You can look these registrations up on the U.S. Copyright Office web site, http://www.loc.gov/copyright/
I saw only 19 registrations, but most of the "biggies" (GNU C, Emacs, GNU tar, diff, make, m4, etc.) are covered. These are existing registrations. There are potentially serious consequences for infringing these - even with no "actual" damages. Plus FSF could register more of their copyrights any time they wanted.
This looks pretty serious to me. I don't see why there's any doubt about FSF's ability to enforce this. My guess is that they'd rather settle it without having to spend any real money. Even with volunteer lawyers, lawsuits can cost a lot.
The guy in the paper needed somthing to write about. This will quietly settle and it will not be big news.
true... i think its mainly because microshaft is increasing requirements to pump the market, and then FSF things like KDE and Gnome have to do the same for end-user appeal, and to not lose ground to it.
Entire campus a Microsoft free zone? Brilliant! Plenty of people would eagerly contribute to that, I sure would.
Linux not being ready for the desktop is pure BS. Some big Linux promoter like IBM would probably be HAPPY to help with a project like this. People would turn out from all over to assist in such an effort.
The big downside of Linux today (and the only one, really) is rapid interchange of documents with MS Office users. This is an artifical barrier that makes it impractical for most businesses to consider Linux for the end user. However, the issue is far less relevant to a campus community which can largely do whatever it damn well wants. In a college/university setting you're probably not sitting there all day getting business-critical MS Office attachments that you have to bat back instantaneously with comments and revisions. OpenOffice does work extremely well with its own documents. It's just the interchange with MS Office that's a little rocky.
You need a little IMAGINATION folks, not petty bean counting bullshit. No, the $2.4 million does not in itself economically justify this, but it is sure a hell of a nice seed to start something with VASTLY more economic, PR and educational potential.
This type of development requires lots of up front money, with only a statistical likelihood of success. If you want this money to come from investors, they need to see a pot of gold potentially at the end. The only alternative is for the money to come from the government or charitable foundations. Actually, in the real world, funding is from a mix of these sources. However, if you cut back patenting as we now know it, you have to push the funding much more toward public sources, which has its own set of problems.
I don't think many people are ready to completely socialize health-care related R&D. What a scary thought.
Remember WordPerfect, where you could press F3 to "Reveal Codes"? Being an MS Word snob from way back, I used to think that style sheets were much more clever than embedding tags into the document.
So they're going back to tag markup. Probably for the better, and certainly much more open, but it seems ironic that their second screenshot looks a WHOLE LOT like a WordPerfect "Reveal Codes" display.
BTW, I don't think a lot of end users will have ANY concept whatsoever of what an "XML schema" is... Good luck!
Nope. Section 904 of the Manual of Patent Examining Procedure says in very clear terms that they're SUPPOSED to search nonpatent literature:
"The examiner, after having obtained a thorough understanding of the invention disclosed and claimed in the nonprovisional application, then searches the prior art as disclosed in patents and other published documents, i.e., nonpatent literature (NPL)."
And the good ones do just that. I've seen it many times.
Anyone wanting to look up the 4,000+ pending published applications whose abstracts mention the word "internet" can click"
here.
Bear in mind that these are just patent applications (not yet patents), though a lot of them will probably get through. Of course there are many other computer, software and busines-method related patents - you can do your own search if you want, here.
Also, slight correction to what I said in my first post above - not ALL applications get published, because it is possible to opt out by giving up most foreign patent rights. Few applicants do this, though, with the result that most applications now are being published.
Geez, why don't the people here learn something about what they're talking about before posting?
Except for some details on the billing, what the poster described is awfully close to the current process! The current BROKEN process.
The problem is that the patent office sometimes does an incompetent search. They get paid for it too (several hundred dollars, actually).
In this case, for example, as is evident from the front page of the patent, the only thing the patent examiner ever looked at to check the patentability of this application was other issued patents and one trade press article from 1996. He/she obviously never heard of CVS, didn't know squat about versioning systems, and just rushed the thing out. The system isn't supposed to work this way, but this kind of thing keeps happening.
Also, the patent applicant is obligated to cite to the examiner everything he/she knows about that might be germane to patentability. So you have to wonder why THEY didn't cite CVS.
No, this will NEVER improve until there is more of a check on the process. But the current process is one-sided - all between the applicant and a government agency, and neither has much incentive to do a thorough search.
So, this situation is unlikely to improve until the process of obtaining a patent is a allowed to be contested procedure - contested by the people who are likely to be impacted by the patent, giving them an opportunity to see a draft of it and object before it has been issued, and to have a real no-holds-barred contested proceeding before a competent panel of experts before the sucker is approved.
No this is not a new idea. It's called "pre-grant opposition" and it's been floated out many times, and always been shot down, either by the patent lawyers or by the "small inventor" lobby. It is the way things are done in Europe, but not here.
As it is, due to a realtively recent change in the law, new patents are now "published" on the Patent Office web site well before they are allowed.
What we need is for Slashdotters to look at THOSE publications and complain about them BEFORE they issue as patents. No formal opposition procedure now exists, but people can still make noise, and it might be helpful.
In this case there was no publication, because the patent application was filed before the publication law took effect. But that will change over time. The norm will be that every new patent will have been published for the world to see, well before it issues. It would be better if there were a formal channel to DO something about it, but we should at least be reading these publications, right now, instead of merely complaining about patents that have already issued.
Aha! Corporations' newly found fear of the securities lawa may be doing some good.
This is only a CYA "risk factor", of course, to protect MS from lawsuits based on hindsight. Nevertheless, it is still a telling concession.
But fundamentally, isn't it the case that MS really CAN'T compete on price without distoring their entire business model? Maybe they should state that as well!
You could put their mail in a putative spam folder and send them an explanatory message with a link to a web page where they can get "authorized" and put on a "buddy list". On that page you could do a variety of things, depending on whether you just wanted to screen out automated mailers or really wanted to pre-qualify the sender.
I'm skeptical about heuristic filters, because of the possibility of the occasional false positive, which could be an embarrasment (or worse).
However, the filtering technology is very much of interest to me, for other reasons... I will take a look at POPFile for sure.
Baysian filtering looks at word occurrence statistics. This is saying just compare the bulk redundancies of a message as compared to a collection of test messages of a known type, without even looking at the "words". May not be the ultimate filter (and I doubt it could be), but it's real interesting, I think, that this appears to have considerably greater than zero accuracy.
OTOH, it seems to me that some other model, such as a scheme that gives legitimate senders explicit advance AUTHORIZATION to send you email, might be what's needed. How to implement that is, well, left as "an exercise for the reader" -- actually, this has been discussed on/.
What this appears to be is XML so that the authorities can trade information they gather via intercept, much like businesses communicate with each other via XML. I suppose the idea is to get law enforcement people using a common markup convention, to get them all on the same page. Not a bad idea, it seems to me.
This has got to take off - the economics will be too compelling, especially when the use of Linux in the educational world develops critical mass.
So, I think there is a market here -- not a very high paying market, but a market - for people offering consulting and training on deploying Linux in a school environment. There are a lot of very cool things going on in that environment, and it is probably more friendly than most to the concept of open source.
Did you read THAT? Lessing was shooting for the votes of Rhenquist, Scalia, Thomas, Kennedy and O'Connor on some "limited federal government" bullshit ("Enumerated Powers")! Give me a break!
Does this man really believe that these justices are driven by pure principle? Did he ever read Bush v. Gore? Remember, these are the same justices that bent over backward to overrule a State supreme court on "Equal Protection" grounds - something only liberal judges ever paid much attention to (unless the issue is reverse discrimination).
Get real, folks. Abstract principles usually bend at the end of the day in the face of politics and economics.
And besides, as Lessig himself acknowledges, under the better principle of constitutional law, he SHOULD have lost this case anyway.
One thing for sure, is that I don't want anyone this naive as MY lawyer, and neither should you.
It was unfortunate indeed that this case was brought as it was in the first place, giving the Supreme Court the opportunity to cement this terrible policy into the fabric of constitutional law.
An in addition there are horrible statements in the majority opinion as to what the purpose of copyright protection is (essentially to make money for copyright owners).
A total disaster from beginning to end.
This stuff needs to be better thought through in advance by people with more sense, or we'll end up with no rights at all.
Strategically embedded print statements - the time-honored way to find bugs.
That plus printing out and actually READING the code. Amazing what you can find by walking out of range of the screen and keyboard and just reading through the stuff.
There's nothing to be afraid of. Our family does not have a copy of Windows running anywhere in the house.
We have a Linksys cable/firewall/wireless access point box, and three machines (or so) with ethernet adapters, all running Debian GNU/Linux.
The server is an old Celeron machine with 256 megs of RAM and an 80 gig drive, and a small old monitor that stays off most of the time. It has a big shared area on its disk, made available to all machines on the network via NFS, for "homework" and personal folders for all members of the family, as well as shared music and photo subdirectories. The server also runs a printer that's shared over the network via CUPS. The server runs Debian "stable" and has been, well, ultra stable.
Each of the workstations mount the server's NFS under a local directory called "server", and treats it as if it were the workstation's own local directory. The word processing software on each workstation (OpenOffice) is set up to point to the appropriate subdir on the server as its default directory. So, everyone's work files reside on the server, where they can be taken care of and backed up.
The workstations are more powerful Athlon and Duron machines, and run the Debian "testing" and "unstable" variants ("unstable" is in fact reasonably stable).
This system does not have all the latest games, but it has a lot of entertaining stuff (esp. from the Debian Jr. project), and it is more than adequate to handle all school-level computing tasks.
Snazzier peripherals like a scanner, CD Writer and Digital Camera are attached to the workstations.
This is a quibble, but in such a case, all you would have is a contract argument, not a copyright infringement case, which is you'd have if you owned a copyright and then licensed under the GPL. I have my doubts whether the GPL could be enforced by someone who was not the copyright owner. It's possible, but in my view problematic.
But it's all pretty theoretical, as any developer using such code would be virtually certain to substantially modify it.
Somebody already suggested this, I think, but the government itself could dual license the code, which would avoid my quibble. I have a funny feeling, though, that this won't be happenming any time soon, in this country.
I was addressing the implication of the first comment that someone could simply slap on a GPL without changing the software. I don't think that would work, for the reason I stated.
Anyway, anyone having any doubts would just take the government's (identical) original pristine distribution under the BSD-like license and not be bothered with the later-applied GPL.
Of course, if the relicensor makes substantial modifications to the software, THOSE may well be released under the GPL. We're in complete agreement there. The re-releaser could apply the GPL on the entire modified program (though doing so would control only the portion contributed by the re-releaser).
In the real world, what the government makes available will almost always be subject to heavy additional development, so the issue of re-licensing is a very pertinent one.
On the larger question of what the government should do, I think it's a close call, but probably a BSD-like approach would have an edge. People who adapt the government product could go either proprietary or GPL, stick with BSD or do whatever else they wanted.
Of course, the government contractor could negotiate to keep the work under the GPL, in which case I would think that's how it would go. Or, of course, the contractor could also negotiate to keep the software proprietary and just give the goverment a limited license (which now happens a lot).
But perhaps the presumption ought to be in favor of a BSD-like approach if the software is going to be created substantially on the government's nickel.
You can't "license" rights that you don't have. The person who takes under a BSD and re-releases under a GPL does not own the copyright - she is not the author, it is not her original work. The only thing the second person can restrict with the GPL is that which she owns, i.e., her own ADDITIONS and CHANGES to the work (if any).
As the poster above pointed out, the government is generally IMMUNE from these kinds of suits.
But it IS a time-honored sport to challenge the validity of patents (in a lawsuit against the patent owner) and then seek to take depositions of the patent examiners who were involved in the grant of the patent. Though the Patent Office will try to limit the examiners' testimony to essentially zero, it still REALLY tweaks them no end to have their examiners called to the witness stand.
This patent, by the way, looks like a complete piece of garbage that had claims tacked onto it years after the fact. These later-thought-up claims had virtually no relationship that I can see to what was actually disclosed when the thing was originally filed.
Debian has some distro-specific tools and scripts for doing things like compiling and installing the kernel, modules, and other source packages. They are easy to use (once you read the FAQ), and work very smoothly. To get the most mileage and have the fewest problems, you use these tools instead of the usual make config, make and make install. It helps keep everything from becoming unhinged. You just have to learn a few new tricks, that's all.
Of course, sticking within the distribution proper sure does work if you don't want to have to think too much. There are about 10,000 packages provided. Look here for a list.
Hope this isn't redundant, but its worth pointing out that Debian goes out of its way to stay free. And it is rock solid stable. These two things mean a lot if you're going to use Linux long term. The "free" part, apart from philosophical issues, means you won't get hit in the future by some software vendor with its hand out demanding to get paid for an "upgrade" of formerly "free" software (this happened too many times with other distros). The stable part means, quite simply, that you can get work done. Debian does not have a monopoly on stability, but it is very stable, especially after being upgraded over a period of time (it STAYS stable).
I don't know why everyone whines about the install. The install isn't bad if you've installed a few distros before and accept most of the defaults. Oh, yes, be sure to select the 2.4 kernel flavor of installation and a journaling file system (e.g., EXT3). Anyway, they're revamping the install. If it's too much for you, use something else.
Red Hat is OK, but I was burned one too many times with RPM dependency conflicts. This kind of thing is very rare in Debian, if you take care to maintain your system "the Debian way."
Yes, I'd like to have xfree 4.2, KDE 3, Gnome 2 and the other latest stuff, and they're all available for Debian if you want to install experimental and unstable packages, but I don't, at least not on a production system. There's nothing missing from the stable and testing distributions that keep me from doing most of what I want to do.
Guess I sound a bit like a true believer, but damn, I like being able to turn off my entire network, say for a weekend out of town, and then turn it on and have every machine come up the way it's supposed to with no fooling around. And know that the whole thing will remain free for the foreseeable future.
Slashdot Mirror
You can look these registrations up on the U.S. Copyright Office web site, http://www.loc.gov/copyright/
I saw only 19 registrations, but most of the "biggies" (GNU C, Emacs, GNU tar, diff, make, m4, etc.) are covered. These are existing registrations. There are potentially serious consequences for infringing these - even with no "actual" damages. Plus FSF could register more of their copyrights any time they wanted.
This looks pretty serious to me. I don't see why there's any doubt about FSF's ability to enforce this. My guess is that they'd rather settle it without having to spend any real money. Even with volunteer lawyers, lawsuits can cost a lot.
The guy in the paper needed somthing to write about. This will quietly settle and it will not be big news.
I can only think of a few crimes that merit the death penalty, and spamming is one of them.
Other crimes that definitely merit death include:
-- Serving popup ads
-- Peeing in the alley behind my apartment building (dogs get clemency)
-- Sending emails without subject lines
true... i think its mainly because microshaft is increasing requirements to pump the market, and then FSF things like KDE and Gnome have to do the same for end-user appeal, and to not lose ground to it.
kudos to intel for actually testing their chip... does anyone remember the pentium pro?
Entire campus a Microsoft free zone? Brilliant! Plenty of people would eagerly contribute to that, I sure would.
Linux not being ready for the desktop is pure BS. Some big Linux promoter like IBM would probably be HAPPY to help with a project like this. People would turn out from all over to assist in such an effort.
The big downside of Linux today (and the only one, really) is rapid interchange of documents with MS Office users. This is an artifical barrier that makes it impractical for most businesses to consider Linux for the end user. However, the issue is far less relevant to a campus community which can largely do whatever it damn well wants. In a college/university setting you're probably not sitting there all day getting business-critical MS Office attachments that you have to bat back instantaneously with comments and revisions. OpenOffice does work extremely well with its own documents. It's just the interchange with MS Office that's a little rocky.
You need a little IMAGINATION folks, not petty bean counting bullshit. No, the $2.4 million does not in itself economically justify this, but it is sure a hell of a nice seed to start something with VASTLY more economic, PR and educational potential.
(The again, maybe the lead post WAS a TROLL...)
According to one of the articles,
"He was courted by the likes of Sega and Nintendo, with one games publisher reportedly offering him a Porsche just for meeting with the company."
This sounds pretty good. Anyone know the story on this?
This type of development requires lots of up front money, with only a statistical likelihood of success. If you want this money to come from investors, they need to see a pot of gold potentially at the end. The only alternative is for the money to come from the government or charitable foundations. Actually, in the real world, funding is from a mix of these sources. However, if you cut back patenting as we now know it, you have to push the funding much more toward public sources, which has its own set of problems.
I don't think many people are ready to completely socialize health-care related R&D. What a scary thought.
Remember WordPerfect, where you could press F3 to "Reveal Codes"? Being an MS Word snob from way back, I used to think that style sheets were much more clever than embedding tags into the document.
So they're going back to tag markup. Probably for the better, and certainly much more open, but it seems ironic that their second screenshot looks a WHOLE LOT like a WordPerfect "Reveal Codes" display.
BTW, I don't think a lot of end users will have ANY concept whatsoever of what an "XML schema" is... Good luck!
Nope. Section 904 of the Manual of Patent Examining Procedure says in very clear terms that they're SUPPOSED to search nonpatent literature:
"The examiner, after having obtained a thorough understanding of the invention disclosed and claimed in the nonprovisional application, then searches the prior art as disclosed in patents and other published documents, i.e., nonpatent literature (NPL)."
And the good ones do just that. I've seen it many times.
Also, slight correction to what I said in my first post above - not ALL applications get published, because it is possible to opt out by giving up most foreign patent rights. Few applicants do this, though, with the result that most applications now are being published.
Geez, why don't the people here learn something about what they're talking about before posting?
Except for some details on the billing, what the poster described is awfully close to the current process! The current BROKEN process.
The problem is that the patent office sometimes does an incompetent search. They get paid for it too (several hundred dollars, actually).
In this case, for example, as is evident from the front page of the patent, the only thing the patent examiner ever looked at to check the patentability of this application was other issued patents and one trade press article from 1996. He/she obviously never heard of CVS, didn't know squat about versioning systems, and just rushed the thing out. The system isn't supposed to work this way, but this kind of thing keeps happening.
Also, the patent applicant is obligated to cite to the examiner everything he/she knows about that might be germane to patentability. So you have to wonder why THEY didn't cite CVS.
No, this will NEVER improve until there is more of a check on the process. But the current process is one-sided - all between the applicant and a government agency, and neither has much incentive to do a thorough search.
So, this situation is unlikely to improve until the process of obtaining a patent is a allowed to be contested procedure - contested by the people who are likely to be impacted by the patent, giving them an opportunity to see a draft of it and object before it has been issued, and to have a real no-holds-barred contested proceeding before a competent panel of experts before the sucker is approved.
No this is not a new idea. It's called "pre-grant opposition" and it's been floated out many times, and always been shot down, either by the patent lawyers or by the "small inventor" lobby. It is the way things are done in Europe, but not here.
As it is, due to a realtively recent change in the law, new patents are now "published" on the Patent Office web site well before they are allowed.
What we need is for Slashdotters to look at THOSE publications and complain about them BEFORE they issue as patents. No formal opposition procedure now exists, but people can still make noise, and it might be helpful.
In this case there was no publication, because the patent application was filed before the publication law took effect. But that will change over time. The norm will be that every new patent will have been published for the world to see, well before it issues. It would be better if there were a formal channel to DO something about it, but we should at least be reading these publications, right now, instead of merely complaining about patents that have already issued.
Aha! Corporations' newly found fear of the securities lawa may be doing some good.
This is only a CYA "risk factor", of course, to protect MS from lawsuits based on hindsight. Nevertheless, it is still a telling concession.
But fundamentally, isn't it the case that MS really CAN'T compete on price without distoring their entire business model? Maybe they should state that as well!
You could put their mail in a putative spam folder and send them an explanatory message with a link to a web page where they can get "authorized" and put on a "buddy list". On that page you could do a variety of things, depending on whether you just wanted to screen out automated mailers or really wanted to pre-qualify the sender.
I'm skeptical about heuristic filters, because of the possibility of the occasional false positive, which could be an embarrasment (or worse).
However, the filtering technology is very much of interest to me, for other reasons... I will take a look at POPFile for sure.
Baysian filtering looks at word occurrence statistics. This is saying just compare the bulk redundancies of a message as compared to a collection of test messages of a known type, without even looking at the "words". May not be the ultimate filter (and I doubt it could be), but it's real interesting, I think, that this appears to have considerably greater than zero accuracy.
/.
OTOH, it seems to me that some other model, such as a scheme that gives legitimate senders explicit advance AUTHORIZATION to send you email, might be what's needed. How to implement that is, well, left as "an exercise for the reader" -- actually, this has been discussed on
What this appears to be is XML so that the authorities can trade information they gather via intercept, much like businesses communicate with each other via XML. I suppose the idea is to get law enforcement people using a common markup convention, to get them all on the same page. Not a bad idea, it seems to me.
This has got to take off - the economics will be too compelling, especially when the use of Linux in the educational world develops critical mass.
So, I think there is a market here -- not a very high paying market, but a market - for people offering consulting and training on deploying Linux in a school environment. There are a lot of very cool things going on in that environment, and it is probably more friendly than most to the concept of open source.
Did you read THAT? Lessing was shooting for the votes of Rhenquist, Scalia, Thomas, Kennedy and O'Connor on some "limited federal government" bullshit ("Enumerated Powers")! Give me a break!
Does this man really believe that these justices are driven by pure principle? Did he ever read Bush v. Gore? Remember, these are the same justices that bent over backward to overrule a State supreme court on "Equal Protection" grounds - something only liberal judges ever paid much attention to (unless the issue is reverse discrimination).
Get real, folks. Abstract principles usually bend at the end of the day in the face of politics and economics.
And besides, as Lessig himself acknowledges, under the better principle of constitutional law, he SHOULD have lost this case anyway.
One thing for sure, is that I don't want anyone this naive as MY lawyer, and neither should you.
It was unfortunate indeed that this case was brought as it was in the first place, giving the Supreme Court the opportunity to cement this terrible policy into the fabric of constitutional law.
An in addition there are horrible statements in the majority opinion as to what the purpose of copyright protection is (essentially to make money for copyright owners).
A total disaster from beginning to end.
This stuff needs to be better thought through in advance by people with more sense, or we'll end up with no rights at all.
Strategically embedded print statements - the time-honored way to find bugs.
That plus printing out and actually READING the code. Amazing what you can find by walking out of range of the screen and keyboard and just reading through the stuff.
There's nothing to be afraid of. Our family does not have a copy of Windows running anywhere in the house.
We have a Linksys cable/firewall/wireless access point box, and three machines (or so) with ethernet adapters, all running Debian GNU/Linux.
The server is an old Celeron machine with 256 megs of RAM and an 80 gig drive, and a small old monitor that stays off most of the time. It has a big shared area on its disk, made available to all machines on the network via NFS, for "homework" and personal folders for all members of the family, as well as shared music and photo subdirectories. The server also runs a printer that's shared over the network via CUPS. The server runs Debian "stable" and has been, well, ultra stable.
Each of the workstations mount the server's NFS under a local directory called "server", and treats it as if it were the workstation's own local directory. The word processing software on each workstation (OpenOffice) is set up to point to the appropriate subdir on the server as its default directory. So, everyone's work files reside on the server, where they can be taken care of and backed up.
The workstations are more powerful Athlon and Duron machines, and run the Debian "testing" and "unstable" variants ("unstable" is in fact reasonably stable).
This system does not have all the latest games, but it has a lot of entertaining stuff (esp. from the Debian Jr. project), and it is more than adequate to handle all school-level computing tasks.
Snazzier peripherals like a scanner, CD Writer and Digital Camera are attached to the workstations.
You can click here to see our setup.
This is a quibble, but in such a case, all you would have is a contract argument, not a copyright infringement case, which is you'd have if you owned a copyright and then licensed under the GPL. I have my doubts whether the GPL could be enforced by someone who was not the copyright owner. It's possible, but in my view problematic.
But it's all pretty theoretical, as any developer using such code would be virtually certain to substantially modify it.
Somebody already suggested this, I think, but the government itself could dual license the code, which would avoid my quibble. I have a funny feeling, though, that this won't be happenming any time soon, in this country.
I was addressing the implication of the first comment that someone could simply slap on a GPL without changing the software. I don't think that would work, for the reason I stated.
Anyway, anyone having any doubts would just take the government's (identical) original pristine distribution under the BSD-like license and not be bothered with the later-applied GPL.
Of course, if the relicensor makes substantial modifications to the software, THOSE may well be released under the GPL. We're in complete agreement there. The re-releaser could apply the GPL on the entire modified program (though doing so would control only the portion contributed by the re-releaser).
In the real world, what the government makes available will almost always be subject to heavy additional development, so the issue of re-licensing is a very pertinent one.
On the larger question of what the government should do, I think it's a close call, but probably a BSD-like approach would have an edge. People who adapt the government product could go either proprietary or GPL, stick with BSD or do whatever else they wanted.
Of course, the government contractor could negotiate to keep the work under the GPL, in which case I would think that's how it would go. Or, of course, the contractor could also negotiate to keep the software proprietary and just give the goverment a limited license (which now happens a lot).
But perhaps the presumption ought to be in favor of a BSD-like approach if the software is going to be created substantially on the government's nickel.
You can't "license" rights that you don't have. The person who takes under a BSD and re-releases under a GPL does not own the copyright - she is not the author, it is not her original work. The only thing the second person can restrict with the GPL is that which she owns, i.e., her own ADDITIONS and CHANGES to the work (if any).
As the poster above pointed out, the government is generally IMMUNE from these kinds of suits.
But it IS a time-honored sport to challenge the validity of patents (in a lawsuit against the patent owner) and then seek to take depositions of the patent examiners who were involved in the grant of the patent. Though the Patent Office will try to limit the examiners' testimony to essentially zero, it still REALLY tweaks them no end to have their examiners called to the witness stand.
This patent, by the way, looks like a complete piece of garbage that had claims tacked onto it years after the fact. These later-thought-up claims had virtually no relationship that I can see to what was actually disclosed when the thing was originally filed.
It's an ABOMINATION.
Debian has some distro-specific tools and scripts for doing things like compiling and installing the kernel, modules, and other source packages. They are easy to use (once you read the FAQ), and work very smoothly. To get the most mileage and have the fewest problems, you use these tools instead of the usual make config, make and make install. It helps keep everything from becoming unhinged. You just have to learn a few new tricks, that's all.
Of course, sticking within the distribution proper sure does work if you don't want to have to think too much. There are about 10,000 packages provided. Look here for a list.
Hope this isn't redundant, but its worth pointing out that Debian goes out of its way to stay free. And it is rock solid stable. These two things mean a lot if you're going to use Linux long term. The "free" part, apart from philosophical issues, means you won't get hit in the future by some software vendor with its hand out demanding to get paid for an "upgrade" of formerly "free" software (this happened too many times with other distros). The stable part means, quite simply, that you can get work done. Debian does not have a monopoly on stability, but it is very stable, especially after being upgraded over a period of time (it STAYS stable).
I don't know why everyone whines about the install. The install isn't bad if you've installed a few distros before and accept most of the defaults. Oh, yes, be sure to select the 2.4 kernel flavor of installation and a journaling file system (e.g., EXT3). Anyway, they're revamping the install. If it's too much for you, use something else.
Red Hat is OK, but I was burned one too many times with RPM dependency conflicts. This kind of thing is very rare in Debian, if you take care to maintain your system "the Debian way."
Yes, I'd like to have xfree 4.2, KDE 3, Gnome 2 and the other latest stuff, and they're all available for Debian if you want to install experimental and unstable packages, but I don't, at least not on a production system. There's nothing missing from the stable and testing distributions that keep me from doing most of what I want to do.
Guess I sound a bit like a true believer, but damn, I like being able to turn off my entire network, say for a weekend out of town, and then turn it on and have every machine come up the way it's supposed to with no fooling around. And know that the whole thing will remain free for the foreseeable future.