[Klug-general] KLUG Project

Karl Lattimer karl at qdh.org.uk
Wed Mar 12 09:13:27 GMT 2008


Some generic advice on running projects :)

* Don't pick something like wine-doors, we have many many development
issues. Its like trying to fit an hexagonal carpet into a square room. 

We have to match an external tool (wine), with a bunch of inconsistent
operating systems, a bunch of users which install via packages built by
others which haven't been built correctly (there is a knack to it), and
developers that want to spend too much of their time scratching their
own itches rather than developing on the overall app, with a hostile
community of zealots and freetards from different camps...

Things have started to calm down now though, but its only been 2 years
eh!

* If you decide to develop an application you'd be far better off
building something simple which scratches an itch we all get from time
to time.

I'm currently working on something new which should be available soon,
the concept for is to present the most useful feature set in a most
accessible way. I don't like to create knowledge barriers to users
starting to use an application, this shall hopefully show how much GNOME
apps can veer from the flock, as other apps of this kind are generally
designed to require technical support contracts or else how would the
vendors make money...

* Commit yourselves to a certain amount of time per week, and
collaborate in real time at least once a month. 

We had communication issues with wine-doors for the first year or so
(and we're pretty much back in that situation at the mo) where we had no
mailing list, no IRC channel emails were difficult to co-ordinate, the
ticket system wasn't being used correctly, some people were going in
their own direction when working on different sections in different ways
and attempts at working on a standard format we could all work to fell
apart.

Then you've got the contributors that refuse to use a mailing list... or
a ticket system... or write useful commit messages... You shouldn't have
that problem here :)

* Don't try and build the Colosseum from scratch, like Oly wanting to
re-invent the wheel on this thread rather than improving an already
existing one. Its obviously not within the scope of any team like this
to build applications of that magnitude.

* Re-use, re-cycle, re-invent! wine-doors was originally meant to
replace winetools, which it pretty much has, with a few problems here
and there we have achieved that goal already. We lost that as our
primary goal a long time ago but... the advantages to us of having code
already available that would be useful to us as we started developing
wine-doors were immeasurable. Remember it's called open source for a
reason.

Hope this helps :)
 K,




More information about the Kent mailing list