FESCo meeting summary for 2009-06-26

[Kevin Kofler]

Matthew Garrett wrote:
> Not in the slightest. We have a finite quantity of developer time.
> Currently more of that is spent on Gnome than on KDE.

But the people working on GNOME are not the same as the people working on
KDE. Supporting KDE better is not going to make anybody stop working on
GNOME.

> No. You're missing documentation. You're missing integration.

There's plenty of KDE documentation at e.g. userbase.kde.org. If you think
we could use more KDE-specific documentation within Fedora, that's the Docs
Team's task, the GNOME-specific documentation wasn't written by the GNOME
maintainers either. But IMHO it's not really the distro's job to document
KDE, having common documentation upstream which is shared by all the
KDE-based distros is the better model and userbase.kde.org provides that.
In any case, writing documentation needs documentation writers who are not
the same people as developers either.

As for integration, we offer a perfectly integrated KDE spin, thank you very
much... We're working really hard on distro integration. For example, why
do you think I wrote that KDM ConsoleKit patch back in F7 times (a modified
version of which got merged upstream in KDE 4.2)? Where are we lacking
integration?

> Where does this better support magically come from? The number of people 
> working on KDE in Fedora is small compared to the number of people
> working on Gnome in Fedora.

The technical support is already there! We just need to work on the
perception level, e.g. the presentation of the download page or the ease of
selecting KDE in the installer (which requires some small technical
changes, but we certainly have the resources to write the relevant Anaconda
patches in KDE SIG, the real problem is getting the idea accepted in the
first place). KDE SIG is already strong enough to make KDE work well
(though I definitely wouldn't object to more help!).

> Why? The natural choice for KDE users right now is either OpenSuse or
> Kubuntu. Why would these users choose Fedora instead?

Because they offer KDE as a first-class choice. If Fedora offered KDE as a
first-class choice, we'd be right there in that list.

And FWIW many people already choose Fedora for KDE. Around 30% of our users
are KDE users. KDE has become an option which is seriously considered by
KDE users thanks to KDE SIG's work. (As I said, some people even consider
us the best KDE distro!) What's missing (and would attract more KDE users)
is the presentation at the perception level (equal treatment on the
download page etc.). On the technical level, KDE SIG is already doing all
that's needed!

> The alternative is to provide enough resources to be able to hire
> several full time KDE developers.

Again, while I definitely wouldn't object to getting more help, we don't
really NEED more developers. We could always use them, but our current team
is working just fine.

As one of the volunteer KDE SIG members, I also take offense at the implied
claim that only full-time developers can do acceptable work.

> My simple assertion is that a desktop maintained by a small number of
> part-time volunteers is unlikely to be of the same quality as one
> maintained by a larger number of full-time workers.

As unlikely as it may be, it is the case. :-) While I'm clearly biased so I
won't claim one or the other is "better", the quality of our KDE 4 is
definitely comparable with openSUSE's.

> To say otherwise would imply that Suse's maintainers are incompetent.

No, it just means there's only a certain amount of developers which are
really needed to provide high-quality packages.

The OpenSUSE KDE folks spend their additional time on things which are nice,
but definitely not essential (e.g. almost daily snapshot builds of KDE,
something we aren't doing even for GNOME) and on upstream development
(which we all benefit from).

        Kevin Kofler