Re: Official Exim 4 package
On Sat, Mar 22, 2003 at 03:16:14PM -0800, Steve Lamb wrote:
> On Sat, 22 Mar 2003 00:11:18 -0600 "Jamin W. Collins"
> <jcollins@asgardsrealm.net> wrote:
> > On Fri, Mar 21, 2003 at 09:08:49PM -0800, Steve Lamb wrote:
> > > Problem is a lot of people ignore those warnings. That's because
> > > most of the time the warning is meaningless in that they warn
> > > people of a config file change that could cause problems and in
> > > 99.9% of the time it doesn't. It leads to the "yeah, yeah,
> > > whatever" syndrome.
>
> > Doesn't mean the warning couldn't/shouldn't be given.
>
> Which is not what I said. What I am pointing out is that the warning
> *isn't enough* for a base package that breaks configuration
> compatibility.
Not what you *stated*, but it was rater effectively implied. The
indication was (at least to me) that you were saying the notices were
ignored and therefor rather useless. And, as I've already stated, v4
could be packaged in such a way that it does not directly upgrade v3.
Thus, it would *not* be a /base/ package.
> > Not so. "exim" is a base package. That doesn't mean that "exim4"
> > would have to be a base package. It would just be another
> > _optional_ package.
>
> Which I addressed by asking "Do we really need to start putting the
> version number in the package name now, IE exim4?" which you deleted.
Because my response was not related to it. But since you seem to really
want me to address it, I see no problem with major changes like this
taking this route. It allows the end user to choose what /they/ want.
> We have a version number field for a reason yet more and more we now
> have the version number in the name which is getting annoying. Do we
> really need exim and exim4? Shouldn't it have been exim3 right away?
> Should we go through and add major version numbers to every package?
When necessary sure. Your definition of /necessary/ will not
necessarily be the same as mine.
> > It need not be structured as an upgrade of a base package.
>
> I think it would be mistake if it weren't. Not without some change
> that makes it clear when we put version numbers in the name and
> when/if we don't *or* that something gets into place on the version
> number field to address this so we don't have to remember to add the
> major on this package but not on that package.
Matter of preference. There are a number of packages in Debian with
their version (or some other indication of their version) in their name
already.
--
Jamin W. Collins
Reply to: