[mythtv-users] rpms for 0.18 and CVS rpms (was: 0.18's available)

[Ryan A. Carris]

On 4/15/05, Axel Thimm <Axel.Thimm at atrpms.net> wrote:
> On Fri, Apr 15, 2005 at 11:18:13AM -0700, Big Wave Dave wrote:
> > On 4/15/05, Brad Templeton <brad+myth at templetons.com> wrote:
> > > On Fri, Apr 15, 2005 at 12:55:17PM -0400, Nate Thompson wrote:
> > > > According to atrpms.net, it is still .17. I'm sure it'll be updated soon.
> 
> The core package (not plugins) are uploaded right now onto bleeding
> for 4xFC, 2xRHEL and 2xRHL. The recent reorganisation of the plugin
> sources requires some more attention from a packaging POV, so they
> will be shipped later.
> 
> > > It might very well be a prudent policy to have as many people run the
> > > cvs or rpm-from-cvs versions at this point, and then build the stable
> > > rpm a week later after all those first-line folks have tried it.
> 
> That should happen *before* a release, and that's why there are CVS
> rpms shortly before releases. I only made them public to the
> mythtv-dev list, as I'm not certain that Isaac wants to see this
> testing done oouside of mythtv-dev (may be too much noise otherwise,
> and he's busy enough before the release anyhow).
> 
> > I think the CVS RPMs being released right-away, and then the actual
> > release RPMs a day or two later might not be a bad idea.  In some ways
> > it would be like issuing "release candidates".
> 
> See above, this has already happened last week, but with a low
> profile. If there is interest and the developers (well, mainly Isaac I
> guess) agree, the CVS rpms could be made more visible before releases.
> --
> Axel.Thimm at ATrpms.net
> 
> 

With Jarod now maintaining the -stable branch, or whatever it is going
to be called, I don't beleive packaging the CVS is necessary.  There
is way to much noise on the user list as it is.  The releases are
usually pretty stable.  If there is a major bug fixed, Jarod will
probably have the fix packaged up real nice for Axel like he did
recently.

Besides, I'm not sure how much rpm's of CVS will help Isaac.  Without
debugging built in and people running it in gdb, there isn't a
backtrace.  And if people "install per Jarod's guide", they don't even
have the frontend and backend running in a window where they can cut
and paste the error messages.