The US legal system only recognizes copyright in the scope of a "work as a whole" and does not distinguish between individual holders of joint copyright. This produces a number of fundimental conflicts with the way software is developed. One of the biggest questions is, if a copyright legal issue involving "work as a whole", such as say, the Hurd kernel, must be resolved in court, and with existing case law, the courts could choose to either recognize each and every individual retaining copyright as singularly having standing to bring or participate in a case, or it could choose to require each and EVERY copyright holder to participate, and either outcome is possible, with existing precidents for both, largely at the discretion of the judge...
Having a single, or limited number of legal copyright holders available makes it possible and practical to engage in legal action where it may be needed, such as if, for example, some company deliberately chooses to ignore the GPL... Another issue that exists with individual copyright holders is that if the licensing needs to be altered, as happened to Mozilla, since the courts only recognize the complete "holders" of a work as one physical entity, one has to track down and gain consent from each and every holder. For the Linux kernel, this could involve receiving notorized approval from several thousand individuals around the world at present, and if even one chooses not to agree...well...then you are stuck. Now, since the Hurd is under the GPL already, it is unlikely to change licensing. However, it is always possible, however remote, that some court may in the future choose to invalidate some key part of the GPL as a whole and require an entirely new license to be created. Should such an event come to pass, then yes, it would be nessisary to change the Hurd license, and, as noted above, with many individual copyright holders, that could be very difficult or impossible to achieve. The key to successful copyright assignment is in having an institution one can trust to hold copyright. Certainly an untrustworthy copyright holder could invalidate the intentions of those who entrusted it with their copyrights. The FSF is, I certainly believe, a fully trustworthy entity to hold copyright, and is not the only one I would trust. I personally believe it is best to have copyright held or shared by a few such trustworthy organizations rather than one alone. However, on any free software project with a large number of contributors, I also believe it is a fatel mistake to not do some form of copyright assignment, and it probably should be done even with projects that have a very small number of contributors. David On Tue, 1 Oct 2002, Jeff Bailey wrote: > On Tue, Oct 01, 2002 at 08:00:45PM +0200, Ludovic Courths wrote: > > > Ok. I didn't expect the process to be this complicated actually. ;) > > I thought that the top line of the files which says "copyright FSF" > > would just be enough. > > The US legal system is ugly *and* evil, and they require us to fill > out forms that legally give the FSF possesion of all the GNU code you > write. > > It's a bit of a pain to do, but it means that the FSF could fight in > court in the event of a copyright violation. Otherwise, all of the > developpers have to get together to fight it in court themselves. > > Tks, > jeff Bailey > > _______________________________________________ Help-hurd mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/help-hurd
