jkp@cs.HUT.FI (Jyrki Kuoppala) (08/09/90)
In article <6--4A8C@xds13.ferranti.com>, peter@ficc (Peter da Silva) writes: >I wrote a soft shareware program for the Amiga called "browser". After >a while, I recieved mail from someone in Germany to the effect that someone >was selling a disk that contained several such programs for more than the >$5 that Fred Fish charges for his Amiga PD collection, with the implication >that I should be horrified at his abuse of my work. > >I'm afraid that I disappointed this worthy gentleman, because I really >didn't care. So long as the package stayed together so my begging letter >was included, I didn't care if they required a 5 year indenture to get my >code. I don't mind that situation either; what I _do_ mind if I post / publish some code to the public domain, then someone else comes and (perhaps improving the code somewhat, perhaps deleting my name, perhaps omiting documentation and source, perhaps doing something else) starts selling the code so that the buyer of that program isn't allowed to redistribute. I think Minix is a very good service to the community, but I see some of the above scene happening with Minix (or perhaps it's my misunderstanding). I think there's a lot of user-contribued stuff distributed with Minix. The authors have allowed free use of that stuff, fine. But I suppose all the software in Minix comes with Prentice-Hall copyright, so you aren't allowed to redistribute that free stuff, either, if you got it with Minix, right ? Please tell me I'm wrong. That's what I think is a very good aspect of the GNU copyright. It guarantees all the users / redistributers the right to get _everything_ of the original work easily and takes away the possibility of a third party taking this freedom away. For these reasons, if I'll make some sizable contributions to Minix (or any other effort-taking software development for that matter), I'll probably put them under the GNU copyright. This may mean that Prentice-Hall won't put them in the official distribution, but the improvements (if they're not too much dependent on Minix) are available for use by free software developers easily, ie. someone can port them to GNU OS or something else to be freely distributable. If I just give them to the public domain, it easily happens that someone takes the software and restricts it's redistribution (like happens currently with X, TeX, Scribe (if I remember right), Ingres, Berkeley version of Unix etc.) //Jyrki
meulenbr@cst.philips.nl (Frans Meulenbroeks) (08/09/90)
[lots trimmed] jkp@cs.HUT.FI (Jyrki Kuoppala) writes: >misunderstanding). I think there's a lot of user-contribued stuff >distributed with Minix. The authors have allowed free use of that >stuff, fine. But I suppose all the software in Minix comes with >Prentice-Hall copyright, so you aren't allowed to redistribute that >free stuff, either, if you got it with Minix, right ? Please tell me >I'm wrong. Hmm. I'm not a lawyer, but I think PH only holds copyright to part of the sources, and a compilation copyright to the rest of the stuff. If you contribute a program with a copyright notice allowing unlimited distribution, PH will surely respect it, and people can take it off the MINIX disks without any problems (my opinion/interpretation). However, if the wording is very strictly and there is a possible implication that it also has effect on the other MINIX stuff it is probably rejected. The stuff on the MINIX disks is the same stuff that is posted by ast. Part of the utilities carries copyright notices of other people or institutions. For instance look at the code or more(1). I think that you can take the more sources from the MINIX disks, and do with it whatever you want as long as you comply with the copyright notice in more.c Sometimes I speak for me. Sometimes not. I definitely do not speak for Philips, PH, Free University, ast, anyone else. -- Frans Meulenbroeks (meulenbr@cst.philips.nl) Centre for Software Technology ( or try: ...!mcsun!phigate!prle!cst!meulenbr)
ast@cs.vu.nl (Andy Tanenbaum) (08/09/90)
In article <meulenbr.650184899@cst> meulenbr@cst.philips.nl (Frans Meulenbroeks) writes: >Hmm. I'm not a lawyer, but I think PH only holds copyright to part of >the sources, and a compilation copyright to the rest of the stuff. >If you contribute a program with a copyright notice allowing unlimited >distribution, PH will surely respect it, and people can take it off >the MINIX disks without any problems This is correct. If you post a program and put in an explicit message that it is in the public domain, then no one can get it out of the public domain. Anyone can modify it (even a single character) and then copyright the modified version and prevent anyone from using THAT version, but the original remains in the public domain. Once pubic domain, always public domain. I would suggest that people who post software and intend it to be in the public domain, put in a notice to that effect. That way people who want to sell it (e.g. P-H) can do so, but they cannot prevent anyone else from doing what they want, including giving it away free. The GNU copyleft, which creates legal obligations on the part of anyone selling it, causes lawyers to go into infinite loops. Statements that something is simultaneously copyright and public domain (see du.c) make as much sense as the law passed by the Indiana State Legislature around 1890 saying that in Indiana pi was legally exactly 3.0. Andy Tanenbaum (ast@cs.vu.nl)
peter@ficc.ferranti.com (Peter da Silva) (08/09/90)
In article <1990Aug8.173146.1206@santra.uucp> jkp@cs.HUT.FI (Jyrki Kuoppala) writes: > I don't mind that situation either; what I _do_ mind if I post / > publish some code to the public domain, then someone else comes and > (perhaps improving the code somewhat, perhaps deleting my name, > perhaps omiting documentation and source, perhaps doing something > else) starts selling the code so that the buyer of that program isn't > allowed to redistribute. I don't mind that either. The choice is this guy is going to use my code and be a black hole, or he's not going to use it, then I let him use it. There are people I know are using bits of my code in commercial programs. More power to them. The total intelligence of the world still rises, and nobody actually loses. And if he makes significant improvements to my code (which wasn't of commercial quality in the first place, most likely) then who am I to deny *him* the fruits of his labors? > For these reasons, if I'll make some sizable contributions to Minix > (or any other effort-taking software development for that matter), > I'll probably put them under the GNU copyright. This will pretty much guarantee that I won't use them. They will also not get into the X distribution, if they're appropriate for that (David Betz had to modify the copyright on XLISP for the Winterp distribution). There are other channels that will close off to you if you do that. You have to consider what your goals are. If it's to maximise the distribution of your software, something like the Berkeley copyright is better. > I just give them to the public domain, it easily happens that someone > takes the software and restricts it's redistribution (like happens > currently with X, X is freely available to anyone who wants it. Unenhanced X. If that's what you want, go for it. > TeX, I believe this, too, is available freely. It's just commercial quality versions, nicely shrink-wrapped and with support, that don't come for free. > [...] Berkeley version of Unix etc.) Berkeley UNIX was never "free", but they're working hard to make it so. -- Peter da Silva. `-_-' +1 713 274 5180. 'U` <peter@ficc.ferranti.com>
jla@wheaties.ai.mit.edu (Joseph Arceneaux) (08/10/90)
In article <+:258U1@xds13.ferranti.com> peter@ficc.ferranti.com (Peter da Silva) writes: >> I don't mind that situation either; what I _do_ mind if I post / >> publish some code to the public domain, then someone else comes and >> (perhaps improving the code somewhat, perhaps deleting my name, >> perhaps omiting documentation and source, perhaps doing something >> else) starts selling the code so that the buyer of that program isn't >> allowed to redistribute. >I don't mind that either. The choice is this guy is going to use my code >and be a black hole, or he's not going to use it, then I let him use it. >There are people I know are using bits of my code in commercial programs. >More power to them. The total intelligence of the world still rises, and >nobody actually loses. No, the people who don't get the code lose. If someone makes interesting improvements and folks start buying that version, sans source, then you have effectively wiped out the `intelligence' of a block of users. Your choice is really whether or not you want to promote the spirit of innovation and the freedom to make contributions, in which your your code was released in the first place. >And if he makes significant improvements to my code (which wasn't of >commercial quality in the first place, most likely) then who am I to deny >*him* the fruits of his labors? But the reason this person could make significant improvements is that he had the code. So if that's a good thing (which it is), then an even better thing would be for all the users of that improved version to have the chance to make the thing even better. >> For these reasons, if I'll make some sizable contributions to Minix >> (or any other effort-taking software development for that matter), >> I'll probably put them under the GNU copyright. >This will pretty much guarantee that I won't use them. It's too bad that you choose to restrict the rights, and intelligence as you put it, of programmers/users in this fashion. I have even heard of researchers who can't use GNU tools because their sponsors worry about being able to someday sell the [conceivable] product. Restricting research in this fashion, for hopes of eventual financial gain, is most reprehensible. >I believe this, too, is available freely. It's just commercial quality >versions, nicely shrink-wrapped and with support, that don't come for >free. This is untrue. GNU Emacs has been ranked superior to various competitors by Un*x publications. The GNU C Compiler seems to generate better code than a variety of commercial compilers. As for support, I have frequently seen 4 or 5 responses, in the same day, to questions posed on the emacs mailing list. This seems to be far better support than is available for most commercial products. Note that this level of support would not be available unless the code was free. >Berkeley UNIX was never "free", but they're working hard to make it so. Code is not `free' if users have to pay exhorbitant license fees to obtain it. Code is Free if everyone who desires it has easy access. So if someone takes Public Domain <whatever>, modifies it and then requires big bucks for their newly copyrighted version, that code is not free. Copyleft is the only instrument which really guarantees the freedom of a program.
henry@zoo.toronto.edu (Henry Spencer) (08/11/90)
In article <7268@star.cs.vu.nl> ast@cs.vu.nl (Andy Tanenbaum) writes: >I would suggest that people who post software and intend it to be in the >public domain, put in a notice to that effect... In general, this is not merely a good idea, it is necessary. In countries abiding by the Berne Convention -- which now includes the US -- it is not necessary to attach a copyright notice for copyright to be in effect. If you want something to be in the public domain, you *must* say so, in order to renounce the implicit copyright. Personally, I don't recommend putting things into PD. A nice permissive copyright notice like the Berkeley one or the C News one is better. That way, you can at least insist on getting credit for what you do. (I do not recommend the GNU copyleft, which is much more restrictive, unless you really support the Cult of Free Software. Many people find it legally difficult or impossible to use copylefted software.) -- It is not possible to both understand | Henry Spencer at U of Toronto Zoology and appreciate Intel CPUs. -D.Wolfskill| henry@zoo.toronto.edu utzoo!henry
jla@wheaties.ai.mit.edu (Joseph Arceneaux) (08/11/90)
In article <1990Aug10.170521.9435@zoo.toronto.edu> henry@zoo.toronto.edu (Henry Spencer) writes: > (I do >not recommend the GNU copyleft, which is much more restrictive, unless you >really support the Cult of Free Software. Many people find it legally >difficult or impossible to use copylefted software.) I recomend using Copyleft if you intend to make a contribution of your code to society. If your intent is indeed to benefit others with your program, then Copyleft will ensure that no one can eventually prohibit some group from thus benefitting.
henry@zoo.toronto.edu (Henry Spencer) (08/12/90)
In article <9849@galapas.ai.mit.edu> jla@ai.mit.edu (Joseph Arceneaux) writes: >> (I do >>not recommend the GNU copyleft, which is much more restrictive, unless you >>really support the Cult of Free Software. Many people find it legally >>difficult or impossible to use copylefted software.) > >I recomend using Copyleft if you intend to make a contribution of your >code to society. If your intent is indeed to benefit others with your >program, then Copyleft will ensure that no one can eventually prohibit >some group from thus benefitting. The question here is not whether you want to "benefit others", because releasing good software will do that regardless of which licensing policy you choose. The underlying issue is how concerned you are about others who do not share your altruistic views, and whether you feel so strongly about what they might do that you are willing to prohibit use of your software by people who won't or can't promise to be good. Not everyone who is constrained by his company's lawyers is a bad guy, and insisting on a promise of good behavior will deny the benefits of your software to many people. Assorted variants of my software, notably my regular-expression package, appear in a variety of commercial no-sources-available software products. This does not bother me; indeed, it pleases me, because the people who buy and use that software are getting a better product this way. -- It is not possible to both understand | Henry Spencer at U of Toronto Zoology and appreciate Intel CPUs. -D.Wolfskill| henry@zoo.toronto.edu utzoo!henry
jla@wheaties.ai.mit.edu (Joseph Arceneaux) (08/12/90)
In article <1990Aug11.234423.3307@zoo.toronto.edu> henry@zoo.toronto.edu (Henry Spencer) writes: >>I recomend using Copyleft if you intend to make a contribution of your >>code to society. If your intent is indeed to benefit others with your >>program, then Copyleft will ensure that no one can eventually prohibit >>some group from thus benefitting. >The question here is not whether you want to "benefit others", because >releasing good software will do that regardless of which licensing policy >you choose. The underlying issue is how concerned you are about others >who do not share your altruistic views, and whether you feel so strongly >about what they might do that you are willing to prohibit use of your >software by people who won't or can't promise to be good. If these folks want to use the software, there's no problem with Copyleft and they do indeed benefit; they are certainly not prohibited from doing so. If on the other hand their intention is to sell the stuff, the only `benefit' is the cash they make from witholding resources from their clients. These people have now prohibited their clients from the benefits (i.e, the source) which they (the so-called vendors) were free to obtain. >Not everyone >who is constrained by his company's lawyers is a bad guy, and insisting >on a promise of good behavior will deny the benefits of your software >to many people. This is true, and not everyone in South Africa believes in Apartheid. That doesn't mean that those citizens shouldn't speak their minds and try to change their government's policies, nor does it mean that the rest of the world should treat the country as though the problem didn't exist.
ast@cs.vu.nl (Andy Tanenbaum) (08/13/90)
In article <9849@galapas.ai.mit.edu> jla@ai.mit.edu (Joseph Arceneaux) writes: >I recomend using Copyleft if you intend to make a contribution of your >code to society. If your intent is indeed to benefit others with your >program, then Copyleft will ensure that no one can eventually prohibit >some group from thus benefitting. If you release your software into the public domain, nobody, not even the original author, can ever prevent anyone from using it in any way, ever. If that is your intention, then a statement simply putting the code into the public domain is enough. The GNU copyleft has a very serious problem of creating legal obligations on the part of anyone using it. I have very carefully avoided using any GNU software in MINIX as well as in Amoeba, even though technically I might have been willing. Although the intention of copyleft may have been to make sure the software was always available to the public, for me it has had just the opposite effect. I can live with public domain and I can live with copyrighted software plus a statement in the code or in a side letter granting permission to use it, but a legal, enforceable obligation to do ANYTHING is going to make any lawyer, including P-H's, sit up and take notice. What exactly, have they contracted to do, what have they gotten in return, what record keeping do they have to maintain to defend a lawsuit, how much in damages might they be liable for, and more. It is anything but simple. Andy Tanenbaum (ast@cs.vu.nl)
jla@wheaties.ai.mit.edu (Joseph Arceneaux) (08/13/90)
In article <7292@star.cs.vu.nl> ast@cs.vu.nl (Andy Tanenbaum) writes: >The GNU copyleft has a very serious problem of creating legal obligations >on the part of anyone using it. I have very carefully avoided using any >GNU software in MINIX as well as in Amoeba, even though technically I might >have been willing. Although the intention of copyleft may have been to >make sure the software was always available to the public, for me it has >had just the opposite effect. Since the effect of Copyleft is indeed to guarantee availability of the software, the only reason to avoid it is if you envision eventually restricting the software from the public. >but a legal, enforceable obligation >to do ANYTHING is going to make any lawyer, including P-H's, sit up and >take notice. What exactly, have they contracted to do, what have they >gotten in return, what record keeping do they have to maintain to defend >a lawsuit, how much in damages might they be liable for, and more. >It is anything but simple. Really, using Copylefted software is very simple--just use it. Only the folks who want to do other things, such as restrict its distribution or sell binary-only versions, need worry about legal obligations. Those, as you point out, are the people with all the lawyers.
jkp@cs.HUT.FI (Jyrki Kuoppala) (08/13/90)
In article <7268@star.cs.vu.nl>, ast@cs (Andy Tanenbaum) writes: >This is correct. If you post a program and put in an explicit message that >it is in the public domain, then no one can get it out of the public domain. >Anyone can modify it (even a single character) and then copyright the modified >version and prevent anyone from using THAT version, but the original remains >in the public domain. Once pubic domain, always public domain. I can uderstand this. However, I think some license agreements say that you are not allowed to redistribute the software you get. To get the software, you must sign the license agreement. So even if all of that software is in the public domain, you are not legally allowed to redistribute it to anyone. I may be wrong, but I think this was the case with the Berkely distribution (BSD) source license some time ago (I don't know if it still is). >The GNU copyleft, which >creates legal obligations on the part of anyone selling it, causes lawyers >to go into infinite loops. Yes, but that's what lawyers are supposed to do. I haven't found any other way to achieve the goal of keeping the software easily available to all in all cases. The Berkely copyright is not that good for my goals because it allows someone to take the code and redistribute only a binary of a modified version; I want the source to be always available to all users of the software, even improved versions (so everyone else can make improvements, too). It seems to be happening that lawyers are becoming more familiar with and start accepting the GNU copyleft. Many hardware and Unix vendors (Mt. Xinu for HP BSD, DG ?) use gcc as their primary or secondary compiler and I think even more include GNU Emacs in their OS distribution (DEC, Berkely, NeXT according to the latest GNU bulletin). I suppose this trend will continue as more GNU software will come to the market and more of the commercial world starts seeing the benefits. //Jyrki
jkp@cs.HUT.FI (Jyrki Kuoppala) (08/13/90)
In article <+:258U1@xds13.ferranti.com>, peter@ficc (Peter da Silva) writes: >I don't mind that either. The choice is this guy is going to use my code >and be a black hole, or he's not going to use it, then I let him use it. >There are people I know are using bits of my code in commercial programs. >More power to them. The total intelligence of the world still rises, and >nobody actually loses. My goals are different; to me it's more important that _all_ users can improve on the code and have the source easily available for other purposes as well (like hiring a third part to make improvements). To me, this is more important than as wide as possible use of the software. As somebody pointed out, the _user_ loses: (s)he can't hire a third party to make improvements (s)he wants because the source isn't available. With your method, the user is dependent on the one commercial vendor and if the vendor doesn't have good enought support then the user is in trouble. It's a bit like if the only one who can fix your car is the guy you bought he car from. With GNU copyleft, anyone skilled enough can fix problems and there's no monopoly for the vendor. >And if he makes significant improvements to my code (which wasn't of >commercial quality in the first place, most likely) then who am I to deny >*him* the fruits of his labors? But you _are_ willing to deny the fruits of his labors from everybody else ? I suppose next we come to the question of whether or not it is profitable to sell and support software with the GNU copyleft or is the more common 'software hoarding' the only way. Apparently many people think the GNU style is profitable. Cygnus Support sells and supports only GNU software and there are independent consultants which offer support for GNU software. Also some companies sell GNU software - one of these is Intel selling the 960 development kit, others include various normal retailers. Not to mention that most of FSF's income comes from selling GNU tapes and manuals. So you can let him and the rest of us, too, to have the fruits of his labors. >> For these reasons, if I'll make some sizable contributions to Minix >> (or any other effort-taking software development for that matter), >> I'll probably put them under the GNU copyright. > >This will pretty much guarantee that I won't use them. They will also >not get into the X distribution, if they're appropriate for that (David >Betz had to modify the copyright on XLISP for the Winterp distribution). >There are other channels that will close off to you if you do that. This might well be; however, I think more is gained by using the GNU copyright because that way even those people buying X for their PC's can get the source and improve on things. >You have to consider what your goals are. If it's to maximise the >distribution of your software, something like the Berkeley copyright >is better. See above. My goal is to make the software's usefulness maximum to everybody. The Berkeley copyright lacks in this. >> I just give them to the public domain, it easily happens that someone >> takes the software and restricts it's redistribution (like happens >> currently with X, > >X is freely available to anyone who wants it. Unenhanced X. If that's what >you want, go for it. Apparently I was a bit unclear; perhaps this messages will clear my intentions. What I meant was the following: I'm interested in getting X for my Amiga to use it as an X terminal for my pc532 board. OK - there is X available for the Amiga. It's a commercial product and costs somewhat. This is not such a big problem (although it _would_ be nice to get it for free - after all, the original X is free) - what is a big problem is that I'm planning to interface the Amiga and the pc532 with a SCSI interface. I don't think it's quite easy to get source to the Amiga X port. If X had the GNU copyleft I wouldn't have the problem. Of course the X consortium can put whatever copyright in its work it wants and I don't blame it for anything, but I want to make sure that people don't have similar problems with other software if I can do anything about it (like if I write some software). >> [...] Berkeley version of Unix etc.) > >Berkeley UNIX was never "free", but they're working hard to make it so. I meant the various utilities from Berkeley that are free and the usage of them by companies using Berkeley software in their OS like Sun etc. Same arguments as the X for Amiga apply. Also, the license agreement for the OS might state that I can't redistribute the software, even if it is declared free by Berkeley. //Jyrki
giguere@csg.uwaterloo.ca (Eric Giguere) (08/13/90)
In article <9861@galapas.ai.mit.edu> jla@galapas.ai.mit.edu (Joseph Arceneaux) writes: >In article <7292@star.cs.vu.nl> ast@cs.vu.nl (Andy Tanenbaum) writes: >>The GNU copyleft has a very serious problem of creating legal obligations >>on the part of anyone using it. I have very carefully avoided using any >>GNU software in MINIX as well as in Amoeba, even though technically I might >>have been willing. Although the intention of copyleft may have been to >>make sure the software was always available to the public, for me it has >>had just the opposite effect. > >Since the effect of Copyleft is indeed to guarantee availability of >the software, the only reason to avoid it is if you envision >eventually restricting the software from the public. Bingo! Well, we might as well all bow our heads in shame... just think, someday you or your employer might try to sell that software you've been working you butt off on... Boy, I can just see it now: "Jetson! What's the hold-up on the widget project?" "Well, Mr. Spacely, I'm afraid our lawyers have run into some trouble. Seems that this whizz-bang compiler we've been using actually prevents us from selling the software we wrote..." "What?!! Who's the idiot who used the compiler?" "Umm.... that would be me, Mr. Spacely... but before you fire me, I think I know a way out..." "This better be good..." "Well instead of selling the SOFTWARE, why don't we sell SUPPORT? If we just make a few changes we can make the software useless without the support. Actually we may even come out ahead... just think, instead of selling software once we can charge exorbitant yearly fees for support! What do you think...?" "Jetson, you're a genius. But don't tell anyone I said so, because I'll deny it. Now stop grovelling and get back to your desk!" ----> Reality check: Alright, perhaps I'm being a little sarcastic. Listen, I don't have anything against the GNU project. I admire their dedication and the products they've come up with. But the sad fact is that someday when I get a wife and kids I'll need money to feed them. I make that money by writing software because it's something I enjoy doing. My employer then sells that software to make the money that funds my paycheque. It's your usual mercantile system. The main point I want to make here is that in the Real World most software companies exist to sell software, not give it away. Because of this they can't use any GNU software in their endeavours, no matter how better it is than any other software. So all you GNU supporters, please don't get pissed off or turn self-righteous at the mere mention of making money off of software. Life's a bitch sometime. -- Eric Giguere giguere@csg.UWaterloo.CA
mitchell (Bill Mitchell) (09/05/90)
I know this horse has been beaten to death several times. Apologies for taking a couple of whacks at it myself. In article <1990Aug8.173146.1206@santra.uucp> jkp@cs.HUT.FI (Jyrki Kuoppala) writes: (stuff deleted) >what I _do_ mind if I post / >publish some code to the public domain, then someone else comes and >(perhaps improving the code somewhat, perhaps deleting my name, >perhaps omiting documentation and source, perhaps doing something >else) starts selling the code so that the buyer of that program isn't >allowed to redistribute. My understanding is that once it is in the public domain, it is public. Any member of the public then is free to do anything he/she chooses with it. He/she might improve it, delete your name, omit or add documentation, omit source, sell the code, use your source freely in his/her product, whatever. Once it's public, it's public. If you release a package with source to the public domain, and someone picks it up and starts selling a binary-only version of the same thing under his own name, that's allowed. His binary-only version is now in direct competition with your public domain version, which is available for free wherever it can be found. If he is to be able to sell any binary-only copies, he probably needs to open distribution channels to potential buyers who are either not reached by the public domain version or are not interested in the source. Fine. An argument might even be made that he has performed a service by opening new distribution channels and deserves to be compensated. Such an argument would, however, be immaterial; as would be the opinions of the argurer about what is or is not deserved. >I think Minix is a very good service to the community me too >, but I see some >of the above scene happening with Minix (or perhaps it's my >misunderstanding). It probably is. Code picked up from comp.os.minix has probably found its way into a number of items being marketed for profit. >I think there's a lot of user-contribued stuff >distributed with Minix. The authors have allowed free use of that >stuff, fine. Not just fine, great! wonderful! kudos and many thanks to those authors. >But I suppose all the software in Minix comes with >Prentice-Hall copyright, so you aren't allowed to redistribute that >free stuff, either, if you got it with Minix, right ? Please tell me >I'm wrong. My possibly flawed understanding is that if the code was released into the public domain by the original author, Prentice-Hall has the right to pick it up and re-release it with whatever restrictions they want to place on the copy they release. You are constrained by their restrictions in your use of a copy of their re-release. You are not so constrained if you obtain your copy from some other source. If you obtain a public domain copy, which has not had any restrictions added to it by whoever might have redistributed it, then you are free to do whatever you want to with it. You can redistribute it to the world for free without taking credit and without adding restrictions of your own. You can redistribute it and charge for it. You can take credit for it. You can place your own restrictions on redistribution of copies obtained from your redistribution. You can incorporate it into a commercial product without crediting the original author. If you choose, and you know who he/she is, you can give the original author credit. You can even pick up a public domain distribution, add the GNU copyright, and redistribute a release of it which has its use and redistribution restricted by the GNU copyright. Fine. >That's what I think is a very good aspect of the GNU copyright. It >guarantees all the users / redistributers the right to get >_everything_ of the original work easily and takes away the >possibility of a third party taking this freedom away. If the distribution were public domain, all users/redistributers already have the right to get everything of the original public domain distribution - if they can find a copy. As I understand it, the GNU copyright requires users/redistributers to make the original work available in its entirety and with no additional restrictions if they redistribute it or market anything including it. I'm not suprised that this requirement retards reuse/redistribution. >For these reasons, if I'll make some sizable contributions to Minix >(or any other effort-taking software development for that matter), >I'll probably put them under the GNU copyright. This may mean that >Prentice-Hall won't put them in the official distribution, but the >improvements (if they're not too much dependent on Minix) are >available for use by free software developers easily, ie. someone can >port them to GNU OS or something else to be freely distributable. If >I just give them to the public domain, it easily happens that someone >takes the software and restricts it's redistribution (like happens >currently with X, TeX, Scribe (if I remember right), Ingres, Berkeley >version of Unix etc.) If you release something restricted by the GNU copyright, why not release it into the public domain, unrestricted, also? Who is hurt? The GNU copyrighted version is still out there and available through whatever redistribution channels it finds, as are any copies of the public domain release which might be passed along without having copyright restrictions added to them. 'nuff said. More than enough, in fact.