up to task lately. Yesterday we had "news" because someone managed to compile a some open source software using the new QT libraries (and did nothing else, from the looks of it). When I manage to build ImageMagick's shared libs under OS X, I'll be counting on it being on the front page.:P
I never thought of that. Why did Apple put the OS itself in charge of regulating the temperature? Why couldn't the fans be controlled by some simple thermostat like mechanism quite apart from the operating system? Apple has been big on software controlling hardware for a while now, but this seems a bit much.
I am curious about the SPEC benchmarks. SPEC CPU2K does not explicitly use the SSE2 instruction set (or and SIMD instruction set) as far as I know -- does SPEC CPU2K use an outside library like, say, the BLAS, or some other library that would be probably be optimized for SIMD instructions?
I know that GCC 3.3 on x86 has support for SSE2, but it does not automatically vectorize computations. If it is not explicitly used, wouldn't -mfpmath=sse just be a digital placebo? I can feed -faltivec to gcc on my OS X box, but it doesn't follow that I then have an altivec accelerated application!:P
While his explanation helps clear things a bit, on this point the benchmarks still seem a bit unfair, and I don't see how it has been addressed.
Slashdot Mirror
What a minute, I've looked through all of the 4 mod level posts here, and I see defenses and attacks on apple, but has anyone actually bothered to try this?
So... what the hell? Of course, what is a little more serious is that this data is all being sent plaintext, but the story as posted doesn't seem to be true, at least based on my casual test.
Also, isn't it considered good form to bother providing a link to the story we're summarizing? I know this is slashdot and no one bothers to read the text anyway, but for those that do, having to copy/paste URLs and browsing the site for the story being discussed is kind of stupid.
Perhaps, but even if you're right, it might suffice if there is a nucleus of high intelligence is preserved even as the rest of humanity goes to rot. In fact, I'd say this is precisely what is happening. While some intelligent people are left to mix with the rest of the population, if they are educated, highly intelligent people do tend to cloister themselves with other intelligent people.
Consider yourself for a moment: how much time do you actually spend mixing with people vastly stupider than yourself in any sort of meaningful interaction? When you first meet a woman, even if she's very pretty, will anything kill your initial animal attraction to her faster than if she opens her mouth and says something stupid? Also, consider the brilliant people you know. Though they are perhaps having fewer children than the remainder of the population, what few children they do have are with other brilliant people. How many theoretical mathematicians do you know married to idiot slobs that spend their days on the couch watching NASCAR or football or whatever shouting "DEFENSE" every 10 minutes or so? It just doesn't happen. Though geographic isolation is breaking down, social and professional isolation is strengthening.
He did, though it was a long time ago. See information on Raskin's Canon Cat. It would be interesting to see him make a more modern computer interface, but he seems content to just make vague complaints nowadays.
"Here I am, brain the size of a planet and they ask me to filter your spam."
Er, my URLs are wrong, but nonetheless the update is available for Jaguar through software update. :)
The update is available for both Panther (10.3.1 or later) and Jaguar (requires 10.2.8).
Let's suppose this hypothetical economy exists. So, needing resources, Joe decides to sell his single computer's downtime for whuffie. He then turns around, takes the wuffie he earned, and with it buys power that is many times greater than the power of his computer. So donating 10 gigaflops buys, what, a teraflop? 10 teraflops?
Let's not allow the whuffie word to confuse us. Not even a cute word can introduce energy into a closed system. Whuffie is no different than any other currency. The value of a single whuffie will, of course, be correlated positively with how many resources are available and negatively with how much whuffie people are spending -- clever people could probably take advantage of whatever fluctuations arise -- but the profit will not be enormous. Anyway, whuffie will tend to hover around the same value from day to day. 1 flop donated will tend to buy 1 flop of computing time (and on average probably less).
In this way whuffie is really useful only as storage of the unused computational potential of your computer. So, Joe decides he lacks the processing power to do it before the heat death of the universe, and suddenly decides to sell his computer's power. He is still screwed: he would have to have that power available anyway to buy equal power.
The idea of an economy based partially on buying and selling computational resources is very interesting, but it will not be magical.
Eh? What sort of dumbass admin would be relying on the stock configurations of openssh and sendmail anyway, and moreover relying on Apple to decide when and how they get upgraded or patched?
Nope. The hardware has room inside--physical slots, I mean--for up to 8 GB of RAM.
Hi there. In cropping off the end of my sentence, you rather missed the point. I was not saying that it was capable of addressing only 8 GB, merely that implicit in its ability to address 8 GB is the ability to address more than 32 bits. In other words, we're not actually disagreeing about anything. :D
Instead, the next major release of the Mac operating system will be a hybrid, much like version 10.2.7
So what? This is already known. We know the OS itself will be capable of addressing 8 GB, that is, more than 32 bits. We also already know it will run on 32 bit processors since Panther made its debut on a G4.
I don't mean to be snarky, and I know that post keynote is always slow, but it seems the ./ Apple news moderation hasn't been up to task lately. Yesterday we had "news" because someone managed to compile a some open source software using the new QT libraries (and did nothing else, from the looks of it). When I manage to build ImageMagick's shared libs under OS X, I'll be counting on it being on the front page. :P
I'm holding out for Alley Cat. In that version the X will be composed of garbage cans and fish bones.
I never thought of that. Why did Apple put the OS itself in charge of regulating the temperature? Why couldn't the fans be controlled by some simple thermostat like mechanism quite apart from the operating system? Apple has been big on software controlling hardware for a while now, but this seems a bit much.
I am curious about the SPEC benchmarks. SPEC CPU2K does not explicitly use the SSE2 instruction set (or and SIMD instruction set) as far as I know -- does SPEC CPU2K use an outside library like, say, the BLAS, or some other library that would be probably be optimized for SIMD instructions?
I know that GCC 3.3 on x86 has support for SSE2, but it does not automatically vectorize computations. If it is not explicitly used, wouldn't -mfpmath=sse just be a digital placebo? I can feed -faltivec to gcc on my OS X box, but it doesn't follow that I then have an altivec accelerated application! :P
While his explanation helps clear things a bit, on this point the benchmarks still seem a bit unfair, and I don't see how it has been addressed.