When you say "server", are you limited to Xeon chips? Is there no AGP port? Can PCI video make up for that? Are there any 32/66 or 64/66 PCI video cards? What is keeping me from saying "damn the torpordoes, full gear ahead!"?
The reason all those "mass air travel will never be feasible" arguments got traction is that they were very nearly right. It didn't really go anywhere until technology improvements and prosperity and governmental investment made the economics work.
Yes, you can make airlining much more expensive without killing it entirely. The Concorde is proof that anyone wishing the service will pay for it, exhorbitantly, even if it's just an increment better in time savings and no more safe.
But you wouldn't have the A300 or the 767. You'd have the same size fuselage, but with a small cabin (just big enough for current first-class traffic, I'd say) and the rest of the space would be packed with crumple zones, armor, fire retardant, impact inflating flotation devices, parachute, ejection gear, and self-deploying medical supplies. And still they would crash and still people would die.
It might be cheaper to make flying cars and call the death toll "traffic fatalities."
Design the egg protection system to work at terminal velocity, and you can drop it from 40,000 feet.
Airliners won't ever work that way. They only exist because the cost/benefit ratio leaves room for profit. Without that serendipity, the business wouldn't be feasible, and we'd be riding everywhere in trains and boats--and crashing and sinking in even greater numbers.
The qualification of airplane and airplane component designs is tortuous, rendering an airframe that is innately and verifiably safer than any other human product.
What Boeing and Airbus really need to do is find a way to make the press safer. Then airplanes could crash with their usual statistical regularity, and the point-failure tragedy would not appear as though it is more significant than the day's dose of 1200 smoking deaths or the month's dose of 3200 automobile deaths.
Did you read what I wrote? I replaced the word "processor" with "system".
It is AMD's responsibility to ensure that motherboard makers don't destroy its reputation or its marketability. Intel spends huge amounts of money on things like RAM timing stability and motherboard qualification. AMD does not. Tom's used a common motherboard and got disastrous results, but published benchmarks anyway and buried the crashes on the last page, demurral and all. AMD couldn't have been happier.
Despite the fact that it has conquered the compatibility and performance issues, shed intellectual property shackles, and begun performing on manufacturing yield, AMD's products lack reliability in fungible systems. Until it gets its act together and stops playing games with its market and its business, it will not be accepted by knowledgeable professionals as a cost-effective alternative.
Way, way back in the bowels of time, I stood in a comic book shop looking at this brand new Cerebus thing, took the entire stock of issue No. 1 up in my hands, thought for a second, then said "Naah," and put them back.
Yabbut, that's the point. You do 400 experiments, then everyone goes ga-ga over one that has a strangeness quotient of 400?
>If they repeat the experiment and get the same results then you will have a P-value of 160,000 to 1 to explain.
Precisely, exactly, and perfectly the reason I said more labs need to try it. Push that improbability right out of the realm of chance result in the number of experiments it is possible to perform in the course of human history. I mean, even if it wasn't true, if it happens every time you do the experiment, it might as well have been true.
>A trivial example would be putting cobalt-60 (beta emitter) in a solenoid. Electrons fly out one end and the antineutrinos come out the other.
Interesting. But then the e-field has to influence the generation of the electron, not just its momentum after it is generated. If you fire a bullet from a gun, the gun goes the other way; if you suck a bullet out of a gun, the gun would tend to come along. So, does the e-field align the electron and the neutrino before the neutrino is fired from the electron? And if the neutrino isn't charged, how the hell does that happen? If there's a proton involved (and there is, because this is neutron decay, right?) then the proton-electron dipole would be aligned, and you're saying the neutrino comes out by going through the proton, or around it, like a snapped rubber band...
Sorry about all the naff questions. I'm good at physics, but I stopped studying nuclear physics at the early graduate level when they started offering me money to do semiconductor design.
I'm sure there will be plenty who step on what these people are doing, because to disagree you only need a different belief and a means to communicate. But in the end, the disagreement almost always ends up finding the right answer.
Falsehood requires a continuing string of liars or dupes, all repeating their tale the same.
The truth is the same no matter who discovers it.
Truth survives the argument by outlasting the lie, because truth can't die with its tellers. In this way, Science evolves an epistemology of things that we expect will be true no matter who tries it. It is the repeatability that is paramount, not the words that describe it or the people who wrote them or the institutions they represent.
You almost got it, but you're thinking passive and you're thinking it has to do with feature changes.
The reason you see open source is that without open source you don't get a port of some code to your computer.
Being able to customize its features is just a bonus.
Open source is a collective term for all the code that's ever been released by someone who developed it on one machine and was asked to port it to another and said no, if you want it, you do it. Or more often was asked "you know, I'd really like to run your menu-driven pizza builder there but I have an Amygdala21 with a 9-bit Fahnestock busk, can I get your code and see if I can hack some diffs into it?"
Emacs, Gnu, X11, Perl, rogue, etc., etc2., etc2.034.3, etc.
A monolithic market with a single architecture and OS would not at all support open source. It would ensure that closed-source would be a viable business model, since you would know that your code's market was every computer in the world.
But computers were designed to be programmed by humans, and it was only a matter of time before someone more human than Microsoft created an API for an Intel microprocessor that didn't make Bill Gates Even More The Richest Man On Earth Again This Year.
So when Bill said he created open source, he wasn't just being stupid or obfuscatory, he was stating the diametric opposite of the truth.
The RIAA's lawyers could convince the government invoke RICO and declare anyone participating as a server in a piracy network to be a racketeer. That's federal time you'd have to serve just because you think "music should be free".
Only the first PCMCIA modems were $300. The outboard modems were only $99.
Toward the end, they started selling the service for $40 and below. That's probably where the new guys will price it. $75 was fine for the mobile power users, but prevented Ricochet from competing well with DSL and cable, and kept away the base of casual users that would have made them more sound financially without taking up a lot of resources.
Technically, Ricochet was terrific for me once they cleaned up one or two network switching problems.
I was one of the people who wrote Ricochet and told them I'd sign back up if they re-lit the network. I travel to work in multiple cities they serve, so it's a great way to keep my main computer on the same networking paradigm without using a ubiquitous 56k provider (one of which also went tits-up on me in the last year...).
For purely non-mobile applications, though, if you have a choice between this and a cable or LOS wireless service at the same or lower price, you are right to go with that. You will get higher throughput. (I've never liked any DSL offer, but YMMV). Until we see what kind of service organization the new company builds, and what the stability looks like, you can't expect them to be any better or worse for downtime and technical support than any other network provider.
--Blair
"And here, I'm sending Bruce Willis out to drill into the asteroid and blow it up, without enough fuel or equipment to get back safely. It's not actually going to hit us, but he doesn't know that..."
Re:Yeah, there Xbox losses are irrelevant
on
Gamecube Guts
·
· Score: 2
Nobody's going to "monopolize the gaming market" by putting out Yet Another Game Console.
What MS really really wants to do is monopolize the set-top computer market. Think of how they could own us if they could replace our tv tuner, video recorder, game console, movie player, music player, and main home computer with one box, and just in time to get the last-mile bandwidth to finally get video-on-demand working to the home.
--Blair
"If this is in their 5-year plan, can you imagine their 500-year plan?"
Slashdot Mirror
When you say "server", are you limited to Xeon chips? Is there no AGP port? Can PCI video make up for that? Are there any 32/66 or 64/66 PCI video cards? What is keeping me from saying "damn the torpordoes, full gear ahead!"?
--Blair
The reason all those "mass air travel will never be feasible" arguments got traction is that they were very nearly right. It didn't really go anywhere until technology improvements and prosperity and governmental investment made the economics work.
Yes, you can make airlining much more expensive without killing it entirely. The Concorde is proof that anyone wishing the service will pay for it, exhorbitantly, even if it's just an increment better in time savings and no more safe.
But you wouldn't have the A300 or the 767. You'd have the same size fuselage, but with a small cabin (just big enough for current first-class traffic, I'd say) and the rest of the space would be packed with crumple zones, armor, fire retardant, impact inflating flotation devices, parachute, ejection gear, and self-deploying medical supplies. And still they would crash and still people would die.
It might be cheaper to make flying cars and call the death toll "traffic fatalities."
--Blair
Didn't you see GATTACA?
No identification method can't be spoofed.
Including your () fingerless ring.
All we'll get by trying to lock them down is eliminating the liberty of people who aren't the problem.
--Blair
"Other than that, crappy movie."
Design the egg protection system to work at terminal velocity, and you can drop it from 40,000 feet.
Airliners won't ever work that way. They only exist because the cost/benefit ratio leaves room for profit. Without that serendipity, the business wouldn't be feasible, and we'd be riding everywhere in trains and boats--and crashing and sinking in even greater numbers.
The qualification of airplane and airplane component designs is tortuous, rendering an airframe that is innately and verifiably safer than any other human product.
What Boeing and Airbus really need to do is find a way to make the press safer. Then airplanes could crash with their usual statistical regularity, and the point-failure tragedy would not appear as though it is more significant than the day's dose of 1200 smoking deaths or the month's dose of 3200 automobile deaths.
--Blair
Did you read what I wrote? I replaced the word "processor" with "system".
It is AMD's responsibility to ensure that motherboard makers don't destroy its reputation or its marketability. Intel spends huge amounts of money on things like RAM timing stability and motherboard qualification. AMD does not. Tom's used a common motherboard and got disastrous results, but published benchmarks anyway and buried the crashes on the last page, demurral and all. AMD couldn't have been happier.
Despite the fact that it has conquered the compatibility and performance issues, shed intellectual property shackles, and begun performing on manufacturing yield, AMD's products lack reliability in fungible systems. Until it gets its act together and stops playing games with its market and its business, it will not be accepted by knowledgeable professionals as a cost-effective alternative.
--Blair
Flaming Carrot!
I can't believe I forgot to list Flaming Carrot.
--Blair
"I'll be over there, nursing my geekiness."
Other apparent influences/analogues/ripoffs/honest mistakes/coincidences, for various reasons:
National Lampoon's Ver-man and the Flit
Zippy the Pinhead
The Badger
Teenage Mutant Ninja Turtles (before they sold out)
Mystery Men
--Blair
"The Bush White House..."
Way, way back in the bowels of time, I stood in a comic book shop looking at this brand new Cerebus thing, took the entire stock of issue No. 1 up in my hands, thought for a second, then said "Naah," and put them back.
--Blair
"Now I live in a van down by the river."
Man, you are way too anal to be that well equipped.
--Blair
"Check his papers."
Tom's Hardware corroborates the belief that AMD systems, if not AMD chips as well, are unstable.
"while all Athlon [systems] suffered from occasional instability
in our tests, the Pentium 4 platform ran without a glitch."
- Athlon XP Meets P4: A Comparison Of All CPUs
I say they should add troubleshooting and reboot time to all benchmark runs and calculate "performance" that way.
--Blair
"This never happened to me before, honest."
>Not via this specific experiment it hasn't!
Yabbut, that's the point. You do 400 experiments, then everyone goes ga-ga over one that has a strangeness quotient of 400?
>If they repeat the experiment and get the same results then you will have a P-value of 160,000 to 1 to explain.
Precisely, exactly, and perfectly the reason I said more labs need to try it. Push that improbability right out of the realm of chance result in the number of experiments it is possible to perform in the course of human history. I mean, even if it wasn't true, if it happens every time you do the experiment, it might as well have been true.
>A trivial example would be putting cobalt-60 (beta emitter) in a solenoid. Electrons fly out one end and the antineutrinos come out the other.
Interesting. But then the e-field has to influence the generation of the electron, not just its momentum after it is generated. If you fire a bullet from a gun, the gun goes the other way; if you suck a bullet out of a gun, the gun would tend to come along. So, does the e-field align the electron and the neutrino before the neutrino is fired from the electron? And if the neutrino isn't charged, how the hell does that happen? If there's a proton involved (and there is, because this is neutron decay, right?) then the proton-electron dipole would be aligned, and you're saying the neutrino comes out by going through the proton, or around it, like a snapped rubber band...
Sorry about all the naff questions. I'm good at physics, but I stopped studying nuclear physics at the early graduate level when they started offering me money to do semiconductor design.
--Blair
Okay, if you do 400 experiments, you can expect 1 would be in the 1-in-400 bin on the tail of the histogram.
I'm sure the Standard Model has endured way more than 400 tests.
A few more labs need to repeat this experiment to make sure the result is accurate.
--Blair
P.S. If a neutrino is chargeless, how do you "fire" one at something?
I'm sure there will be plenty who step on what these people are doing, because to disagree you only need a different belief and a means to communicate. But in the end, the disagreement almost always ends up finding the right answer.
Falsehood requires a continuing string of liars or dupes, all repeating their tale the same.
The truth is the same no matter who discovers it.
Truth survives the argument by outlasting the lie, because truth can't die with its tellers. In this way, Science evolves an epistemology of things that we expect will be true no matter who tries it. It is the repeatability that is paramount, not the words that describe it or the people who wrote them or the institutions they represent.
--Blair
We had to body cast Patrick Warburton, and build a giant rubber suit, and remote control antennae...
And here I thought they just painted him.
--Blair
You almost got it, but you're thinking passive and you're thinking it has to do with feature changes.
The reason you see open source is that without open source you don't get a port of some code to your computer.
Being able to customize its features is just a bonus.
Open source is a collective term for all the code that's ever been released by someone who developed it on one machine and was asked to port it to another and said no, if you want it, you do it. Or more often was asked "you know, I'd really like to run your menu-driven pizza builder there but I have an Amygdala21 with a 9-bit Fahnestock busk, can I get your code and see if I can hack some diffs into it?"
Emacs, Gnu, X11, Perl, rogue, etc., etc2., etc2.034.3, etc.
A monolithic market with a single architecture and OS would not at all support open source. It would ensure that closed-source would be a viable business model, since you would know that your code's market was every computer in the world.
But computers were designed to be programmed by humans, and it was only a matter of time before someone more human than Microsoft created an API for an Intel microprocessor that didn't make Bill Gates Even More The Richest Man On Earth Again This Year.
So when Bill said he created open source, he wasn't just being stupid or obfuscatory, he was stating the diametric opposite of the truth.
--Blair
My Clark-Nova was talking to me in the '60s, but that was its job.
--Blair
The RIAA's lawyers could convince the government invoke RICO and declare anyone participating as a server in a piracy network to be a racketeer. That's federal time you'd have to serve just because you think "music should be free".
--Blair
Does it sound like Kavita Maharaj?
Because I swear, sexy though it is, her voice is synthesized.
--Blair
Disney has the killer-app movie of the fall kids' market.
Lucas has the killer-pap movie to sell for next summer's kids' market.
Disney needs cash.
Lucas has all the money.
--Blair
"Do the math. Show your work."
Only the first PCMCIA modems were $300. The outboard modems were only $99.
Toward the end, they started selling the service for $40 and below. That's probably where the new guys will price it. $75 was fine for the mobile power users, but prevented Ricochet from competing well with DSL and cable, and kept away the base of casual users that would have made them more sound financially without taking up a lot of resources.
Technically, Ricochet was terrific for me once they cleaned up one or two network switching problems.
I was one of the people who wrote Ricochet and told them I'd sign back up if they re-lit the network. I travel to work in multiple cities they serve, so it's a great way to keep my main computer on the same networking paradigm without using a ubiquitous 56k provider (one of which also went tits-up on me in the last year...).
For purely non-mobile applications, though, if you have a choice between this and a cable or LOS wireless service at the same or lower price, you are right to go with that. You will get higher throughput. (I've never liked any DSL offer, but YMMV). Until we see what kind of service organization the new company builds, and what the stability looks like, you can't expect them to be any better or worse for downtime and technical support than any other network provider.
--Blair
Coming Soon from Infogrames:
CIV IV: The Solar System Strikes Back!
--Blair
"And here, I'm sending Bruce Willis out to drill into the asteroid and blow it up, without enough fuel or equipment to get back safely. It's not actually going to hit us, but he doesn't know that..."
Nobody's going to "monopolize the gaming market" by putting out Yet Another Game Console.
What MS really really wants to do is monopolize the set-top computer market. Think of how they could own us if they could replace our tv tuner, video recorder, game console, movie player, music player, and main home computer with one box, and just in time to get the last-mile bandwidth to finally get video-on-demand working to the home.
--Blair
"If this is in their 5-year plan, can you imagine their 500-year plan?"
Fortunately, it sounds like things will be able to be smoothed over if we miss the deadline
That is the real secret to success in software management.
--Blair
"Choose your parents well."
/. bt, dt:
Interesting Keyboard/Mouse Combo
Can I get the nifty "Desk and Elbow" skin they're using on the demo model?
That looks like it rocks.
--Blair