Re: post-release package update policy

[Marcus Daniels <marcus@sysc.pdx.edu>]

>>>>> "IanJ" == Ian Jackson <ian@chiark.chu.cam.ac.uk> writes:

IanJ> David Engel's proposed stable tree which only gets bugfixes
IanJ> provides this too.  We don't need to cast the release in stone.

So call it R6.0, R6.1, R6.2, etc (and call the experimental version
R7alpha, or "experimental", "development", or the like). The arguments
for an unfrozen the stable tree are reasonable enough, and probably
more consistent with the ethic of Free software.  However, the
combined bitstring of all the package release dates and/or version
numbers does uniquely identify a system -- this huge number can be
thought of as a version number.  Sure, it could be generated on the
fly for purposes of reporting bugs or for communicating needed
information to support people.  But if R6 doesn't have minor release
numbers, and the documentation doesn't emphasize that "R6" is not a
sufficient identification it will create the image of administrative
sloppiness and create confusion-- especially novice users who think
they are going to rely on vendors or other support people.  "But I
have R6 -- what do you mean I'm not telling you enough?!"

IanJ> Just because there is no one thing you can point to
IanJ> that is Linux 1.2.x doesn't mean that it isn't a useful concept.

But there is "x"!!