You will always be able to install on a local machine as long as there is such a thing (and there will always be a market, so someone will fill that "hardware niche" if it comes to that), because there will always be plenty of hackers wanting to get the guts of the system out right there on their own box. OSS ensures that there's no company out there that can say "You have to pay $X per day to use the app now".
Good Ideas expose their hidden Good Features over time -- Open Source is no exception.
I completely concur. Reading this article I got a distinct feeling that this cast of "stars" wasn't much asset-wise, particularly if the headliner is marca. Most of these guys were merely in the right place at the right time -- the real stars of the Netscape show left a long time ago. And they're going forward to tackle a vaguely defined area which is currently the arena of the "big boys", trying to be guerillas again -- only, they're not guerillas any more.
Sure, every VC company wants a piece of this, just to hedge their bets; but this won't be a winner for them margin-wise. The winners for VC's are still the guerilla company's: there's a lot of room to grow from ground zero.
You're right -- it's a 'SELECT * from ORDERS where account number = 'cookienumber' AND WHERE orderstatus = pending ORDER by orderdate;'
If it cost them 1000's of hours to come up with this, anyone know the hotline to the guy who does their IT/MIS/etc hiring? I'll do it in 2 hours (gotta get in a game of quake ya know), and only charge them for 500.
Hopefully as little will come of this patent case as came of that.
NO! On the contrary -- hopefully this thing will go the length: all the way to a court battle where the patent is finally thrown out as the overly broad nonsense it is -- within full view of country, press, and (insert favorite diety).
Perhaps then the USPTO will (1) get resources to better deal with patents, (2) be overhauled, (3) or be legislated to stop granting software patents.
I agree that VB brings "programming" to the Common Man. It also allows Programmers to get results in an MS world. More often than not (IMHO) it enables an accountant/secretary/administrative assistant to hack together a poorly-designed stopgap solution to an office automation problem which will ultimately have to be scrapped (after being kludged 15-20 times in-house) in order to provide a robust solution for real growth -- resulting in much higher IT costs overall than if they had done it "right" from the start.
As far as SOAP (MS-XML) is concerned, I cringed when I first read about it. After seeing what MS has done to HTML, the CSS-DOM, their application OM (not to be confused with an AOM), Java, Javascript,... am I forgetting other major standards they've bastardized and butchered?... I guarantee you that it will be just broken enough for someone to bring another anti-trust/class-action suit against MS over predatory/uncompetitive behavior. I'd even be willing to put $ on it.
You can get XML support in Java to varying degrees. Mozilla (whose components are to be usable inside a JVM, or callable from C++ add-ons) will have the best XML support available when it is released (and don't tell me it's vaporware -- I'm posting this message with it...) -- even on the MS platform.
The pure economics of Microsoft's practices will ensure continued loss of Good developers: charging so much for OS and development tools that one could have bought a decent server for the $; said developer tools being ever-changing kludges encasing buggy and chimeric APIs; unportable standards implementations.... Move to an OpenSource OS (*BSD, Linux,...) and buy that extra server, use proven tools which obey the standards...
People may mumble about PHB's (and yes, corporate blindness exists especially in the halls of the F500), but my employers are thrilled to purchase equipment that is more secure, more robust, better utilized, and more readily administerable -- instead of forking out $K's every year for the same old MS software that has higher version numbers, requires more RAM, and crashes more often.
As an owner of Applied Cryptography I pulled it out to see if the remaining numbers in that sequence made sense as page numbers -- try the first word of each page, last word of each page; nothing sensible; took the numbers as pairs in hopes of one being page number, the next being word on the page ; no dice -- it can't possibly work on the other books (look at the numbers)., Also there are different numbers after the ISBNs for each book, so pairing, etc., won't work.
Then took the numbers modulo 26 to see if they map to the alphabet. Then tried modulo the title length... no dice.
Took the first letter and/or last letter from each page with that number. Nope.
went and got some pot roast and that's where it stands.
Fine, so long as the company you are working for (which has enough revenue to justify using the big iron) is happy with throwing away your code when they realize they want to move up to the next big piece of iron a year later, which may or may not be a Sun box, or capable of supporting your server-dependent tweaks.
IMHE(xperience), the programmer resources are vastly more expensive than the iron (if they are not then your dollars are being spent in the wrong places -- or you have management which excels in getting great programmers for cheap prices (in which case they aren't getting as good a deal as they think)). Re-engineering code for a move to a new platform because you NEED to move to the new platform is very expensive.
Good management will hire good people to do it right the first time (portable, with the right algorithms, and rock solid code) and deploy it on the right systems with an eye towards potential changes in platforms down the road.
One might argue that isolating platform-dependent code into its own "module" (depending upon what a module is in your language of choice) is the solution -- your server-dependent tweaks go there, and that's what changes in the future. Point taken, but then why are we arguing about not using Java then: that's exactly what JNI was designed for.
The "official Sun blessing" was given back in August 1998, and did next to nothing with regard to speeding the emergence of a Java2 environment on Linux. The fact that IBM is going to soon release a good JDK on Linux in a few months has spurred Sun to _finally_ give in and release an official (and, IMHO probably (intentionally) buggy) Java for Linux port, so as not to completely lose face. You can bet your bottom one that 1.2.3 will be out for most other platforms shortly thereafter, again leaving Linux a few revisions behind.
Why? Because, while Sun wants people to be able to leave M$-land, they want them to leave it for Sun-land, and that generally doesn't mean going to Linux.
I'll get used to it, but you can bet I'll be watching my p's undt q's: I've (at least once to my knowledge, maybe more often) gotten contracts over people on nothing more than grammar (i.e., we both seemed qualified, but I knew the difference between "then" and "than"...).
So, sure, let grammar/spelling go down the toilet -- it will keep me in the money.
Just glad to see people admit that the NT4 tools are crapware. I did spend a couple of hours looking through the Win2K installation he had going. I'll stick by my original post.
You'll note I didn't state that Win2K is NT4, just that for such a "revolutionary" change it still looks, acts, walk, smells, and quacks like the same old duck.
In some other related posts it has been noted that Linux is just coming into its own regarding some of the GUI features windows has had for some time. Of course, Windows is still not coming into many of the features Linux (and *nix before it) have had for many more years (security, solid TCP/IP stack, easy remote administration, true multi-user operation, etc.). And I have 1500 true-type fonts available on my desktop Linux box, with 36 full-screen virtual desktops, intense themability, ssh-remotable (with Kerberos key management) X sessions (across a firewall which accepts only ssh sessions authenticated by RSA). I've paid only the price of hardware, which is lower than it would have been for the hardware to run a comparable (well, there isn't really such a thing, but if there were) Windows setup.
Oh, almost forgot (that machine is a PII): 12:44pm up 43 days, 14:01, 15 users, load average: 1.19, 1.25, 1.26 (sorry the uptime is so low -- I upgrade the kernel every so often).
At the risk of responding to yet another troll, I refuse to admit that IE is a good product (in any of a number of senses of the word "good"). Unfortunately, neither is most of its competition... for the moment.
I've begun using mozilla more and more regularly due to the fact that Netscape on Linux is a piece of trash, and I just don't like using lynx for all my browsing.
When I started to really pick Mozilla apart (especially under the hood) I realized that in a few months Microsoft will have finally lost their browser war. Adobe will lose massive "market share" to the GIMP within the next year, even on the Win32/64 platforms.
And, to top it off, I was playing with the most recent build of Win2k at a friend's house, alongside an NT4.0 machine. To tell the truth, I couldn't tell which was which (other than the "Win 2000 build number...." at the corner of the screen). The biggest improvement I could see was that windows minimize really quickly -- they still start up slow as dogshit through a panty, and the disks still sound like someone is testing a cache-busting head scheduler on them everytime you access the file system.
Leading me (along with the exponential-feeling progress of the major linux distros and GNU software in the past year) to comment more than once to people that "Microsoft is in trouble.".
Fact of the matter is, IMNSHO, within 2 years they will be falling back upon the Office Suite which was their foot in the door for so many years. Whatever the outcome of the court case, it will be moot,as many predicted -- but not for the reasons they predicted. People thought that Microsoft would be penalized too late for issues (like the bundling of IE) that would be unimportant. Instead, the outcome of the case will be moot because Microsoft is already on the downhill slope of a long and steep descent into ruin.
Yes, it's pre-alpha and it crashes occasionally (only about 30% more often than my Netscape 4.08/4.61/4.7 installs over the past few months).
There is an argument about possible "bloat" -- you got a problem, don't load the components you don't want (this is the first browser in which you will *ever* have the opportunity to do that) so "can it" (maybe there's a market for making "Mozillas distros" with various configurations... hmm....).
Every once in a while there comes along something so cool in a product that that one feature alone justifies everything else. There are arguably a number of them, but I just came across this one for the first time and almost pissed myself. Under "View". there's a "Translate" menu. Go to your favorite website and select a translation. This is not some "cut-and-paste babelfish" hack -- this thing (using a 3rd party service) re-renders the entire page (properly even), with the text translated. It's like you hit the official German/Spanish/Japanese site of wherever you were located, but you didn't...
damnnation... they were nice enough to only take IMAGES of his hard drive? I thought whenever the Feds came they generally take all your goods and you likely never see them again (regardless of guilt).
Such a document does exist. We include it in every contract/consulting proposal we do. Our customers lose too much time and $ trying to make NT work that we need to convince them up-front.
Check out the Kirch Paper, the ultimate tool against FUD.
They make it a crime to send unsolicited commercial e-mail to a recipient whose ISP has a posted policy forbidding it. Tying in the source ISP might be part of the issue, but this is hard to pass the courts (free speech, prior restraint, all that sorta stuff tends to get in the way). At most forcing the source ISP to submit usage/registration records under force of court order is probably sufficient. Of course for obliging ISPs "conspiracy to commit a felony" (if the crime is a felony) is likely sufficient to keep ISPs from "knowingly" harboring spammers.
As far as tracing spam, yes, Virginia, much of the unsolicited email out there is essentially forged. However, most forgeries are poor, and few forgeries are truly hard to trace. In addition, open SMTP relays are becoming harder and harder to find. In addition to any legislation that exists, resources like the RealTime BlackHole List make it harder and harder for the spammer to even send or relay spam.
Of course this discussion is completely independent of whether I believe illegalizing spam is a good idea. I personally think the government shouldn't have its nose in the issue, and it reeks of censorship. Given a little more time users will be more savvy, tools like the BlackHole List will be more prevalent, and spam-ridden ISPs (like AOL) will be forced to filter more actively or lose a noticeable number of customers to places (like Mindspring/Earthlink) which do more filtering. I have had a Perl source and content-based spam filter in place for over two years now and have filtered over 700 spam mails automatically (about 10 false-positives...). Between that and the judicious use of spam-drops (like the hotmail address listed above) my life is generally spam-free.
Same story here. Had MCI, got rid of it because it was poor service for too much $. Kept getting billed (was getting a bill from AT&T and MCI for the same exact calls!), complained, called, wrote letters, bitched, kept getting billed while being told "no you don't have to pay that it's all taken care of", get bills with credit threats, etc., threatened to "gruesomely execute every last one of you shitheads", finally got the state's Public Service Commission to do them in and the bills stopped (it's just a matter of them wanting to stop the bills badly enough...).
That was about 18 months ago. Friend across the country related an almost identical story to me a few weeks ago (just happened to him). I get a call from some poor telemarketer drone lady (it's not her fault she picked an awful employer) telling me I should "switch to MCI/WorldCom". After telling her "no thanks, not interested" she proceeded to try to sell me on the deal and I told her I had MCI before and it was awful. Her response: "Well, we're MCI/WorldCom now, and things are different". I couldn't help it: "What you ruin my credit first then send me the 6 months of bills for service I didn't order - instead of waiting until later to ruin my credit? Listen, I know it's not your fault, but if anything MCI has gotten worse since they've gotten bigger."
It is. Given a proposed solution (i.e., a factorization) it is trivially verifiable in polynomial time, regardless of which of any number of reasonable modern computational models you wish to use.
2) That NP != P
Granted, since otherwise the problem is bounded by a polynomial.
3) That we live in a P world
How very technical of you. What is that supposed to mean?
If what you are trying to arrive at are sufficient conditions for a lower-bound on the complexity of integer factorization you are likely wasting your time. If I remember correctly it is known that Factorization is in the intersection of NP and co-NP (someone correct me if I've forgotten), but it has not been shown NP-complete (and is thought not to be).
But this says little about time required to solve (lower bounds) which are a more difficult matter all together. Depending upon your computational model it may be a trivial problem (consider the model where a machine can perform factorizations in one step -- then this is a constant-time problem). The point being that some magical computational system (here everyone reads "quantum computing", but it could be any magic-like technology -- and preferably less like vaporware in my opinion) or algorithmic revelation may reduce the problem quickly to tractability.
My advisor and I used genetic algorithms to do optimizations...
It's late and it's been a few years. I meant simulated annealing instead of genetic algorithms. We looked at both and found SA better than GA for our particular optimization problem. Doesn't affect the post otherwise:-)
My advisor and I used genetic algorithms to do optimizations for an NP-complete problem which arises in building certain types of genetic sequencing "chips" (the paper is here (postscript). This is a practical application which seems much harder to me. Our optimizer ran on a *single* Sparcstation 5 in a matter of minutes.
The Intelligent Robotics Laboratory at Vanderbilt University (where I spent a summer) was doing things back in 1994 that were much more difficult to do than this project without having to resort to GA's (which are generally considered by real algorithmicists to be a last resort (it was in our paper cited above)). And neither of these were the "cutting edge" environments you get at places like CalTech, MIT, or some industrial labs.
And, I wish I could give you a link (but I can't remember the fellow's name) to the research (which may have appeared on Slashdot) being done in simulating evolution in physical environments. The main researcher in question gave a seminar at my former employer (a company which does mathematical modelling of complex phenomena) a year and a half ago showing film clips of "evolved" computer "life forms" which solved physical motion problems in ways eerily similar to extant "real" creatures. Same lab (I've searched for 45 minutes now for a link or an old email about the presentation and am coming up dry so don't ask -- if I find it I'll post it) was doing visualizations of "evolved programs" where they were finding evolved (GA's) redundancies in coding operations similar to those found in actual natural DNA/RNA.
So using 1000 pentiums to make a lego bridge via GA's is newsworthy? Bah! Gimme a break.
Slashdot Mirror
You will always be able to install on a local machine as long as there is such a thing (and there will always be a market, so someone will fill that "hardware niche" if it comes to that), because there will always be plenty of hackers wanting to get the guts of the system out right there on their own box. OSS ensures that there's no company out there that can say "You have to pay $X per day to use the app now".
Good Ideas expose their hidden Good Features over time -- Open Source is no exception.
Sure, every VC company wants a piece of this, just to hedge their bets; but this won't be a winner for them margin-wise. The winners for VC's are still the guerilla company's: there's a lot of room to grow from ground zero.
If it cost them 1000's of hours to come up with this, anyone know the hotline to the guy who does their IT/MIS/etc hiring? I'll do it in 2 hours (gotta get in a game of quake ya know), and only charge them for 500.
NO! On the contrary -- hopefully this thing will go the length: all the way to a court battle where the patent is finally thrown out as the overly broad nonsense it is -- within full view of country, press, and (insert favorite diety).
Perhaps then the USPTO will (1) get resources to better deal with patents, (2) be overhauled, (3) or be legislated to stop granting software patents.
As far as SOAP (MS-XML) is concerned, I cringed when I first read about it. After seeing what MS has done to HTML, the CSS-DOM, their application OM (not to be confused with an AOM), Java, Javascript, ... am I forgetting other major standards they've bastardized and butchered?... I guarantee you that it will be just broken enough for someone to bring another anti-trust/class-action suit against MS over predatory/uncompetitive behavior. I'd even be willing to put $ on it.
You can get XML support in Java to varying degrees. Mozilla (whose components are to be usable inside a JVM, or callable from C++ add-ons) will have the best XML support available when it is released (and don't tell me it's vaporware -- I'm posting this message with it...) -- even on the MS platform.
The pure economics of Microsoft's practices will ensure continued loss of Good developers: charging so much for OS and development tools that one could have bought a decent server for the $; said developer tools being ever-changing kludges encasing buggy and chimeric APIs; unportable standards implementations.... Move to an OpenSource OS (*BSD, Linux,...) and buy that extra server, use proven tools which obey the standards...
People may mumble about PHB's (and yes, corporate blindness exists especially in the halls of the F500), but my employers are thrilled to purchase equipment that is more secure, more robust, better utilized, and more readily administerable -- instead of forking out $K's every year for the same old MS software that has higher version numbers, requires more RAM, and crashes more often.
Look up the word "meme". This is nothing novel.
Hey, I resent that! I *do* shower occasionally. Maybe "grungy" but "nasty" is definitely going too far!
As an owner of Applied Cryptography I pulled it out to see if the remaining numbers in that sequence made sense as page numbers -- try the first word of each page, last word of each page; nothing sensible; took the numbers as pairs in hopes of one being page number, the next being word on the page ; no dice -- it can't possibly work on the other books (look at the numbers)., Also there are different numbers after the ISBNs for each book, so pairing, etc., won't work.
Then took the numbers modulo 26 to see if they map to the alphabet. Then tried modulo the title length ... no dice.
Took the first letter and/or last letter from each page with that number. Nope.
went and got some pot roast and that's where it stands.
IMHE(xperience), the programmer resources are vastly more expensive than the iron (if they are not then your dollars are being spent in the wrong places -- or you have management which excels in getting great programmers for cheap prices (in which case they aren't getting as good a deal as they think)). Re-engineering code for a move to a new platform because you NEED to move to the new platform is very expensive.
Good management will hire good people to do it right the first time (portable, with the right algorithms, and rock solid code) and deploy it on the right systems with an eye towards potential changes in platforms down the road.
One might argue that isolating platform-dependent code into its own "module" (depending upon what a module is in your language of choice) is the solution -- your server-dependent tweaks go there, and that's what changes in the future. Point taken, but then why are we arguing about not using Java then: that's exactly what JNI was designed for.
Why? Because, while Sun wants people to be able to leave M$-land, they want them to leave it for Sun-land, and that generally doesn't mean going to Linux.
I'll get used to it, but you can bet I'll be watching my p's undt q's: I've (at least once to my knowledge, maybe more often) gotten contracts over people on nothing more than grammar (i.e., we both seemed qualified, but I knew the difference between "then" and "than"...).
So, sure, let grammar/spelling go down the toilet -- it will keep me in the money.
You'll note I didn't state that Win2K is NT4, just that for such a "revolutionary" change it still looks, acts, walk, smells, and quacks like the same old duck.
In some other related posts it has been noted that Linux is just coming into its own regarding some of the GUI features windows has had for some time. Of course, Windows is still not coming into many of the features Linux (and *nix before it) have had for many more years (security, solid TCP/IP stack, easy remote administration, true multi-user operation, etc.). And I have 1500 true-type fonts available on my desktop Linux box, with 36 full-screen virtual desktops, intense themability, ssh-remotable (with Kerberos key management) X sessions (across a firewall which accepts only ssh sessions authenticated by RSA). I've paid only the price of hardware, which is lower than it would have been for the hardware to run a comparable (well, there isn't really such a thing, but if there were) Windows setup.
Oh, almost forgot (that machine is a PII):
12:44pm up 43 days, 14:01, 15 users, load average: 1.19, 1.25, 1.26
(sorry the uptime is so low -- I upgrade the kernel every so often).
I've begun using mozilla more and more regularly due to the fact that Netscape on Linux is a piece of trash, and I just don't like using lynx for all my browsing.
When I started to really pick Mozilla apart (especially under the hood) I realized that in a few months Microsoft will have finally lost their browser war. Adobe will lose massive "market share" to the GIMP within the next year, even on the Win32/64 platforms.
And, to top it off, I was playing with the most recent build of Win2k at a friend's house, alongside an NT4.0 machine. To tell the truth, I couldn't tell which was which (other than the "Win 2000 build number ...." at the corner of the screen). The biggest improvement I could see was that windows minimize really quickly -- they still start up slow as dogshit through a panty, and the disks still sound like someone is testing a cache-busting head scheduler on them everytime you access the file system.
Leading me (along with the exponential-feeling progress of the major linux distros and GNU software in the past year) to comment more than once to people that "Microsoft is in trouble.".
Fact of the matter is, IMNSHO, within 2 years they will be falling back upon the Office Suite which was their foot in the door for so many years. Whatever the outcome of the court case, it will be moot,as many predicted -- but not for the reasons they predicted. People thought that Microsoft would be penalized too late for issues (like the bundling of IE) that would be unimportant. Instead, the outcome of the case will be moot because Microsoft is already on the downhill slope of a long and steep descent into ruin.
Sorry :-)
There is an argument about possible "bloat" -- you got a problem, don't load the components you don't want (this is the first browser in which you will *ever* have the opportunity to do that) so "can it" (maybe there's a market for making "Mozillas distros" with various configurations... hmm....).
Every once in a while there comes along something so cool in a product that that one feature alone justifies everything else. There are arguably a number of them, but I just came across this one for the first time and almost pissed myself. Under "View". there's a "Translate" menu. Go to your favorite website and select a translation. This is not some "cut-and-paste babelfish" hack -- this thing (using a 3rd party service) re-renders the entire page (properly even), with the text translated. It's like you hit the official German/Spanish/Japanese site of wherever you were located, but you didn't...
HOT
damnnation... they were nice enough to only take IMAGES of his hard drive? I thought whenever the Feds came they generally take all your goods and you likely never see them again (regardless of guilt).
Check out the Kirch Paper, the ultimate tool against FUD.
Why does autorpm not do what you want?
They make it a crime to send unsolicited commercial e-mail to a recipient whose ISP has a posted policy forbidding it. Tying in the source ISP might be part of the issue, but this is hard to pass the courts (free speech, prior restraint, all that sorta stuff tends to get in the way). At most forcing the source ISP to submit usage/registration records under force of court order is probably sufficient. Of course for obliging ISPs "conspiracy to commit a felony" (if the crime is a felony) is likely sufficient to keep ISPs from "knowingly" harboring spammers.
As far as tracing spam, yes, Virginia, much of the unsolicited email out there is essentially forged. However, most forgeries are poor, and few forgeries are truly hard to trace. In addition, open SMTP relays are becoming harder and harder to find. In addition to any legislation that exists, resources like the RealTime BlackHole List make it harder and harder for the spammer to even send or relay spam.
Of course this discussion is completely independent of whether I believe illegalizing spam is a good idea. I personally think the government shouldn't have its nose in the issue, and it reeks of censorship. Given a little more time users will be more savvy, tools like the BlackHole List will be more prevalent, and spam-ridden ISPs (like AOL) will be forced to filter more actively or lose a noticeable number of customers to places (like Mindspring/Earthlink) which do more filtering. I have had a Perl source and content-based spam filter in place for over two years now and have filtered over 700 spam mails automatically (about 10 false-positives...). Between that and the judicious use of spam-drops (like the hotmail address listed above) my life is generally spam-free.
That was about 18 months ago. Friend across the country related an almost identical story to me a few weeks ago (just happened to him). I get a call from some poor telemarketer drone lady (it's not her fault she picked an awful employer) telling me I should "switch to MCI/WorldCom". After telling her "no thanks, not interested" she proceeded to try to sell me on the deal and I told her I had MCI before and it was awful. Her response: "Well, we're MCI/WorldCom now, and things are different". I couldn't help it: "What you ruin my credit first then send me the 6 months of bills for service I didn't order - instead of waiting until later to ruin my credit? Listen, I know it's not your fault, but if anything MCI has gotten worse since they've gotten bigger."
It is. Given a proposed solution (i.e., a factorization) it is trivially verifiable in polynomial time, regardless of which of any number of reasonable modern computational models you wish to use.
2) That NP != P
Granted, since otherwise the problem is bounded by a polynomial.
3) That we live in a P world
How very technical of you. What is that supposed to mean?
If what you are trying to arrive at are sufficient conditions for a lower-bound on the complexity of integer factorization you are likely wasting your time. If I remember correctly it is known that Factorization is in the intersection of NP and co-NP (someone correct me if I've forgotten), but it has not been shown NP-complete (and is thought not to be).
But this says little about time required to solve (lower bounds) which are a more difficult matter all together. Depending upon your computational model it may be a trivial problem (consider the model where a machine can perform factorizations in one step -- then this is a constant-time problem). The point being that some magical computational system (here everyone reads "quantum computing", but it could be any magic-like technology -- and preferably less like vaporware in my opinion) or algorithmic revelation may reduce the problem quickly to tractability.
It's late and it's been a few years. I meant simulated annealing instead of genetic algorithms. We looked at both and found SA better than GA for our particular optimization problem. Doesn't affect the post otherwise :-)
The Intelligent Robotics Laboratory at Vanderbilt University (where I spent a summer) was doing things back in 1994 that were much more difficult to do than this project without having to resort to GA's (which are generally considered by real algorithmicists to be a last resort (it was in our paper cited above)). And neither of these were the "cutting edge" environments you get at places like CalTech, MIT, or some industrial labs.
And, I wish I could give you a link (but I can't remember the fellow's name) to the research (which may have appeared on Slashdot) being done in simulating evolution in physical environments. The main researcher in question gave a seminar at my former employer (a company which does mathematical modelling of complex phenomena) a year and a half ago showing film clips of "evolved" computer "life forms" which solved physical motion problems in ways eerily similar to extant "real" creatures. Same lab (I've searched for 45 minutes now for a link or an old email about the presentation and am coming up dry so don't ask -- if I find it I'll post it) was doing visualizations of "evolved programs" where they were finding evolved (GA's) redundancies in coding operations similar to those found in actual natural DNA/RNA.
So using 1000 pentiums to make a lego bridge via GA's is newsworthy? Bah! Gimme a break.