Re: Gnome development thinking!

[Henrik Brink <hb pro-it dk>]

tir, 2003-09-02 kl. 11:38 skrev Michael Meeks:
> Hi Henrik,
> 
> On Mon, 2003-09-01 at 12:09, Henrik Brink wrote:
> > Why is the development-process so (relatively) hard to join? I shouldn't
> > contact you to say that I want to help out. It should be an obvious
> > thing!
> 
> 	:-) I think a good driver for development is solving problems you
> personally have an interest in - that should lead you to one particular
> project eg. D/BUS or OpenOffice.org - and you contact people, read the
> code and get stuck in.

Yes, I can see what you mean. As you've probably guessed, I have no
experience in Gnome coding, and I just thought I would make a proposal
for every aspect og the Gnome development. 
I still want better translation, and then I'll have to join every
project that doesn't have danish translation. I know a couple of people,
including myself, that could do these translations very easy if there
wasn't so much work, just to contact people etc.
Many new people will feel the same way, and better translation will mean
more Gnome-users and so on.

> 
> > Imagine a GUI application. This program allows you to join development
> > groups (workgroups) in which you are skilled, and presents some possible
> > tasks. If you take the task, it disapears from the list, and when you
> > are finished it shows up again for the other people in your workgroup to
> > comment on your work.
> 
> 	There are serious scalability problems with distributed lock
> management; particularly - some people are all talk and no action, and
> this type of person may be relied upon to take as many exciting sounding
> tasks as possible - and never finish them ;-) Ultimately, there are
> enough real problems out there in need of fixes that the likelihood of
> duplication of effort seems fairly low [1].
> 
> 	Of course - if you want to start bold new endeavours, make visionary
> brush sweeps of code etc. then you're going to conflict with loads of
> people ;-) so don't. Just join an existing project and make small
> iterative improvements, then you'll soon be able to see where the action
> is, what is needed and take on bigger projects.
> 
That was not what I meant. I agree a lot with the idea of having as few
projects as possible, and for the most people to work together as
possible.

> 	That at least, is my experience ;-) oh, and if you want a task - I
> happen to know that a gtk+ frontend for the x3270 IBM terminal program 
> would be rather good news for Gnome; you'd learn a lot about gtk+ I
> guess, but then it's better to hack with other people - at least at
> first.
> 
> 	HTH,
> 
> 		Michael.
> 
> [1] - see http://www.gnome.org/~michael/todo.html for a longish list of
> things that never get done; indeed, you could have been working on
> almost any-one of these for a year and not had a conflict ;-)