Timna, for the record, is the name of an ancient (Egyptian, I think) copper mine (get it yet?) in what is currently Israel. I actually visited the place last year. A little surreal.
Anyway the Israel references come because Intel has a big fab and research center over there.
I don't see anything in the article about a "compact-car sized meteor" as the submitter suggests. It seems that once again Slashdot has posted unconfirmed, unreliable information. Meteors strike the earth all the time (mostly falling in the ocean unobserved). This is nothing to report.
Scientific American ran an article two years ago about different search and indexing strategies for the web. You can read the online rendition
here. It discusses many of the techniques used in Google, and some techniques that (as far as I can tell) have not yet been implemented in real-world search engines. Some cool ideas here, such as classifying large link-providers and link-receivers as "information hubs." If possible, I suggest getting the hard-copy of this article since it includes a few nice graphics and is a much easier read than the online version.
Anyway, none of the ideas used by Google and now Teoma are really new -- academics have been doing this stuff for a while now.
Don't get too worked up about this. As soon as someone, somewhere, manages to get positive power output from a fusion reactor, they will die in a "mysterious accident." Their notes will be burned in a "inexplicable fire." The buildings housing computers containing pertinent information will be involved in "an unfortunate, unexplained explosion." The petroleum companies have too much invested in world pollution to just sit around and let this happen.
The best way to prevent this from happening is complete disclosure of all knowledge pertaining to fusion, and large-scale mirroring of that information all over the world.
My company directly competes with Adobe in certain arenas, so I'm a little biased against them, but this is a very liberal move on Adobe's part. One that I actually didn't expect them to make when I first heard of this.
The question is, what are Adobe's real motivations for this settlement? By settling they reserve their right to bring a suit in the future, while still getting KIllustrator to change its name.
It's a bit premature to judge, but either Adobe is beginning to fear free software, or they are beginning to embrace it. I'll be watching this closely now.
When drunk drivers who put *lives* at risk don't get that sort of time, much less financial penalty, (especially on a first offense!), this becomes an abuse of the law and of law enforcement.
And this surprises you? Have you forgotten that this country's government caters to corporations and not individuals? What does it matter if one individual dies, when there are crimes being committed against corporations? The law isn't being abused; it has been purposefully constructed to protect (to a zealous degree) the "rights" of corporate entities at the expense of individuals and human dignities.
You can do a few things: vote in the next election, move to another country, or turn to a life of crime trying to outwit the f***ers. But don't waste your breath here on Slashdot.
The argument is intended to set a theoretical speed limit, not to describe a method of extracting energy from the mass in a laptop. The gist of this is that, provided a certain amount of energy, it is possible to compute only so fast. The argument is ok, even if he is abusing physics a little bit.
The real problem comes on the next page where he starts talking about entropy. Claude Shannon (considered to be the father of information theory) made a grave mistake in calling his measure of information density "entropy." He should have just called it "information density." You can learn about this definition of entropy in any good data compression or information theory textbook. It is unfortunate because many people seem to think that physical entropy and informational entropy are somehow equivalent, when in fact they are not.
There may, in truth, be a relationship between physical entropy and informational entropy but the relationship has not yet been discovered, as this article would lead us to believe. If it ever is discovered, the form of the relationship will certainly not be as simple as is stated here.
This is all very dubious and actually somewhat amusing.
Some people seem confused. This license does not prevent developers from using free software to develop products using this SDK. It prevents developers from using Potentially Viral Software to develop such products, and it prevents developers from redistributing the SDK in conjunction with such Potentially Viral Software.
At the risk of repetition, here is Microsoft's definition of Potentially Viral Software:
"Potentially Viral Software" means software which is licensed pursuant to terms that: (x) create, or
purport to create, obligations for Microsoft with respect to the Software or (y) grant, or purport to grant, to any third party any
rights to or immunities under Microsoft's intellectual property or proprietary rights in the Software.
Ok, this makes sense. Microsoft is preventing people from redistributing the SDK in conjunction with software that is licensed under terms that could be predatory toward Microsoft's products. Microsoft is also stipulating that such "viral" software may not be used in the process of developing products that make use of their SDK. Great, good for Microsoft. They are covering their ass.
Microsoft, however, further states that Publically Available Software falls under this PVS category. Untrue. By using Emacs, for example, to create a source module that calls Microsoft's SDK, I do not infect that source module with a free license. Emacs is not distributed with licensing terms to the effect of "any source code produced with Emacs thereafter shall be Free Software." I can create anything I damn well please in Emacs.
What is really going on here? Is Microsoft really that afraid of PAS? Nope. They are hoping to scare developers away from open source by misrepresenting the contents of open source licenses. I'd bet a lot of proprietary developers have never even read the GPL, and they'll believe whatever Microsoft feeds them regarding open source.
So this SDK license is really serving two purposes for Microsoft: cover Microsoft's ass in case such "viral" software actually does exist somewhere (which I highly doubt), and also spread Fear, Uncertainty, and Doubt amongst the diehard Microsoft developers regarding open source licensing conditions. It is a calculated misrepresentation of open source, and it's damned clever. By hiding this crap in a license agreement, Microsoft makes it appear that they are afraid of open source -- and therefore, other developers should also be afraid of it.
Since when does the government develop ANYTHING? The government did not develop the space shuttle, the government did not develop the communications satellite, or GPS, or the M16 machine gun, or any number of other things the government has interests in. The government allocates funds for private industry to do the development. This is what capitalism is all about.
Not to mention that IBM, as a technology entity, has interests in pure research, just like most other huge companies. Despite what you might think, large companies often have humanitarian and scientific interests that are not bound by capitalistic fantasies.
The example cited in the article about faster web-searching might become reality, then again, it might not. It's still only a single application, and if the time-to-fruition really is along the lines of 20 years as the quantum computing community believes, we'd be idiots to assume that the web will look anything like it does today, or even exist at all, by then. Speculating on how to apply developing technologies to the internet in 20 years is like a person in the 1940's speculating how to apply new propellor designs to new planes in the 1960's, which ended up using air-breathing jet engines instead of inefficient propellors. Except that the internet is changing a hell of a lot faster than that. It's pointless to waste time speculating about the internet in 20 years.
It is because a quantum bit ("qubit") can be in multiple states simultaneously. This is almost synonymous with simulating "multiple problem variants" simultaneously.
Here's a good starting point for the non-physicist:
No. Have you ever even read the literature on quantum computation? Sure, it's small, it uses very little power, and it's incredibly powerful.
But how do you propose we program a quantum device to do everyday tasks? Do think it's as simple as writing some C code? No way in hell. Go read the literature.
I thought the Slashdot readership (regardless of intelligence level) was generally opposed to wanton commercialization. Now we're talking about commercializing QUANTUM COMPUTERS? Get off it already.
Why the hell would any normal person ever need something like this? "Ooooh daddy, can I get the QZ-5900 please?!?! I want to calculate solar radiation flux! I want to simulate nuclear detonations! I want to solve the traveling salesman problem for 29 billion routes!"
These are the kinds of problems that quantum computers are geared to compute. Not for playing games, not for doing spreadsheets, and definitely not anything for the store shelves. The only thing that might even tangentially intersect with the common interests is decryption, and I assure you the government will take care of preventing that kind of technology from ever getting into our hands. If we ever see quantum computers in commercial hardware, it will be in very specialized devices that do tasks that 95% of civilization doesn't care about.
If I have a switch that is broken and sending packets into the void, this is what I do:
I unplug it.
The FBI's black box is no different than a malfunctioning switch. The ISP tech walks over to the rack, unplugs their precious box, and look, we have connectivity again. You need an IQ of about 5 to be able to figure this out.
The "rotating" DNA helix of M&M's made total sense to me, I don't see why anyone else missed it. It wasn't that the HELIX was rotating in space, it was staying still and the rotating LIVING QUARTERS was rotating around it. The camera tracked the living quarters, giving the helix the appearance of rotating. You could really do this (very carefully) if you were on a rotating ring in space. Although I admit this was about the best it got in the movie, don't moan about the "impossible" effect; it's possible.
My other major beef with this movie was, why are the humans controlling everything? On a mission to Mars I really doubt there will be humans sitting at consoles punching buttons with the split-second timing needed to perform course corrections. I would think the computer would be doing all that while the crew sat back and drank whiskey. Let's see, what else? The gravity rotation didn't shut down when they entered Mars orbit, this is bad. Imagine that you are flying a giant gyroscope. You try to turn. It doesn't work.
If the aliens all escaped in their cute little spaceships, as the hologram would lead us to believe, why didn't they just come to earth? They seeded earth with new life, but where did THEY go? And you are telling me that this futuristic spacecraft was incapable of alerting the crew to the ENORMOUS FREAKING HOLES in the fuel line? Or how about the guy's face freezing immediately after removing his helmet? There's nothing in space to conduct the heat away from your body -- you can't describe it as "cold" because there's nothing there to be cold. Heat radiates away as infrared in space, and this doesn't occur fast enough to cause the effect depicted in the movie.
Once again Slashdot ruins a site. I wonder how many bottles of Advil will be consumed at linuxplanet.com today. Why the hell can't you guys mirror it? And if you can't mirror it (for copyright reasons) why don't you at least inform the unsuspecting, hard-working site admins that you are about to DoS them?
Yes, Slashdot is a great big DoS engine. From an objective standpoint, what Slashdot does to sites is no different than little script kiddies with packet machine guns. Except it's legal.
"Preferred" is so subjective that this GPL clause really doesn't mean anything at all. What if the author prefers to name all his variables x1, x2, x3, x4, x5, GlowInTheDarkCondom, etc? Are you seriously going to take someone to court and say "This man should be fined or thrown in jail because he uses bad identifiers?"
Suppose that every time a[b] occurs in the code I replace it with b[a] (which means the same but is a little more opaque). Are you suggesting that by doing this I might be heading for legal trouble because other people might have difficulty understanding it? Even if it's MY OWN SOURCE? Do you understand the source to GCC? Are you going to sue people because it is not commented better?
Admittedly there is a difference between dark code and obfuscated code, but if a software author wants to write scribbles and release it under GPL then I really doubt anyone is going to complain. "Preferred form" is just a way to ensure that the code is released in its original format as opposed to, say, the assembly language output of the compiler, which isn't quite so useful for making modifications.
Enforcing this GPL clause is exactly the same as enforcing coding standards. If you don't buy that, read over the language again and think about exactly what "preferred form for making modifications" means. Preferred by who? What sort of modifications?
Here's a nice example from the Linux kernel (mm/page_alloc.c, on or around line 76):
/* * Buddy system. Hairy. You really aren't expected to understand this * * Hint: -mask = 1+~mask */
You aren't expected to understand it. Why don't you go after Linus for writing unreadable code? An implementation of a buddy allocator doesn't HAVE to be unreadable; Linus could have made this code much clearer and understandable but he chose not to (probably in the interest of a straight-line execution path in the common case). I would not "prefer" this code if I was the one modifying it. In fact, I wouldn't "prefer" the majority of the kernel nor would I prefer much of the vast body of other GPL'd source out there. So, for me, this code is not in the "preferred form." So do I have the right to sue Linus for every cent he has? OF COURSE I DON'T.
The opinions expressed herein are solely my own and do not reflect the opinions or policies of my employer.
As I sit here trying to follow the link unsuccessfully, the following came to mind: the web site serving these pictures is not equipped to deal with the sort of traffic this seemingly innocent Slashdot article has generated, and, more than likely, the poor scientists there are getting absolutely nothing useful accomplished. The web server is probably using the same uplink as the rest of the machines at the site, and if I can't even get through to the web server, then I really doubt the scientists are able to get OUT to the internet. I wish the Slashdot community (and those people who select which articles to post) would think about this. These people are trying to conduct SCIENCE and they are providing these (very LARGE) images as a free service out of kindness. Our curiosity is no justification for crippling their operation for hours.
I'll look at the pictures next week, after the mob has left. I'll be considerate.
Slashdot Mirror
Anyway the Israel references come because Intel has a big fab and research center over there.
Back in the beginning people did think Pentium was a stupid name. Now, nothing weird about it. You just get used to it.
Anyway don't make fun of it. Plenty of people live in Tualatin, drive on the Tualatin Valley highway, etc...
I don't see anything in the article about a "compact-car sized meteor" as the submitter suggests. It seems that once again Slashdot has posted unconfirmed, unreliable information. Meteors strike the earth all the time (mostly falling in the ocean unobserved). This is nothing to report.
Anyway, none of the ideas used by Google and now Teoma are really new -- academics have been doing this stuff for a while now.
The best way to prevent this from happening is complete disclosure of all knowledge pertaining to fusion, and large-scale mirroring of that information all over the world.
The question is, what are Adobe's real motivations for this settlement? By settling they reserve their right to bring a suit in the future, while still getting KIllustrator to change its name.
It's a bit premature to judge, but either Adobe is beginning to fear free software, or they are beginning to embrace it. I'll be watching this closely now.
And this surprises you? Have you forgotten that this country's government caters to corporations and not individuals? What does it matter if one individual dies, when there are crimes being committed against corporations? The law isn't being abused; it has been purposefully constructed to protect (to a zealous degree) the "rights" of corporate entities at the expense of individuals and human dignities.
You can do a few things: vote in the next election, move to another country, or turn to a life of crime trying to outwit the f***ers. But don't waste your breath here on Slashdot.
The real problem comes on the next page where he starts talking about entropy. Claude Shannon (considered to be the father of information theory) made a grave mistake in calling his measure of information density "entropy." He should have just called it "information density." You can learn about this definition of entropy in any good data compression or information theory textbook. It is unfortunate because many people seem to think that physical entropy and informational entropy are somehow equivalent, when in fact they are not.
There may, in truth, be a relationship between physical entropy and informational entropy but the relationship has not yet been discovered, as this article would lead us to believe. If it ever is discovered, the form of the relationship will certainly not be as simple as is stated here.
This is all very dubious and actually somewhat amusing.
At the risk of repetition, here is Microsoft's definition of Potentially Viral Software:
"Potentially Viral Software" means software which is licensed pursuant to terms that:
(x) create, or purport to create, obligations for Microsoft with respect to the Software or
(y) grant, or purport to grant, to any third party any rights to or immunities under Microsoft's intellectual property or proprietary rights in the Software.
Ok, this makes sense. Microsoft is preventing people from redistributing the SDK in conjunction with software that is licensed under terms that could be predatory toward Microsoft's products. Microsoft is also stipulating that such "viral" software may not be used in the process of developing products that make use of their SDK. Great, good for Microsoft. They are covering their ass.
Microsoft, however, further states that Publically Available Software falls under this PVS category. Untrue. By using Emacs, for example, to create a source module that calls Microsoft's SDK, I do not infect that source module with a free license. Emacs is not distributed with licensing terms to the effect of "any source code produced with Emacs thereafter shall be Free Software." I can create anything I damn well please in Emacs.
What is really going on here? Is Microsoft really that afraid of PAS? Nope. They are hoping to scare developers away from open source by misrepresenting the contents of open source licenses. I'd bet a lot of proprietary developers have never even read the GPL, and they'll believe whatever Microsoft feeds them regarding open source.
So this SDK license is really serving two purposes for Microsoft: cover Microsoft's ass in case such "viral" software actually does exist somewhere (which I highly doubt), and also spread Fear, Uncertainty, and Doubt amongst the diehard Microsoft developers regarding open source licensing conditions. It is a calculated misrepresentation of open source, and it's damned clever. By hiding this crap in a license agreement, Microsoft makes it appear that they are afraid of open source -- and therefore, other developers should also be afraid of it.
Very damned clever. I'm impressed.
Not to mention that IBM, as a technology entity, has interests in pure research, just like most other huge companies. Despite what you might think, large companies often have humanitarian and scientific interests that are not bound by capitalistic fantasies.
The example cited in the article about faster web-searching might become reality, then again, it might not. It's still only a single application, and if the time-to-fruition really is along the lines of 20 years as the quantum computing community believes, we'd be idiots to assume that the web will look anything like it does today, or even exist at all, by then. Speculating on how to apply developing technologies to the internet in 20 years is like a person in the 1940's speculating how to apply new propellor designs to new planes in the 1960's, which ended up using air-breathing jet engines instead of inefficient propellors. Except that the internet is changing a hell of a lot faster than that. It's pointless to waste time speculating about the internet in 20 years.
Here's a good starting point for the non-physicist:
Quantum Computation: A Tutorial
But how do you propose we program a quantum device to do everyday tasks? Do think it's as simple as writing some C code? No way in hell. Go read the literature.
Why the hell would any normal person ever need something like this? "Ooooh daddy, can I get the QZ-5900 please?!?! I want to calculate solar radiation flux! I want to simulate nuclear detonations! I want to solve the traveling salesman problem for 29 billion routes!"
These are the kinds of problems that quantum computers are geared to compute. Not for playing games, not for doing spreadsheets, and definitely not anything for the store shelves. The only thing that might even tangentially intersect with the common interests is decryption, and I assure you the government will take care of preventing that kind of technology from ever getting into our hands. If we ever see quantum computers in commercial hardware, it will be in very specialized devices that do tasks that 95% of civilization doesn't care about.
If I have a switch that is broken and sending packets into the void, this is what I do:
I unplug it.
The FBI's black box is no different than a malfunctioning switch. The ISP tech walks over to the rack, unplugs their precious box, and look, we have connectivity again. You need an IQ of about 5 to be able to figure this out.
The "rotating" DNA helix of M&M's made total sense to me, I don't see why anyone else missed it. It wasn't that the HELIX was rotating in space, it was staying still and the rotating LIVING QUARTERS was rotating around it. The camera tracked the living quarters, giving the helix the appearance of rotating. You could really do this (very carefully) if you were on a rotating ring in space. Although I admit this was about the best it got in the movie, don't moan about the "impossible" effect; it's possible.
My other major beef with this movie was, why are the humans controlling everything? On a mission to Mars I really doubt there will be humans sitting at consoles punching buttons with the split-second timing needed to perform course corrections. I would think the computer would be doing all that while the crew sat back and drank whiskey. Let's see, what else? The gravity rotation didn't shut down when they entered Mars orbit, this is bad. Imagine that you are flying a giant gyroscope. You try to turn. It doesn't work.
If the aliens all escaped in their cute little spaceships, as the hologram would lead us to believe, why didn't they just come to earth? They seeded earth with new life, but where did THEY go? And you are telling me that this futuristic spacecraft was incapable of alerting the crew to the ENORMOUS FREAKING HOLES in the fuel line? Or how about the guy's face freezing immediately after removing his helmet? There's nothing in space to conduct the heat away from your body -- you can't describe it as "cold" because there's nothing there to be cold. Heat radiates away as infrared in space, and this doesn't occur fast enough to cause the effect depicted in the movie.
Is it worth it to continue? I think not.
Once again Slashdot ruins a site. I wonder how many bottles of Advil will be consumed at linuxplanet.com today. Why the hell can't you guys mirror it? And if you can't mirror it (for copyright reasons) why don't you at least inform the unsuspecting, hard-working site admins that you are about to DoS them?
Yes, Slashdot is a great big DoS engine. From an objective standpoint, what Slashdot does to sites is no different than little script kiddies with packet machine guns. Except it's legal.
---
My opinions are my own and no one else's.
"Preferred" is so subjective that this GPL clause really doesn't mean anything at all. What if the author prefers to name all his variables x1, x2, x3, x4, x5, GlowInTheDarkCondom, etc? Are you seriously going to take someone to court and say "This man should be fined or thrown in jail because he uses bad identifiers?"
Suppose that every time a[b] occurs in the code I replace it with b[a] (which means the same but is a little more opaque). Are you suggesting that by doing this I might be heading for legal trouble because other people might have difficulty understanding it? Even if it's MY OWN SOURCE? Do you understand the source to GCC? Are you going to sue people because it is not commented better?
Admittedly there is a difference between dark code and obfuscated code, but if a software author wants to write scribbles and release it under GPL then I really doubt anyone is going to complain. "Preferred form" is just a way to ensure that the code is released in its original format as opposed to, say, the assembly language output of the compiler, which isn't quite so useful for making modifications.
Enforcing this GPL clause is exactly the same as enforcing coding standards. If you don't buy that, read over the language again and think about exactly what "preferred form for making modifications" means. Preferred by who? What sort of modifications?
Here's a nice example from the Linux kernel (mm/page_alloc.c, on or around line 76):
/*
* Buddy system. Hairy. You really aren't expected to understand this
*
* Hint: -mask = 1+~mask
*/
You aren't expected to understand it. Why don't you go after Linus for writing unreadable code? An implementation of a buddy allocator doesn't HAVE to be unreadable; Linus could have made this code much clearer and understandable but he chose not to (probably in the interest of a straight-line execution path in the common case). I would not "prefer" this code if I was the one modifying it. In fact, I wouldn't "prefer" the majority of the kernel nor would I prefer much of the vast body of other GPL'd source out there. So, for me, this code is not in the "preferred form." So do I have the right to sue Linus for every cent he has? OF COURSE I DON'T.
The opinions expressed herein are solely my own and do not reflect the opinions or policies of my employer.
As I sit here trying to follow the link unsuccessfully, the following came to mind: the web site serving these pictures is not equipped to deal with the sort of traffic this seemingly innocent Slashdot article has generated, and, more than likely, the poor scientists there are getting absolutely nothing useful accomplished. The web server is probably using the same uplink as the rest of the machines at the site, and if I can't even get through to the web server, then I really doubt the scientists are able to get OUT to the internet. I wish the Slashdot community (and those people who select which articles to post) would think about this. These people are trying to conduct SCIENCE and they are providing these (very LARGE) images as a free service out of kindness. Our curiosity is no justification for crippling their operation for hours.
I'll look at the pictures next week, after the mob has left. I'll be considerate.