Raul Miller <[EMAIL PROTECTED]> writes: > The following had no copyright files. Some of them had symbolic links ... > I think that current policy allows this, and I think that policy is a > mistake
I think it's not explicitly mentioned in the policy but the de facto policy seems to be that it's ok if the package Depends on the package containing the copyright file. That seems reasonable, it's usually implemented by linking the /usr/doc directory itself to the doc directory of the base package. Otherwise multi-binary packages become quite a pain since you would have to duplicate lots of files or set up lots of otherwise empty /usr/doc directories. > Most of the remainder had GPL licenses. But I was shocked to find that > most had *only* GPL licenses. If that was really the case, then these > should go in main. Otherwise, I think most of these need the LGPL license Note that in many cases our requirements may be stricter than the the scope of the GPL and `derived work' rules. For example, the emulator programs are in contrib because they typically need a ROM image to be useful. However I would not expect them to be derivative works of the ROM images (thouh possible the reverse, the ROMs could arguably be derivatives of the architecture they were designed for). I don't think the GPL would normally require distribution of the ROM images. Similarly with the doom engine and the WAD files. etc. Also, it is not necessary to move them to the LGPL and in many cases it may be inappropriate and mean giving up much of the GPL protections. In most cases a simple exception for the required non-free module is all that's necessary. Something like: ``You may distribute this package under the terms of the GPL. As a special exception you may distribute it without including the source of the Motif library even though the GPL would normally require this.'' The Motif library being typical of the type of problem in these packages. You might check news logs for RMS's example wordings. There may be details I've lost in mine. Incidentally I think I've already notified a couple of these authors of these issues but I forget which. (explorer and/or dfm I think). greg -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
