Re: A suggestion for the woody freeze

[Adrian Bunk <bunk@fs.tum.de>]

On Thu, 17 Jan 2002, Tomas Pospisek's Mailing Lists wrote:

> On Thu, 17 Jan 2002, Adrian Bunk wrote:
>
> > Jan 25.-27. bug squashing party in preparation of the freeze
> >
> > Feb 1.     freeze; start of the first bug squashing cycle
> >            every upload to frozen must be approved by the release manager
>
> My personal opinion on this is that there is very little motivation for
> joe developer to help with the freeze. The average developer is runing
> either testing or unstable and accepts the fact that he's spending
> 10min/day fixing little bugs and can do an apt-get update every day. He
> scratches his itch and that's it.

The main question is:

What are we doing our work for?

Do we want to make a high quality stable distribution or do we do our work
only for itself without planning to release?
If the latter is true we should announce officially that there will be no
more stable release of Debian.

I do personally use a mixture of testing and unstable at home but at work
I want to use a stable distribution because there it's important that the
machines don't break because a buggy package went into testing.

If there will no longer be regular stable releases of Debian this means
that you have to switch production machines to other distributions.  :-(

> IMHO it'd help if there was some global freeze in the sense that it'd be
> hard to even get packages into >unstable<. That'd mean that before a
> stable release isn't out new packages will not make it in. The same way
> kernel freeze works. That'd mean that if a developer would like some new
> version or update of a certain software or a new program to be
> available it'd mean :
>
> 	* wait until the release is out
> 	* or help out to make that happen faster and we can move on then.

I hope this is not needed.
But yes, if the problem exists in the way you describe it it might even be
needed.

> 2¢
> *t

cu
Adrian