[Kolab-devel] Re: Fedora packaging

[Karel Volný]

Hi,

> Although concrete decisions related to policy may differ somewhat, there
> might be some benefit at looking at plans for Debian and perhaps openSUSE,
> too.

you mean distro plans or Kolab plans? - any link what to read, please?

> I haven't really looked at the Debian packaging much since December, but I
> did write up an approach that kept setup-kolab and made it work slightly
> better with the way packages are installed and configured in Debian, and it
> is conceivable that Fedora behaves similarly enough that the general
> dependency structure and approach is at the very least worth looking at:
>
> http://blogs.fsfe.org/pboddie/?p=623

"Take MySQL as an example. Upon installing the appropriate Debian package providing the MySQL server, the user is prompted to set up the administrative credentials for the server. After this brief interaction, MySQL should be available for general use without further configuration work, although it may be the case that some tuning might be beneficial. It seems to me that Kolab could be delivered with the same convenience in Debian."

Exactly my point! (in principle)

BTW, thanks for the diagram, great introduction for me.

> I put some Debian packages related to this work in the following place:
>
> http://www.boddie.org.uk/downloads/kolab/
>
> From what I understand of the discussion about this in December, things like 
> setup-kolab are not exactly what one would normally expect in Debian, but at 
> the same time there is surely a benefit to not have to do the configuration 
> work from scratch every time someone wants to support a distribution: setup-
> kolab surely does useful enough work in most situations and can at least
> serve as the template for whatever gets deployed on a distribution, but it's
> easier to keep it around and make it do the right thing, which shouldn't be
> too hard despite distributions wanting to call services different names 
> and other such annoying but relatively trivial stuff.

In an ideal world, it would be just a matter of changing a few globals, like where the distros like to put the stuff, how do they call things, what initsystem they use etc. and running the same script everywhere. In reality ... I can hardly imagine it would work that way, but we can at least try to split the steps into generic and distro-specific.

> The other sticking point appeared to be delivering Kolab to work with the 
> official distribution packages for things like cyrus-imapd and roundcube (on 
> Debian, at least) where there may need to be some improvements from
> unofficial packages ported over to the official ones.

Yes, that's the thing I'd like to see resolved too. While Red Hat has taken the opposite direction with Software Collections, I highly doubt that average users would be much happy having multiple instances of basically the same code, having to wonder which one is getting the fixes they care about, different places to track bugs, and how do the alternative packages fit in the rest of the system ...

K.

-- 
Karel Volný
QE BaseOs/Daemons Team
Red Hat Czech, Brno
tel. +420 532294274
(RH: +420 532294111 ext. 8262074)
xmpp kavol at jabber.cz
:: "Never attribute to malice what can
::  easily be explained by stupidity."