[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Updating fis-gtm package to 6.1



On Fri, Feb 07, 2014 at 09:01:44PM -0500, Bhaskar, K.S wrote:
> >Is this a vote to keep three fis-gtm versions at the same time inside a
> >Debian release?  If I understood Luis correctly we are at first
> >targeting at some educational VistA installation.  From your explanation
> >it seems to become clear that if you want to support a Clinic you
> >somehow need a support company that might perfectly profit from our
> >fis-gtm packages but the maintenance of all previous versions need to be
> >put on their shoulders.  We should not try to push the burden on the
> >Debian release team / security team.  That's simply not how Debian
> >works.
> 
> [KSB3] I don't think there will be an excessive burden on the Debian
> release / security team.  Let me propose that we see how it goes and
> if it looks like it will be a burden, we'll think of a different
> strategy.

Please trust my experience that it is not about what you think but what
is accepted by the gatekeepers of debian (=ftpmaster):  They will not
accept your plan and will refuse multiple versions of the same software
if the number goes (far) beyond the usual migration procedure.
 
> >>Because bugs in GT.M are very infrequently encountered in
> >>production,
> >I'm sorry but as long as any software is not written by the magical good
> >of infallibility it will have bugs.  So we need to be prepared to fix it
> >and we should try to keep the burden for people who need to do the
> >actual work at a bearable amount.
> 
> [KSB3] The upstream team will fix material bugs in a timely fashion.
> Yes, GT.M has many bugs that the development team knows about, but
> remember that GT.M has been in daily production use since 1986, and
> available as free / open source software since 2001.  But the track
> record for the last ten or more years is that it is very rare for
> users to encounter bugs in production environments.  So again, let
> me propose that we see how it goes in practice.

I have no reason to doubt your statement but you forget that also
non-upstream bugs might create some unexpected work.  My suggestion is
that we will go for the moment with fis-gtm 6.0 and 6.1.  If it really
turns out that the 6.0 packages remain as static as you are predicting I
do not see any problem from users perspective to fetch the packages from

   http://snapshot.debian.org/package/fis-gtm/6.0-003-2/

and they can pin their system via sources.list and apt preferences to
this version which solves the user problem without creating any trouble
on the Debian side.  As I said:  It is not me who is vetoing your plan -
there are some people with a more general overview who need to accept
it.

Kind regards

         Andreas.

-- 
http://fam-tille.de


Reply to: