On Tue, Sep 13, 2005 at 03:37:58PM -0400, michael chang wrote: > I don't suppose we'd know what most of the existing code is licenced > under as a whole, would we? (Ideally, that would seem to be the best > choice as a licence that we use...)
The main codebase is BSD, however some of the code was lifted from glibc, which is under the Lesser GPL. If we were dynamically linking this would be fine, but LGPL code can only be compiled directly into LGPL or GPL licenced projects. The code is the lib/getopt stuff in the main codebase. libconfuse should be ok, because the library is purely linked in, though I'm a little vague on if LGPL allows static linking. This isn't a problem for my packages as I dynamically link as much as possible. The lib/tcp.c file is also LGPL, as is lib/dotconf.c, and some other bits of lib/ that were derived from GNet. There's also some Apache Softwware Licence stuff in there too which if not acknowledged in the docs may constitute a licence breach. lib/readdir.c is ASL v1.1. APR is ASL v2.0 but again I'm dynamically linking that. libmetrics seems to be a world of pain where licencing is concered. There's some explicitly GPL stuff in there, srclib/libmetrics/linux/fsusage.c specifically, so libmetrics isn't even self-consistant when it comes to licencing and is probably undistributable anywhere as it stands. Ideally every source file would have a copyright statement to define who owns the copyright and an explicit licence statement or pointer to a readme for the licence, though obviously this isn't always practicable. Debian are very anal about this stuff to ensure that their distribution really is free, but it looks to me that as it stands, debian notwithtanding, there are some major licence issues with the codebase in it's current form, which may render it actually illegal to compile at all. Of course IANAL and not even really a software licence expert so I could be being overcautious, or even have not noticed some more incompatible source licences in the mix. Stu -- | Stuart Teasdale | Atmospheric, Oceanic and Planetary Physics | | System Manager | University of Oxford | | "You don't need a weatherman to know which way the wind blows" |
