Hi, On 6 Sep 1999, Manoj Srivastava wrote: > [...] > Next stage: : /usr/doc is illegal, and we start filing bugs against > packages still using /usr/doc. Symlinks are still > required. > [...]
This is exactly what I am trying to avoid. The purpose of requiring symlinks is that the user is able to look for docs in a single place, be it /usr/share/doc or /usr/doc, during the transition period. If we reduce the transition period to potato only, the symlinks should not be "required" really. The technical committee has decided (among other things) that /usr/doc must still be the single directory a user will have to look for docs in potato. In my opinion, this exceptional measure should not last longer than one release, i.e. potato, or else the whole FHS transition will last much more than expected (the FHS transition will be "complete" when all docs are in /usr/share/doc and /usr/doc is empty). So, we should probably set /usr/share/doc as the single directory a user will have to look as a sort-of-release-goal for potato+1. [ When I say sort-of-release-goal I do not mean that the release is delayed because of this, I mean that bugs are filed etc. ]. This way, developers who like symlinks will do the transition by making symlinks, and developers who dislike symlinks (remember that there are a *lot* of them) should be able to do the transition anytime during the unstable release of potato+1 in one shot. If we know in advance that we want all packages to use /usr/share/doc in potato+1, we will be able to start uploading pure-FHS packages as soon as potato+1 is created. Question: Does this proposal sound "compatible" enough with what the technical committee has just decided so that it may be approved by the ordinary policy procedures? If so, I will probably make a formal proposal and look for seconds. Thanks. -- "54bf9995152351fd7f5eea85a0d98137" (a truly random sig)
