On Wed, 2002-11-20 at 18:20, Jakob Bohm wrote: > On Wed, Nov 20, 2002 at 01:04:22PM -0500, David Turner wrote: > > On Mon, 2002-11-18 at 17:45, Jakob Bohm wrote: > > > On Mon, Nov 18, 2002 at 12:54:27PM -0500, David Turner wrote: > > > > On Mon, 2002-11-18 at 10:08, Henning Makholm wrote: > > > > > > > > > > All portions of governed files not labeled otherwise are owned by > > > > > > Hans > > > > > > Reiser, and by adding your code to it, widely distributing it to > > > > > > others or sending us a patch, and leaving the sentence in stating > > > > > > that > > > > > > licensing is governed by the statement in this file, you accept > > > > > > this. > > > > > > > > > > Adding things to the files without due notice of the change is > > > > > *forbidden* by the GPL. Essentially this notice seems to say that you > > > > > get the *additional* right to do such additions if you transfer your > > > > > copyright to Hans Reiser. Formally that amounts to a dual-licensing > > > > > scheme which is fine by the DFSG as long as one of the alternatives > > > > > (i.e. GPL) is free. > > > > > > > > This is somewhat bogus. Reiser demands a specific form for the change > > > > notices required by GPL (2)(a) -- you have to remove a sentence. And > > > > this is certainly a requirement in addition to the GPL, which conflicts > > > > with section (6). > > > > > > > > Fortunately, this doesn't actually work in the US. You can't transfer > > > > copyright implicitly in the US -- you need signed paperwork (17 USC 204 > > > > (a)). > > > > > > > > > > I respectfully thing you are reading the license all wrong here. > Oops, I meant think not thing ;-) > > > > > > IANAL, TINLA, IANADD, but here is my reading: > Still applies, > > > > > > The files are available under GPL, with no limitations. > > > > That's where we disagree. See below. > > > > > ... Eliding myself > > > > > > So far this should not be a problem. > > > > > > The Author claims, that if "You" modify a file without deleting > > > the reference to the extended license and replacing it with pure > > > GPL (or whatever), your changes become subject to the extended > > > license which grants the Author the permission to make the code > > > available under any license he chooses. > > > > That's not quite correct. He claims that he "owns" it. The only > > possible meaning of that is a transfer of copyright. It turns out that > > this doesn't actually work, but it's an attempt. > > > > > This is slightly more controversial, but essentially it works > > > the same way as LGPL->GPL conversion: If you modify an LGPL file > > > without changing the LGPL statement, your are putting your code > > > under LGPL (or it is not distributable). However you can change > > > the license from LGPL to pure GPL by simply changing the license > > > statement. > > > > Yes, but that's written into the license, while this is tacked on in a > > random separate file. > > > > I see it a little differently: > > An LGPL library is subject to the LGPL which can be paraphrased > as: > > "This license applies to files which begin with a conspicuous > notice saying so. By distributing etc. this code you > agree to Xxxx and get the right to Yyyy. If you want to, you > may also irreversibly change the license to the GPL for your > copy and copies subsequently made from that copy, you do > that by changing the notice so it points to GPL, not this > file". (The LGPL is written in very formal terms by top > lawyers). > > The reiserfs library is subject to the README license which can > be paraphrased as: > > "This license applies to files which begin with a conspicuous > notice saying so. By distributing this code you agree to > grant the original author all the rights he would have if he > held the copyright in it and gain the possibility that he > might pay you money in return. If you want to, you > may also irreversibly change the license to the GPL for your > copy and copies subsequently made from that copy, you do > that by changing the notice so it points to GPL, not this > file". (The README is written in less formal terms and may > not have been grammar-checked by a lawyer). > > In neither case is the code really under the GPL. In both cases > the license is explicitly GPL-compatible because it allows > derived works (such as a compiled kernel binary or a forked > version of glibc or reiserfs) to be placed under pure GPL. In > both cases distributing with the original notice intact grants > additional rights to some recipients (with very different > benefits in return). In both cases exercising the GPL-only > option for your changed copy implies a fork of the Program > because the original author wants to keep his version under the > original license. > > Hope this makes my comparison clearer.
OK, I agree [1] that this is a GPL-compatible license. [1] speaking only for myself, not FSF. -- -Dave Turner Stalk Me: 617 441 0668 "Once a man has tasted freedom he will never be content to be a slave." - Walt Disney
