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

Re: Bug#123769: Problems for evolution into woody



On Wed, 2001-12-19 at 07:33, Grant Bowman wrote:
> 
> So, back to the issue at hand.  The update_excuses looks like this:
> 
> * evolution/alpha unsatisfiable Depends: libgnome-pilot1 (>= 0.1.63) ['gnome-pilot']
> * out of date on arm: evolution, evolution-dev, libcamel-dev, libcamel0 (from 0.99.2-1)
> * evolution/i386 unsatisfiable Depends: libgnome-pilot1 (>= 0.1.63) ['gnome-pilot']
> * out of date on ia64: evolution, evolution-dev, libcamel-dev, libcamel0 (from 0.99.0-2)
> * out of date on m68k: evolution, evolution-dev, libcamel-dev, libcamel0 (from 0.15-3)
> * out of date on mips: evolution, evolution-dev, libcamel-dev, libcamel0 (from 0.99.0-2)
> * out of date on mipsel: evolution, evolution-dev, libcamel-dev, libcamel0 (from 0.99.2-1)
> * out of date on powerpc: evolution, evolution-dev, libcamel-dev, libcamel0 (from 0.99.2-1)
> * out of date on s390: evolution, evolution-dev, libcamel-dev, libcamel0 (from 0.99.2-1)
> * out of date on sparc: evolution, evolution-dev, libcamel-dev, libcamel0 (from 0.99.0-2)
> * Not considered
> * Depends: evolution gnome-pilot
> 
> Would fixing libgnome-pilot1 or changing this to a recommend instead of
> depend allow evolution to move into woody?  MUST all platforms be fixed
> simultaneously?

If it has ever built for a platform then it is expected to build for it
again and therefore it is a bug if it doesn't.

Yes, libgnome-pilot1 will need to build on these platforms before
evolution will be considered for installation in testing.  I suspect
that everything will need to build on all of the "out of date" platforms
as well.

If you want to use this I would recommend that either you (a) use
'unstable' (personally, I would recommend this) or (b) use the
/etc/apt/apt.conf hack to let you use a mix of 'testing' and 'unstable'
packages.

Filing bugs against evolution will only get people's backs up because
evolution is not the problem here.  What you are doing is like filing
bugs against Sawfish because it depended on a particular new version of
glibc that wouldn't compile against sparc.

If you really want to help, make libgnome-pilot1 compile on the missing
architectures.  If you can't help on that, hassling the developer will
do nothing more than annoy a bunch of fairly ornery volunteers.

Given the stability of the interfaces to evolution (gtkhtml, camel, ...)
I know that if _I_ was the DD responsible I wouldn't be too stressed for
the missing architectures at this stage.  Once we get the base system
ready in January or so, _that's_ when I'd start seriously looking at it
and deciding that either (A) Alpha is not important, or (B) now is the
hour to make it all work on alpha.  To do it now is really makework, or
potentially makework, and we're all volunteers.  At this time of the
year there is plenty of other stuff to do in our real lives.

"Testing" is not normally a distribution that is actually runnable.  The
distributions that are usable are either 'stable' (should be always
usable) or 'unstable' (expect occasional breakage).  Maybe that doesn't
seem right, but 'testing' as a distribution, has only been around since
'potato' was released.  There are tricks to make it 'mostly usable' but
you will have a lot less pain with 'unstable', and bug reports against
'unstable' are a lot more useful to developers.

Regards,
					Andrew McMillan
-- 
--------------------------------------------------------------------
Andrew @ Catalyst .Net.NZ Ltd, PO Box 11-053, Manners St, Wellington
WEB: http://catalyst.net.nz/        PHYS: Level 2, 150-154 Willis St
DDI: +64(4)916-7201    MOB: +64(21)635-694    OFFICE: +64(4)499-2267
       Are you enrolled at http://schoolreunions.co.nz/ yet?



Reply to: