Re: Diety UI draft
Raul Miller <rdm@test.legislate.com> wrote:
> > The presence, in a packages file, of either a newer version of
> > the package or a package which superceeds the package should be
> > required before a package may be considered obsolete.
Behan Webster <behanw@verisim.com> wrote:
> How about the case of a package being dropped from the distribution?
> By your method, the package will never be marked obsolete even though
> (at least in my book) it has been obsoleted.
I don't see a problem with this -- it'll be marked depreciated, which
means roughly the same thing.
> To pick a nit, I don't think marking a package obsolete if there is
> a newer version is particularly useful. Having a newer version of
> a package is usually considered an upgrade, isn't it? Or perhaps
> that's not what you meant above.
Installing the newer version is considered an upgrade. I'd like
to be able to easily distinguish between packages which ought to
be upgraded from packages which I shouldn't expect to upgrade. This
is particularly useful when problems occur (or when a lot of time
will be involved before the upgrade).
> I believe there is a difference between a package being obsolete and
> a package being replaced. This difference is made in the current
> design.
I think you're referring to the fact that when it's replaced the old
one won't be installed any longer?
> But like I said, the absolute worst that can happen is that a package
> can be mistakenly shown to be obsolete (no other action taken, except
> looking for a replacement package). Is that so terrible?
>
> The only time I can ever imagine packages to be mistakenly shown to
> be obsolete is if you alternate between 2 very limited source lists
> that (somehow) have very little overlap in packages instead of a
> larger superset of the two. I'm not sure why someone would want to
> do such.
I'm not concerned with the exact language so much as I'm concerned with
understanding my installation. If there's some other way that whenever
I see the symbol "obsolete", I can distinguish between replacement available
and replacement not available, I'm happy.
--
Raul
--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
debian-devel-request@lists.debian.org .
Trouble? e-mail to templin@bucknell.edu .
Reply to: