Updates and delays in signing packages
mjg at redhat.com
Sat Jul 18 03:23:47 UTC 2009
On Sat, Jul 18, 2009 at 06:45:21AM +0530, Rahul Sundaram wrote:
> On 07/18/2009 06:30 AM, Matthew Garrett wrote:
> > If software is that unstable then I think it's reasonable to ask whether
> > it's ready to be shipped in a stable distribution release.
> I think it is. It is not crashing and burning for all the users using it
> including me all the time. Let's take the latest bug report I have received
> It requires you to access the help in a particular way to get it to
> crash but I consider a crash a problem that needs to be fixed quickly.
If it's not a common pathway then I don't think it's urgent - certainly
not urgent enough that waiting an extra week is a big problem.
> This doesn't make it in unstable enough to be excluded compared to rest
> of what we have been included by default in the past and Gnote is not
> the default in any stable release. Anyway, you are sort of missing the
> point by focusing on the example given without understanding that for a
> certain class of software (new project where upstream is active, user
> demand new features etc) with frequent updates, the current system isn't
> working out well. I doubt I am the only maintainer with this problem. I
Most of our users wait 6 months between releases. If they can wait that
long to get a new piece of software in the first place, the difference
between an upload a week and an upload every two weeks shouldn't be much
of an issue. I just really don't see how feature requests can be that
high a priority during a stable release.
Matthew Garrett | mjg59 at srcf.ucam.org
More information about the devel