Re: Bizarre changelog entries in some gnome packages
On Tue, Mar 23, 2004 at 03:42:33PM -0300, Humberto Massa wrote:
> Matt Zimmerman wrote:
>
> >On Tue, Mar 23, 2004 at 02:41:53PM +0100, Mathieu Roy wrote:
> >>And on what basis do you decide what is history and what is not?
> >>These malformed changelog entries are definitely a part of the
> >>package's history.
> >They do not document any change to the package, except one(?) of
> >the fragments which mentions orphaning. The others are worthless.
>
> The entries themselves are changes to the packages
Oh yes, of course they are. And like any other changes to the package, they
can be edited if it improves the overall quality of the package. I do not
consider changelogs to be works of art, soapboxes, personal journals, or
sacred relics, and I certainly don't think that they're inappropriate to
change if the result is a more accurate history of the package.
This doesn't mean that I would modify the changelog for just any reason, but
if I forgot to mention something that changed, I would not hesitate to add
it to the entry for that version as a correction. If a maintainer fixes a
security vulnerability without a cross-reference, I encourage them to add
the CVE name to the corresponding changelog entry in a future upload.
>, and document history in a broader sense.
The line of text starting with " *" in this entry:
gnome-session (2.4.2-2) unstable; urgency=low
* order to
-- Christian Marillat <marillat@debian.org> Fri, 19 Mar 2004 15:50:59
+0100
is content-free. The maintainer thinks it is a funny joke, but if the next
maintainer were to replace that entry with one which actually listed the
changes in version 2.4.2-2 of the package, I would support that change.
--
- mdz
Reply to: