As to API calls -- that's where various layers of abstraction come in useful. If properly designed, a program should not be hard to port to any platform that offers the necessary features.
As a person who works almost entirely on maintaining the UNIX port of a Windows/UNIX application where 90% of its developers only work on Windows, I can tell you that first-hand that this isn't as easy as you say. First, you have to design the software from ground-up to be portable. That means making sure that you never call system-specific APIs where possible and always step through the abstraction layer. If your application wasn't designed to be ported at the very beginning, then going back through the codebase and retooling all calls to native system calls is extremely time-consuming and leaves behind terribly unmaintainable code. Second, even if you did do a nice abstraction layer, platform-specific demands for features will eventually render your abstraction layer harder and harder to maintain. It isn't that it's impossible to have cleanly implemented abstraction layer that doesn't build up horrible kluges over time -- it's just extremely time-consuming and thus considered economically unfeasable in the real world.
Because of the slow decay of portability layers over time, it's actually far easier to just write single-platform code. Backend code tends to be easy to write portably, but front-end code is always nightmarish when going between GUI and graphics libraries. OpenGL vs. DirectX, Motif vs. Win32, COM vs. CORBA -- the small incompatibilities between the way these libraries work always results in some horrible glue code.
I am sure Microsoft knows what the hell they are doing, given that they have been in the business for quite a while.
Me too. This is why the Mac Business Unit has a completely separate codebase for IE for Mac. This is one reason why IE for Mac was far more standards-compliant and far more bug-free than IE for Windows. They didn't share ANY code.
Ever notice that Mac Office and Windows Office look nothing like each other? I don't think it's because of porting difficulties [...]
The fact that the two products don't resemble each other at all should be the first warning that they have a lot of code that is not shared between the two platforms. MS tried straight-up porting the suite back in the 90s, and it was disasterously received on the Mac for 3 reasons -- it didn't look like a Mac app at (including the previous versions of Word, Excel, etc.), it was slow and a memory hog, and it was buggy. All of these were because of the disasterous backporting effort to take the Windows codebase and use it on the Mac. They essentially wrote a huge library that contained large chunks of the Windows APIs ported to run on top of the MacOS APIs. (This is ironic since all of these applications started on the Mac and were ported to Windows from there before the differences between the two were considered to be "unmaintainable.") Over the years since then, the Mac version of Office has eliminated more and more ported Windows code and gone back to using more and more MacOS-only code. There were serious porting difficulties with their software. That's why they prefer to just treat the two as separate codebases whenever possible.
Cheap is a relative term. It takes nearly half a day to deposit a millimeter's width. Also despite being one of the hardest substances on Earth, diamond is very brittle and would not hold up well to being dropped while in the thin, yet wide shape of a CD.
Except that you've just eliminated the need for musicianship in the vocalist. Now, basically, songwriting and backup bands are all that's important. (Hey.... Maybe this isn't such a bad thing after all. Freakin' prima-donna front men...)
The technology is already in use to create pop acts and has been used throughout the late 90s for this very purpose. The very pop acts that the other poster listed are known to use autotuners. Instead of finding good singers, all the recording industry has to do is find pretty people who can sing passably well enough to work with the autotuner. It's essentially the final nail in the coffin of actual talent in pop acts.
This "levels the playing field" in an industry supposedly based on bringing the best of the best to the national stage. Standards in the recording industry have been slipping for years. One reason for this is that truly talented and popular acts are hard to keep control of contractually. They have enough creativity and talent to jump ship to other labels or create their own labels and survive for years. This is pretty much the opposite of what the recording industry wants. This technology will allow them to take more talentless nobodies and propel them to the national stage with the implicit understanding that without the company that sponsored them, they are nothing. More talented yet more difficult to control musicians can be left more and more to the wayside. In essence, it allows a talent-based industry to get away with selling products without true talent, thus cutting long-term expenses for them.
This technology's a gift from the heavens in karaoke bars, but it'll be just another step in purging all vitality and talent from popular music.
Not necessarily, no. Finding them inserting data into the network would be like finding a needle in a haystack. If they do it properly, they could insert data into the network at any point and have it propogate to the other systems to mail out. Look at how Gnutella propogates searches sometime. Make it push-based and remove the need for an identifying address for the origin to be returned, and you've got a distribution scheme ideal for spamming. You just feed new emails and new addresses into the network and watch it go.
Nearly everyday traffic on the 6 lane interstate suddenly comes to a full stop. When you get going again and drive up a little ways, there was no reason to stop, no accident, no debris in the road.
Along with most people here, I've seen pretty good anecdotal evidence to believe that aches and pains can predict the weather. Ultimately, the only way you're going to find out is if you track changes in barometric pressure each day and log all aches and pains for a while.
However, while you're doing so, I'd recommend that you also log your caffeine intake. Over a decade of nearly migrane-level headaches has become just a part of my past since I discovered that caffeine withdrawl was causing my problem and simply swore off of caffeine permanently. Most people who have aches and pains before the weather have them in their joints. I've never heard of headaches as a symptom before.
Check your caffeine intake levels as an indicator.
I also pointed out the possibility Microsoft is using SCO as a proxy in an effort to exterminate one of the few remaining threats to their monopoly.
That wasn't a good idea. Now you've entered conspiracy-theory land, where the authorities simply stop listening to you. There is absolutely no proof whatsoever that MS is behind this, and you just sound like another drooling paranoid anti-MS zealot when you go around repeating stuff like this. I violently dislike MS too, but give credit where credit's due. SCO is probably doing all of this on their own for the profit of their own executives.
All MS did was license some UNIX stuff early on, right after the case got started. That's circumstantial evidence at best -- correlation, not causation. You're not going to win any friends among prosecutors if you make unfounded accusations like that.
Well... maybe. The cheapest man won. I'm not disparaging GCC as a technical project, but the reason that all those other developers went out of business is because one of their competitors cost nothing instead of cositng hundreds or thousands. This is essentially the reason why Internet Explorer defeated Netscape 4 way back in the day when Netscape costed money. This is why Doublespace destroyed Stacker, why Windows Media Player is killing Real, and why you don't hear anything about webservers other than Apache and IIS.
There are many features that GCC doesn't implement or doesn't implement well. Auto-parallelization has and probably never will be a supported feature in GCC due to the technical hurdles. Auto-vectorization support is new, but many commercial competing compilers had support for this in the past. Nowdays, though, only hardware-vendor compilers can keep the marketshare needed to support such features. GCC is a weak compiler for RISC chips like the PowerPC compared to IBM and even Apple's Mac OS 9-only MrC compiler. It lags behind Intel's compiler for x86 systems in performance as well. Don't even get me started on g++ and its sundry issues, especially the way that it refuses to play nice with other compilers and follow system ABIs.
Okay, maybe I'm disparaging GCC as a project now, but it's still an amazing achievement. GCC isn't perfect; no software is. However, there were plenty of niches that other vendors could've filled if they weren't competing with GCC's price. The only competing compilers which are still alive are either (a) free or (b) sold by the vendor of the expensive, non-x86 hardware that they work with.
And anyone who isn't male, or isn't white, or can't sing the Star Spangled Banner from memory, or doesn't have a licensed copy of the Holy Bible with a licensed Bible Reader application from Microsoft that is certified to comply with DRM standards.
...And Republican. Aw, crap. I guess it's "-1, Redundant" for me...
Basically, the last time that a major non-Windows worm threatened the stability of internet was back when the majority of computers on the Internet weren't running Windows. There have been numerouswormssincethen for UNIX & Linux, but their market penetration has been low enough not to seriously hurt the whole internet. This is not as good of a thing as you indicate.
Dr. Evil:...We demand payment for our ONE TRILLION lines of UNIX code. [puts pinky finger to edge of mouth] [IBM board bursts into laughter.] IBM CEO: One trillion lines of code? [laughs] There isn't that many lines in the entire GNU system! I mean, you might as well demand payment for a billion-jillion-bazillion lines of code!
Whew! I'm glad this had to do with bad beer. When I saw it in Science Slashbox, I was scared that it had to do with giving blood at pressure-hose speeds!
Well, it shouldn't matter anyway; these weapons aren't true grasers, though the principle is oddly symmetric.
A graser, like any "?aser" device works by stimulating energized electrons to transition to a lower shell immediately (instead of at a random time) by smacking another photon into it, causing the atom emit a photon (always of a certain frequency) in the same direction that the original photon was moving. The gamma decay device works by stimulating the nucleus is a very similar way with X-rays until it raises the chances that the nucleus will randomly decay.
It's kind of like a graser, but with the nucleus instead of the electron shells. That and once an atom has served its purpose once, it's no longer useable for the same trick thanks to having decayed. Though it technically fits each letter in the acronym GRASER, the gamma decay weapon deserves another name entirely.
Incorrect, according to the Honorverse web encyclopedia. You're probably thinking of grav lances or the fact that grasers in the Honorverse use gravity lenses for focus and aiming. GRASERs are in fact weapons based on Gamma Rays Amplified by Stimulated Emission of Radiation. This is done with nuclear stimulation because nothing else can provide the energy to stimulate the emission of gamma rays. Saying gamma rays are "just very highly energetic photons" is like say that that supersonic craft are "just very fast planes." There are engineering problems that require vastly different approaches that common, less extreme implementations of the basic ideas (lasers or planes).
By the way, "microwave lasers" are usually called MASERs. The L in LASER only refers to visible light.
Actually, this is dust and dirt being worn into the tar by drivers. You ever notice how it's darker in the middle and lighter where the tires normally go over it? This is why.
Doping diamond would work the same way, if you can get it to work at all.
Quoth the article:
The third big challenge has been the most daunting for materials scientists: To form microchip circuits, positive and negative conductors are needed. Diamond is an inherent insulator - it doesn't conduct electricity. But both Gemesis and Apollo have been able to inject boron into the lattice, which creates a positive charge. Until now, though, no one had been able to manufacture a negatively charged, or n-type, diamond with sufficient conductivity. When I visit Butler in Washington, he can barely contain his glee. "There's been a major breakthrough," he tells me. In June, together with scientists from Israel and France, he announced a novel way of inverting boron's natural conductivity to form a boron-doped n-type diamond. "We now have a p-n junction," Butler says. "Which means that we have a diamond semiconductor that really works. I can now see an Intel diamond Pentium chip on the horizon."
Funny how the article is actually filled with informative content like that...
Well, any two-bit sci-fi author can tell you that carbon and silicon are similar enough that life can be based on either carbon -or- silicon....and any good organic chemist will tell you that that's bunk. The extra shell of electrons in silicon keeps silicon atoms too far apart to easily form double-bonds and makes triple-bonds impossible as well as forcing single-bonds between silicon atoms to be much weaker that carbon-carbon single-bonds. This property prevents silicon-based chemistry from forming the basic geometries that make up the protein, carbohydrate, and nucleic acid structures that are the basis for all carbon-based life on Earth. Silicon-based life is a wishful fancy that is not grounded in reality.
Most people that make any significant contributions to their field do so before they're 30.
In 300 years, current modes of human cognition will be outdated and irrelevant to the people actually getting work done. If we manage to catch the wave of increasing longevity and ride it that far, we will either no longer be anything like we are now, or we will be fossils and relics kept around by our successors as we do children and the elderly today. The future of progress will be in enhanced intelligence. Whether this intelligence is machine or augmented biology is irrelevant. Once it becomes possible to be smarter than human, it will be a societal inevitability.
People who grow useless past age 30 will be a thing of the past unless the next generation's model of thinking is as vastly superior to theirs as a modern computer is to it's decade old ancestor.
What does this say about SCO's belief in the lawsuit?
They could be hedging their bets regardless of whether or not they think their case has any merits. Let's look at all the possible outcomes on a two-way axis of SCO winning or losing and their execs selling now or after the trial.
1) SCO loses but the execs cashed out now: They profit greatly while the price is currently inflated, and unless a later trial can prove that they knew their case was baseless, you probably can prove beyond a reasonable doubt that they're guilty of insider trading. They get rich.
2) SCO loses but the execs wait to cash out: Not only did they miss a chance to cash out while their stock options were valuable, but they may no longer have a company that can pay them their salary.
3) SCO wins and the execs cashed out now: Congratulations! Not only are they rich from their earlier transactions, but as the heads of a now richly profitable company with next to no overhead costs, they can safely vote themselves huge salary increases and loads of stock options with the blessings of the new, enriched shareholders who rushed in after the lawsuits started. They are the head of a company with at least a few years of bright future with profitable litigation and extortion (*cough* I-mean-licensing) ahead of them.
4) SCO wins but the execs waited to cash out: You're still in charge of a the market leader in squeezing customers over IP legally filtched from other people's hard work and dedication, and you've got plenty of options to now vest at even higher values.
In situations #1, #3, and #4, SCO executives become rich. While #4 is the most profitable, it's the most risky because it has the chance of becoming #2, which nets them no significant gain. If they sell-out now, they become rich either way. Essentially, SCO's powers-that-be are taking the safest route possible to riches. Insider trading is very tricky to prove. You have to prove that the person absolutely knew about a forthcoming change in the fortune of a company that no outsider could've known about. This is why so many Enron and Worldcom people got off, and why SCO people will probably get off scot-free.
They are taking the most rational move, according to game theory. They should only hold their investments if they know 100% that they will win. No matter how solid they think their case is, as rational people, they should realize that going up against IBM is never a sure bet. Thus, they take the appropriate action and get what money they can now, knowing that a win will only reward them with more money while a loss will only close the door on an opportunity that they will have already taken advantage of.
Slashdot Mirror
As to API calls -- that's where various layers of abstraction come in useful. If properly designed, a program should not be hard to port to any platform that offers the necessary features.
As a person who works almost entirely on maintaining the UNIX port of a Windows/UNIX application where 90% of its developers only work on Windows, I can tell you that first-hand that this isn't as easy as you say. First, you have to design the software from ground-up to be portable. That means making sure that you never call system-specific APIs where possible and always step through the abstraction layer. If your application wasn't designed to be ported at the very beginning, then going back through the codebase and retooling all calls to native system calls is extremely time-consuming and leaves behind terribly unmaintainable code. Second, even if you did do a nice abstraction layer, platform-specific demands for features will eventually render your abstraction layer harder and harder to maintain. It isn't that it's impossible to have cleanly implemented abstraction layer that doesn't build up horrible kluges over time -- it's just extremely time-consuming and thus considered economically unfeasable in the real world.
Because of the slow decay of portability layers over time, it's actually far easier to just write single-platform code. Backend code tends to be easy to write portably, but front-end code is always nightmarish when going between GUI and graphics libraries. OpenGL vs. DirectX, Motif vs. Win32, COM vs. CORBA -- the small incompatibilities between the way these libraries work always results in some horrible glue code.
I am sure Microsoft knows what the hell they are doing, given that they have been in the business for quite a while.
Me too. This is why the Mac Business Unit has a completely separate codebase for IE for Mac. This is one reason why IE for Mac was far more standards-compliant and far more bug-free than IE for Windows. They didn't share ANY code.
Ever notice that Mac Office and Windows Office look nothing like each other? I don't think it's because of porting difficulties [...]
The fact that the two products don't resemble each other at all should be the first warning that they have a lot of code that is not shared between the two platforms. MS tried straight-up porting the suite back in the 90s, and it was disasterously received on the Mac for 3 reasons -- it didn't look like a Mac app at (including the previous versions of Word, Excel, etc.), it was slow and a memory hog, and it was buggy. All of these were because of the disasterous backporting effort to take the Windows codebase and use it on the Mac. They essentially wrote a huge library that contained large chunks of the Windows APIs ported to run on top of the MacOS APIs. (This is ironic since all of these applications started on the Mac and were ported to Windows from there before the differences between the two were considered to be "unmaintainable.") Over the years since then, the Mac version of Office has eliminated more and more ported Windows code and gone back to using more and more MacOS-only code. There were serious porting difficulties with their software. That's why they prefer to just treat the two as separate codebases whenever possible.
Cheap is a relative term. It takes nearly half a day to deposit a millimeter's width. Also despite being one of the hardest substances on Earth, diamond is very brittle and would not hold up well to being dropped while in the thin, yet wide shape of a CD.
Except that you've just eliminated the need for musicianship in the vocalist. Now, basically, songwriting and backup bands are all that's important. (Hey.... Maybe this isn't such a bad thing after all. Freakin' prima-donna front men...)
The technology is already in use to create pop acts and has been used throughout the late 90s for this very purpose. The very pop acts that the other poster listed are known to use autotuners. Instead of finding good singers, all the recording industry has to do is find pretty people who can sing passably well enough to work with the autotuner. It's essentially the final nail in the coffin of actual talent in pop acts.
This "levels the playing field" in an industry supposedly based on bringing the best of the best to the national stage. Standards in the recording industry have been slipping for years. One reason for this is that truly talented and popular acts are hard to keep control of contractually. They have enough creativity and talent to jump ship to other labels or create their own labels and survive for years. This is pretty much the opposite of what the recording industry wants. This technology will allow them to take more talentless nobodies and propel them to the national stage with the implicit understanding that without the company that sponsored them, they are nothing. More talented yet more difficult to control musicians can be left more and more to the wayside. In essence, it allows a talent-based industry to get away with selling products without true talent, thus cutting long-term expenses for them.
This technology's a gift from the heavens in karaoke bars, but it'll be just another step in purging all vitality and talent from popular music.
Not necessarily, no. Finding them inserting data into the network would be like finding a needle in a haystack. If they do it properly, they could insert data into the network at any point and have it propogate to the other systems to mail out. Look at how Gnutella propogates searches sometime. Make it push-based and remove the need for an identifying address for the origin to be returned, and you've got a distribution scheme ideal for spamming. You just feed new emails and new addresses into the network and watch it go.
Nearly everyday traffic on the 6 lane interstate suddenly comes to a full stop. When you get going again and drive up a little ways, there was no reason to stop, no accident, no debris in the road.
Ah, it's time for a little MLP...
A layman's description of why traffic backs up for no apparent reason and how to stop it.
An article about an actual German study on the the physics of gridlock.
Never mind. Someone did beat him to the punch. Shame on the moderators for not modding them up either.
I'm surprised no one beat him to the punch on this one. Shame on the rest of you.
Along with most people here, I've seen pretty good anecdotal evidence to believe that aches and pains can predict the weather. Ultimately, the only way you're going to find out is if you track changes in barometric pressure each day and log all aches and pains for a while.
However, while you're doing so, I'd recommend that you also log your caffeine intake. Over a decade of nearly migrane-level headaches has become just a part of my past since I discovered that caffeine withdrawl was causing my problem and simply swore off of caffeine permanently. Most people who have aches and pains before the weather have them in their joints. I've never heard of headaches as a symptom before.
Check your caffeine intake levels as an indicator.
I also pointed out the possibility Microsoft is using SCO as a proxy in an effort to exterminate one of the few remaining threats to their monopoly.
That wasn't a good idea. Now you've entered conspiracy-theory land, where the authorities simply stop listening to you. There is absolutely no proof whatsoever that MS is behind this, and you just sound like another drooling paranoid anti-MS zealot when you go around repeating stuff like this. I violently dislike MS too, but give credit where credit's due. SCO is probably doing all of this on their own for the profit of their own executives.
All MS did was license some UNIX stuff early on, right after the case got started. That's circumstantial evidence at best -- correlation, not causation. You're not going to win any friends among prosecutors if you make unfounded accusations like that.
Not only is she smart, but she's damned cute too! She's got my vote!
I'm glad to see that we the American people continue to be issue-driven in our selection of candidates.
Well... maybe. The cheapest man won. I'm not disparaging GCC as a technical project, but the reason that all those other developers went out of business is because one of their competitors cost nothing instead of cositng hundreds or thousands. This is essentially the reason why Internet Explorer defeated Netscape 4 way back in the day when Netscape costed money. This is why Doublespace destroyed Stacker, why Windows Media Player is killing Real, and why you don't hear anything about webservers other than Apache and IIS.
There are many features that GCC doesn't implement or doesn't implement well. Auto-parallelization has and probably never will be a supported feature in GCC due to the technical hurdles. Auto-vectorization support is new, but many commercial competing compilers had support for this in the past. Nowdays, though, only hardware-vendor compilers can keep the marketshare needed to support such features. GCC is a weak compiler for RISC chips like the PowerPC compared to IBM and even Apple's Mac OS 9-only MrC compiler. It lags behind Intel's compiler for x86 systems in performance as well. Don't even get me started on g++ and its sundry issues, especially the way that it refuses to play nice with other compilers and follow system ABIs.
Okay, maybe I'm disparaging GCC as a project now, but it's still an amazing achievement. GCC isn't perfect; no software is. However, there were plenty of niches that other vendors could've filled if they weren't competing with GCC's price. The only competing compilers which are still alive are either (a) free or (b) sold by the vendor of the expensive, non-x86 hardware that they work with.
And anyone who isn't male, or isn't white, or can't sing the Star Spangled Banner from memory, or doesn't have a licensed copy of the Holy Bible with a licensed Bible Reader application from Microsoft that is certified to comply with DRM standards.
...And Republican. Aw, crap. I guess it's "-1, Redundant" for me...
Never, huh?
Basically, the last time that a major non-Windows worm threatened the stability of internet was back when the majority of computers on the Internet weren't running Windows. There have been numerous worms since then for UNIX & Linux, but their market penetration has been low enough not to seriously hurt the whole internet. This is not as good of a thing as you indicate.
It's from the underwear gnomes episode of South Park and was one of the most biting satirical indictments of the dot-com era's business models.
I keep flashing back to Austin Powers 2...
...We demand payment for our ONE TRILLION lines of UNIX code. [puts pinky finger to edge of mouth]
Dr. Evil:
[IBM board bursts into laughter.]
IBM CEO: One trillion lines of code? [laughs] There isn't that many lines in the entire GNU system! I mean, you might as well demand payment for a billion-jillion-bazillion lines of code!
There's no step three!
</Jeff Goldblum>
Whew! I'm glad this had to do with bad beer. When I saw it in Science Slashbox, I was scared that it had to do with giving blood at pressure-hose speeds!
Visible to superhumans, maybe. What about infrared, ultraviolet, and x-ray lasers?
Valid point.
Well, it shouldn't matter anyway; these weapons aren't true grasers, though the principle is oddly symmetric.
A graser, like any "?aser" device works by stimulating energized electrons to transition to a lower shell immediately (instead of at a random time) by smacking another photon into it, causing the atom emit a photon (always of a certain frequency) in the same direction that the original photon was moving. The gamma decay device works by stimulating the nucleus is a very similar way with X-rays until it raises the chances that the nucleus will randomly decay.
It's kind of like a graser, but with the nucleus instead of the electron shells. That and once an atom has served its purpose once, it's no longer useable for the same trick thanks to having decayed. Though it technically fits each letter in the acronym GRASER, the gamma decay weapon deserves another name entirely.
My inner evil marketroid recommends Gradec.
Incorrect, according to the Honorverse web encyclopedia. You're probably thinking of grav lances or the fact that grasers in the Honorverse use gravity lenses for focus and aiming. GRASERs are in fact weapons based on Gamma Rays Amplified by Stimulated Emission of Radiation. This is done with nuclear stimulation because nothing else can provide the energy to stimulate the emission of gamma rays. Saying gamma rays are "just very highly energetic photons" is like say that that supersonic craft are "just very fast planes." There are engineering problems that require vastly different approaches that common, less extreme implementations of the basic ideas (lasers or planes).
By the way, "microwave lasers" are usually called MASERs. The L in LASER only refers to visible light.
Actually, this is dust and dirt being worn into the tar by drivers. You ever notice how it's darker in the middle and lighter where the tires normally go over it? This is why.
Doping diamond would work the same way, if you can get it to work at all.
Quoth the article:
Funny how the article is actually filled with informative content like that...
Well, any two-bit sci-fi author can tell you that carbon and silicon are similar enough that life can be based on either carbon -or- silicon. ...and any good organic chemist will tell you that that's bunk. The extra shell of electrons in silicon keeps silicon atoms too far apart to easily form double-bonds and makes triple-bonds impossible as well as forcing single-bonds between silicon atoms to be much weaker that carbon-carbon single-bonds. This property prevents silicon-based chemistry from forming the basic geometries that make up the protein, carbohydrate, and nucleic acid structures that are the basis for all carbon-based life on Earth. Silicon-based life is a wishful fancy that is not grounded in reality.
Most people that make any significant contributions to their field do so before they're 30.
In 300 years, current modes of human cognition will be outdated and irrelevant to the people actually getting work done. If we manage to catch the wave of increasing longevity and ride it that far, we will either no longer be anything like we are now, or we will be fossils and relics kept around by our successors as we do children and the elderly today. The future of progress will be in enhanced intelligence. Whether this intelligence is machine or augmented biology is irrelevant. Once it becomes possible to be smarter than human, it will be a societal inevitability.
People who grow useless past age 30 will be a thing of the past unless the next generation's model of thinking is as vastly superior to theirs as a modern computer is to it's decade old ancestor.
What does this say about SCO's belief in the lawsuit?
They could be hedging their bets regardless of whether or not they think their case has any merits. Let's look at all the possible outcomes on a two-way axis of SCO winning or losing and their execs selling now or after the trial.
1) SCO loses but the execs cashed out now:
They profit greatly while the price is currently inflated, and unless a later trial can prove that they knew their case was baseless, you probably can prove beyond a reasonable doubt that they're guilty of insider trading. They get rich.
2) SCO loses but the execs wait to cash out:
Not only did they miss a chance to cash out while their stock options were valuable, but they may no longer have a company that can pay them their salary.
3) SCO wins and the execs cashed out now:
Congratulations! Not only are they rich from their earlier transactions, but as the heads of a now richly profitable company with next to no overhead costs, they can safely vote themselves huge salary increases and loads of stock options with the blessings of the new, enriched shareholders who rushed in after the lawsuits started. They are the head of a company with at least a few years of bright future with profitable litigation and extortion (*cough* I-mean-licensing) ahead of them.
4) SCO wins but the execs waited to cash out:
You're still in charge of a the market leader in squeezing customers over IP legally filtched from other people's hard work and dedication, and you've got plenty of options to now vest at even higher values.
In situations #1, #3, and #4, SCO executives become rich. While #4 is the most profitable, it's the most risky because it has the chance of becoming #2, which nets them no significant gain. If they sell-out now, they become rich either way. Essentially, SCO's powers-that-be are taking the safest route possible to riches. Insider trading is very tricky to prove. You have to prove that the person absolutely knew about a forthcoming change in the fortune of a company that no outsider could've known about. This is why so many Enron and Worldcom people got off, and why SCO people will probably get off scot-free.
They are taking the most rational move, according to game theory. They should only hold their investments if they know 100% that they will win. No matter how solid they think their case is, as rational people, they should realize that going up against IBM is never a sure bet. Thus, they take the appropriate action and get what money they can now, knowing that a win will only reward them with more money while a loss will only close the door on an opportunity that they will have already taken advantage of.