Re: question for sources.list
On Sat, Sep 03, 2005 at 12:09:39AM +0200, David Jardine wrote:
> On Fri, Sep 02, 2005 at 02:48:33PM -0600, Paul E Condon wrote:
> >
<snip>
> >
> > "Don't use release code names in configuring the apt system."
> >
> > seems to me to be a good general rule.
> >
> > As a comment to those who have used code names and think they
> > work: They work only if you have only one release in your sources.list,
> > and then they don't really work, they just don't matter because
> > there is never a choise of release to be made.
> >
> > Personally, I was rather disappointed when I first made this discovery
> > in apt/preferences. I thought I had figured out a neat lazy-man's way
> > to handle the transitions from one stable release to the next. But I
> > won't describe it because I have established that it doesn't work.
> >
>
> Paul, I think you were one of those (forgive me if I'm wrong) who
> shot me down a couple of months ago for suggesting that such words
> as "stable", "testing" and "unstable" might better be reserved as
> purely descriptive words for real things like "sarge" or "etch".
>
> I confess my vast ignorance of how all the developers, release
> managers, etc do their work on debian, but isn't the whole point of
> the operation to provide us, as users, with a system that is as
> easy as possible to use (without MS-type dumbing down, of course)?
>
> When we start with debian, we learn about "stable", "testing", etc
> (by the way, what's the correct word for these things? Are they
> different releases? distributions? flavours? versions? ...?). We
> decide what sort of life we want to lead and pick the appropriate
> thing ("woody", sarge", "etch"...). When a new thingummy becomes
> the official stable and the others move up, we take advice (from
> this list, probably) about when to dist-upgrade. We're in control
> of our system and know that a word like "stable" is not suddenly
> going to point to something quite different.
>
> You say you were "disappointed" that you couldn't base things on
> the code names. Well, yes, aren't we all?
>
I don't think I shot you down. In my opinion, release code names ought
to be the primary key by which a release is designated. I might have
had another opinion in the past, If so, it is now inoperative.
And, I reaffirm my report concerning what actually works in Debian as
it exists today. I'm not a developer and I'm not qualified to be one,
so I am not qualified to judge what level of effort would be required
to move Debian to the situation I would like, nor what other projects
might be delayed. Debian is already the slowest distribution on
release cycles. Personally, I don't mind slow release cycles, but I am
aware that many other users would like more frequent releases. All
this work has to fit with the resources available.
So, until some time in the far future, people should not say to
newbies that release code names and release status names ('stable',
'testing', etc.) are interchangeable. They are not. Existing support
for release code names is, in fact, quite restricted by comparison.
--
Paul E Condon
pecondon@mesanetworks.net
Reply to: