The responses to my post mostly don't seem to have an appreciation for the importance and massively revolutionary consequences of AI. It would change everything. And I don't mean change a lot, I mean REALLY change everything. For example poverty would be completely eliminated. DEATH would be eliminated. If it's possible in the near future, then it needs to be done as soon as we can. With the billions available, IBM may be able to make a version of Blue Gene optimized for AI that could actually work.
Uh, exactly HOW is this supposed to work? All we need to eliminate poverty and death is to have something smarter tell us what to do? (Assuming that artificial intelligence would be smarter than the smartest humans, which is Not Necessarily The Case.)
Maybe the problems we have can't be cured by just being smarter, anyway.
But, even if they do make changes to the kernel, I suspect they have some way of getting around the license.
They wouldn't have to get around the license. If they don't distribute the binaries outside the organization, they don't have to release source outside the organization.
I seriously doubt they'll be selling those to anybody off the street.
No need, most employees are already using Windows.
Re:You don't know anything about railroads, do you
on
By Road and Rail?
·
· Score: 4, Informative
An excellent and accurate response, except for one thing...
you have a dual purpose buss rolling along a rail route at, ummm, what, 80kmph? It weighs, what? 10 tons? 20 tons? Then right behind it is a kilometer long train full of, oh, I dunno - NAPTHA - that's roaring along at what?140kmph? ANd it weighs how many hundreds of tons? And takes how long to stop?
A good rule of thumb for stopping distance is roughly 1 meter per kph in daytime, about 1 1/4 at night; I've heard that at 80kph (which is exactly 50mph for those of us in the States) the distance is about 81 meters (about 245 feet) and at night it's about 95m (about 300 feet).
The stopping distances quoted above are for automobiles on dry pavement. Trains take quote a bit more distance:
150-car freight train stopping distance
30 mph =3,500 feet or 2/3 of a mile
50 mph =8,000 feet or 1 1/2 miles
8-car passenger train stopping distance
60 mph =3,500 feet or 2/3 of a mile
79 mph =6,000 feet or 1 1/8 miles
(Data from various Operation Lifesaver websites...)
Got any proof of that? Proof either way is hard to come by -- being a commercial closed-source program, the makers of Fritz don't publish anything, and most of the internals of Deep Blue (such as it's evaluation terms) are still covered by NDAs with IBM.
Of course, "Deep Fritz" didn't exist in 1995. It was plain old single-processor Fritz. Deep Blue wasn't around then either -- a version of Deep Thought was enetered, an early predecessor of Deep Blue and Deep Blue II.
While it's likely that there was some wasted nodes in those 200 million nps (some loss of efficiency is expected in multi-processor game tree search algorithms), given the fact that Deep Blue II beat Gary Kasparov, it's unlikely that many "wrong" moves were encountered.
The machine that Fritz beat wasn't the Deep Blue that Kasparov played -- it was a version or two before that. Far fewer processors, for one thing, and earlier versions of the special chess processors.
IBM Corporation has absolutely nothing to gain. If they enter some version of Deep Blue II and win, it's because they were expected to. If they draw even a single game, several years (and millions of dollars) of marketing and public hype goes out the window.
Summarizing -- A group of Carnegie-Mellon doctoral students (Feng-hsiung Hsu, Murray Campbell and Thomas Anantharaman) (mostly hardware guys!) built Chiptest in 1985. Special-purpose hardware with a chip design originally loosely based on Hans Berliner's Hitech. Evolving development leads to "Deep Thought" (someone was a Hitchhiker fan) in 1988, Hsu and Campbell joining IBM in 1989, Deep Thought II in '91, Deep Blue in '93, and Deep Blue II in '97.
Deep Blue was not a very aesthetic name, but worth millions and millions to IBM in publicity, you can be sure.
However, in the real world, you decide what you want to do : write neat new software that requires new hardware to run efficiently, or make neat new hardware that requires the software to be completely rewritten.
Slashdot Mirror
They wouldn't have to get around the license. If they don't distribute the binaries outside the organization, they don't have to release source outside the organization.
I seriously doubt they'll be selling those to anybody off the street.
No, the thing that makes MS bad is what makes Apple successful.
No need, most employees are already using Windows.
The stopping distances quoted above are for automobiles on dry pavement. Trains take quote a bit more distance
150-car freight train stopping distance
30 mph =3,500 feet or 2/3 of a mile
50 mph =8,000 feet or 1 1/2 miles
8-car passenger train stopping distance
60 mph =3,500 feet or 2/3 of a mile
79 mph =6,000 feet or 1 1/8 miles
(Data from various Operation Lifesaver websites...)
Since 1992 -- http://www.chessclub.com/about.html#history
Online ratings and the works.
From what I can see, this appears to be the perfect prior art.
Got any proof of that? Proof either way is hard to come by -- being a commercial closed-source program, the makers of Fritz don't publish anything, and most of the internals of Deep Blue (such as it's evaluation terms) are still covered by NDAs with IBM.
Of course, "Deep Fritz" didn't exist in 1995. It was plain old single-processor Fritz. Deep Blue wasn't around then either -- a version of Deep Thought was enetered, an early predecessor of Deep Blue and Deep Blue II.
While it's likely that there was some wasted nodes in those 200 million nps (some loss of efficiency is expected in multi-processor game tree search algorithms), given the fact that Deep Blue II beat Gary Kasparov, it's unlikely that many "wrong" moves were encountered.
The machine that Fritz beat wasn't the Deep Blue that Kasparov played -- it was a version or two before that. Far fewer processors, for one thing, and earlier versions of the special chess processors.
IBM Corporation has absolutely nothing to gain. If they enter some version of Deep Blue II and win, it's because they were expected to. If they draw even a single game, several years (and millions of dollars) of marketing and public hype goes out the window.
Uh, no.
This page from IBM has a timeline of the development of Deep Blue.
Summarizing -- A group of Carnegie-Mellon doctoral students (Feng-hsiung Hsu, Murray Campbell and Thomas Anantharaman) (mostly hardware guys!) built Chiptest in 1985. Special-purpose hardware with a chip design originally loosely based on Hans Berliner's Hitech. Evolving development leads to "Deep Thought" (someone was a Hitchhiker fan) in 1988, Hsu and Campbell joining IBM in 1989, Deep Thought II in '91, Deep Blue in '93, and Deep Blue II in '97.
Deep Blue was not a very aesthetic name, but worth millions and millions to IBM in publicity, you can be sure.
However, in the real world, you decide what you want to do : write neat new software that requires new hardware to run efficiently, or make neat new hardware that requires the software to be completely rewritten.