sperber(a)informatik.uni-tuebingen.de (Michael Sperber [Mr. Preprocessor]) writes:
dv> Well, at least patchlevels should be only bug fixes of
releases
dv> considered stable and distributed as such.
Well, but nothing holds us to that requirement,
C'mon! You *do* know what I mean. If you tell Joe "this is a
new patch level for 21.0". Joe will understand "this is a new release with
some bug fixes". Of course nothing holds us to that requirement, except the
common sense people give to the notion of patch level. Now if you start
claiming a new patch level is released, and the options file has changed
syntax, expect to be flamed.
and I'm not going to sign any papers saying that we'll only
fix bug. We may
very well *introduce* bugs, for all we know.
That's exactly why the notions of feature/code freeze exist, and why
it's good to firmly adhere to this policy regarding releases[1]. In real life,
nothing is all white or all black. You're never completely sure that you wont
introduce a new bug by fixing another one. However, strongly filtering the
patches (like Tomo's) for 21.1 is the way to limit the risk.
Footnotes:
[1] that's also why it'd be good to have more frequent minor releases.
--
/ / _ _ Didier Verna
http://www.inf.enst.fr/~verna/
- / / - / / /_/ / E.N.S.T. INF C201.1 mailto:vernaļ¼ inf.enst.fr
/_/ / /_/ / /__ / 46 rue Barrault Tel. (33) 01 45 81 73 46
75634 Paris cedex 13 Fax. (33) 01 45 81 31 19