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

Re: [DebianGIS] Re: Mapserver?

On Wed, Oct 11, 2006 at 03:36:57PM +0200, Fabio Tranchitella wrote:
> Same here for the python bindings, but at this point I do not want to go
> ahead without the authorization of the group. I'm sure we can support
> these releases in etch, and there are still two months before the
> release (if we release in time). 
> I'm going to take the responsability for these packages, togheter with
> the package maintainers, to assure that the packages in etch will have
> the required quality for a debian release.
> If somebody **really** think it is not possible for a technical reason,
> excluding the "we miss the man power" resons, then please show me what
> is the next step in your opinion and I'll be glad to go ahead in that
> direction.

If we do need to upload a new geos for that, we will have to pass the NEW
queue (due to usuale name bumping), so I doubt we can go anywhere... 
Aba announce about releasing plan is maybe a bit understimated IMHO 
(if we had really to reduce from ~250 to ~50 RC bugs, I doubt 2-3 weeks are enough),
but realistically, we have a very short period to align ports and
dependecies. Also changes are quite deep on the basis of previous mails
and tests shown 'til now seem only about eventual regressions.
Personally, I did not evaluate the geos/mapserver inners so deep to
estimate if we could have still hidden issues. My considerations are
of general order.

BTW I strongly support (and I'll do my best in the next months, after my current
shortage in free time) a more strict backporting cycle for etch after release 
for (at least) all server-oriented packages. Under this POV, releasing now is of little 
interest. I would stabilize other packages which need more care.

Francesco P. Lovergine

Reply to: