Burning music CDs
in [UK Linux]
|
From: unruh on 9 Mar 2010 14:21 On 2010-03-09, Richard Kettlewell <rjk(a)greenend.org.uk> wrote: > unruh <unruh(a)wormhole.physics.ubc.ca> writes: >> On 2010-03-08, Paul Martin <pm(a)nowster.org.uk> wrote: >>> unruh wrote: > >>>> What I see is unreasonableness on all sides. I find Debian's stance to >>>> be equally unreasonable, long after 2006. >>> >>> The GPL and CDDL are incompatible. This is confirmed by the author of >>> the CDDL. Joerg doesn't agree. >> >> Oh dear. And you complain about Joerg. There is an ancient proverb >> having to do with motes and beams that you should perhaps read. >> As I said I find the stance of both to be rediculous, and childish. >> (PS, your claim re the CDDL is silly. The legal status of that license >> is not determined by rumours about what the supposed author claimed or >> did not claim, but by the language. > > So look at the language, neither document is hard to find. Nor is > finding an incompatibility. e.g. CDDL s3.1 insists you include a copy > of the CDDL with the work; GPL2 s6 insists you don't add any > restrictions beyond those in the GPL. Neither are unreasonable given > what they're trying to achieve, but you can't satisfy both at once. What the hell are you talking about? GPL also states that you need to make available a copy of the GPL. But they ARE different licenses. Noone is arguing that all works have to satisfy ALL possible licenses at once. One of the key issues under copyright law is whether or not linking with a work makes it a "derivative work". That is a (vague) term under copyright law, a definition which lies at the heart of the SCO claims against Linux. They take an extremely broad definition of what works are derivative, as do some of the anti-Joerg proponents. There is an argument by a lawyer versed in copyright law that says linking does NOT make a work a derivative work.Since the restrictions in the GPL obtain their force only through copyright law ( they are explicitly NOT a contract), the definitions of copyright law apply. I believe that a finding by any court that any linking made anything a derivative work would be a disaster for all software. > >> It is certainly less incompatible that GPL3 is with GPL2, which WERE >> designed to be incompatible, and are. Has Debian stated that they >> refuse to include any GPL3 programs since the kernel is GPL2 and Linux >> has stated he will never release it under GPL3? > > Programs aren't linked with the kernel, so licence compatibility > problems don't arise. The kernel source has explicitly stated this for > many years. Download it and check if you don't believe me. What has this to do with anything? > >>>> Yes, Joerg is difficult to get along with. But you are not marrying >>>> him. You are providing software. And Debian is screwing its >>>> customers by their position on the inclusion of cdrecord, and it >>>> seems simply out of pique. >>> >>> The GPL and CDDL are incompatible. This is confirmed by the author of >>> the CDDL. Joerg doesn't agree. Therefore nobody has a usable licence >>> to distribute cdrecord code produced by Joerg after May 2006. >> >> Horseshit. The CDDL IS a license. > > The relevant point is that cdrecord is a mixture of both the CDDL and > the GPL. Again, download it and check if you don't believe me. So the > incompatibility matters. So what? Some of the programs are GPL, some are CDDL. So? >
From: Mark Hobley on 9 Mar 2010 16:08 unruh <unruh(a)wormhole.physics.ubc.ca> wrote: > One of the key issues under copyright law is whether or not linking with > a work makes it a "derivative work". That is a (vague) term under > copyright law, a definition which lies at the heart of the SCO claims > against Linux. If you are distributing compiled binary applications build against the work, then some organizations may argue that the binary application is a derivative work. However, distributed source code, can be distributed without the work nd compiled against an alternative work (and hence is probably not a derivative because it contains no components of the work). Mark. -- Mark Hobley Linux User: #370818 http://markhobley.yi.org/
From: unruh on 9 Mar 2010 16:44 On 2010-03-09, Paul Martin <pm(a)nowster.org.uk> wrote: > In article <slrnhpd7u1.ttp.unruh(a)wormhole.physics.ubc.ca>, > unruh wrote: >> On 2010-03-09, Richard Kettlewell <rjk(a)greenend.org.uk> wrote: >>> unruh <unruh(a)wormhole.physics.ubc.ca> writes: >>>> On 2010-03-08, Paul Martin <pm(a)nowster.org.uk> wrote: >>>>> unruh wrote: >>> >>>>>> What I see is unreasonableness on all sides. I find Debian's stance to >>>>>> be equally unreasonable, long after 2006. >>>>> >>>>> The GPL and CDDL are incompatible. This is confirmed by the author of >>>>> the CDDL. Joerg doesn't agree. >>>> >>>> Oh dear. And you complain about Joerg. There is an ancient proverb >>>> having to do with motes and beams that you should perhaps read. >>>> As I said I find the stance of both to be rediculous, and childish. >>>> (PS, your claim re the CDDL is silly. The legal status of that license >>>> is not determined by rumours about what the supposed author claimed or >>>> did not claim, but by the language. >>> >>> So look at the language, neither document is hard to find. Nor is >>> finding an incompatibility. e.g. CDDL s3.1 insists you include a copy >>> of the CDDL with the work; GPL2 s6 insists you don't add any >>> restrictions beyond those in the GPL. Neither are unreasonable given >>> what they're trying to achieve, but you can't satisfy both at once. > >> What the hell are you talking about? GPL also states that you need to >> make available a copy of the GPL. But they ARE different licenses. >> Noone is arguing that all works have to satisfy ALL possible licenses at >> once. > > They do if there are two licences within a product. No. If they were liscences on the same program, then it would. But as everyone in Linux insists, the GPL explicitly allows "mere aggregation" ( a term which has no meaning withing copyright law, and thus has no meaning within a license under copyright law), which mkisofs and cdrecord are. > >> One of the key issues under copyright law is whether or not linking with >> a work makes it a "derivative work". That is a (vague) term under >> copyright law, a definition which lies at the heart of the SCO claims >> against Linux. > > SCO have no case, as well you know. Of course I do, which makes me bewildered when people in the Linux community make what looks like the same argument. > >> They take an extremely broad definition of what works are >> derivative, as do some of the anti-Joerg proponents. There is an >> argument by a lawyer versed in copyright law that says linking does NOT >> make a work a derivative work.Since the restrictions in the GPL obtain >> their force only through copyright law ( they are explicitly NOT a >> contract), the definitions of copyright law apply. I believe that a >> finding by any court that any linking made anything a derivative work >> would be a disaster for all software. > > Where's linking in all this? Why did you bring it up? Do you actually > understand the term? Do you? > >>>> It is certainly less incompatible that GPL3 is with GPL2, which WERE >>>> designed to be incompatible, and are. Has Debian stated that they >>>> refuse to include any GPL3 programs since the kernel is GPL2 and Linux >>>> has stated he will never release it under GPL3? >>> >>> Programs aren't linked with the kernel, so licence compatibility >>> problems don't arise. The kernel source has explicitly stated this for >>> many years. Download it and check if you don't believe me. > >> What has this to do with anything? > > It's fundamental to your understanding. Go and read the licences. > >>>> Horseshit. The CDDL IS a license. >>> >>> The relevant point is that cdrecord is a mixture of both the CDDL and >>> the GPL. Again, download it and check if you don't believe me. So the >>> incompatibility matters. > >> So what? >> Some of the programs are GPL, some are CDDL. So? > > The GPLv2 requires that all of the package is distributed under the > GPL: No. The GPL cannot say anything about other works. They have no legal right to do so. The GPL can only talk about the work itself, and "derived works" But derived works is a term under copyright law, and its meaning is unclear. The law has been all over the place on it. SCO's case base based on a very liberal interpretation of "derived work" and interpretation which would almost certainly fail in court (it is disturbing that AFAIK the case is still dragging on). The question is as to how far the GPL can cast its "net" in applying its derived work doctine. > > "These requirements apply to the modified work as a whole. If > identifiable sections of that work are not derived from the Program, > and can be reasonably considered independent and separate works in > themselves, then this License, and its terms, do not apply to those > sections when you distribute them as separate works. But when you > distribute the same sections as part of a whole which is a work based > on the Program, the distribution of the whole must be on the terms of > this License, whose permissions for other licensees extend to the > entire whole, and thus to each and every part regardless of who wrote > it." This statement is all based on the definition of derived work which this section tries to capture by its use of the term-- work based on the Program" That is close to the definition of derived work under copyright law, and is the bone of contention. > > That's pretty much unambiguous. In fact it is legally extremely ambiguous, and may well fall outside copyright law. Since this is a license, anything which falls outside of copyright law is null and void. > > " 4. You may not copy, modify, sublicense, or distribute the Program > except as expressly provided under this License. Any attempt > otherwise to copy, modify, sublicense or distribute the Program is > void, and will automatically terminate your rights under this License. > However, parties who have received copies, or rights, from you under > this License will not have their licenses terminated so long as such > parties remain in full compliance." > > " 6. Each time you redistribute the Program (or any work based on the > Program), the recipient automatically receives a license from the > original licensor to copy, distribute or modify the Program subject to > these terms and conditions. You may not impose any further > restrictions on the recipients' exercise of the rights granted herein. > You are not responsible for enforcing compliance by third parties to > this License." > > These two clauses mean that only Joerg can distribute cdrecord (with > its twin licences), assuming all the GPL and CDDL code is SOLELY > his. If there's anyone else's GPL code in there, not even he has a > licence to redistribute under the GPL. ???? This is totally nuts. The author of the work has the right to allow anyone to distribute under any license he wishes. He has stated innumerable times that the work IS distrubtable. His statements in case of copyright law are determinative in the permissions they allow ( they may not be in the restrictions they place). He has stated that he has the permission of all of the license holders to distribute the works he does under the licenses he uses. If you have evidence to dispute that, I would suggest you display it. mkidofs is a separate program, and is under GPL. Almost everyting else is CDDL. Where exactly are you claiming that there is a conflict. >
From: unruh on 9 Mar 2010 16:55 On 2010-03-09, Mark Hobley <markhobley(a)hotpop.donottypethisbit.com> wrote: > unruh <unruh(a)wormhole.physics.ubc.ca> wrote: > >> One of the key issues under copyright law is whether or not linking with >> a work makes it a "derivative work". That is a (vague) term under >> copyright law, a definition which lies at the heart of the SCO claims >> against Linux. > > If you are distributing compiled binary applications build against the work, > then some organizations may argue that the binary application is a derivative > work. However, distributed source code, can be distributed without the work Some organizations claim that if you have ever been within 100 miles of the work, it is a derivative work. That does not make it law. If the library is a separate library, with a defined API a good case could be made that the result is not derivative, but is a separate work. AFAIK no court case has yet decided on the issue. http://www.rosenlaw.com/Rosen_Ch06.pdf gives an analysis of the GPL and the limitations to the interpretation of what a derivative work is. Now, since no court has ruled on the issue, this is not determative, but it is suggestive. > nd compiled against an alternative work (and hence is probably not a > derivative because it contains no components of the work). What "contains a component of the work" means is unclear. Does a bookstore contain "components" of all its books and thus the layout of the bookstore is copyright under all of the copyright liceses of all the books in the store? This is simply a warning that words like "contains" can be tricky and are not straightforward under copyright law. > > Mark. >
From: Mark Hobley on 9 Mar 2010 19:08
Paul Martin <pm(a)nowster.org.uk> wrote: > One licence applies to the program used to build cdrtools, which is > forms part of cdrtools, and a different licence applies to the source > code that's being build. However, one cannot separate the build tools > from the source code, as has Joerg objected to that. Can't one of us just knock up a script to replace the build tools, and then remove the original build tools from the package? Or is it not that easy? Mark. -- Mark Hobley Linux User: #370818 http://markhobley.yi.org/ |