Is anybody is fixing bugs?

Frank Osterfeld frank.osterfeld at gmx.de
Sun Jun 17 09:58:24 CEST 2007


On Sunday 17 June 2007, Richard Kakis wrote:

> It could be worse.  They might close the bug as WONTFIX within 24 hours.
>   Yes, someone should contact you within 30 days.

"Someone should". Replace that by "I will" :)

> > Unfortunately, this is not how Free Software (and specifically KDE)
> > works. You can't force people to fix bugs or react to reports within
> > a certain timeframe -- most of the people are volunteers after all.
>
> Volunteer != 'do what ever you want'.

Volunteer = Spending lots of time besides your day job on something you love 
and care about, and prepare being told by passive spectators that you're 
doing not enough, or the wrong thing.

(Well, there are lots of positive aspects to it as well, otherwise people 
wouldn't do it)

> If I volunteer to help Habitat for Humanity build a house, I expect to
> be told what to do. 

That never worked for OSS. The people who ask for jobs are rarely the ones who 
stay around longer and do sustainable work. This is just not how 
(self-)motivation works in OSS.

> I might be offered a choice of jobs, but anarchy
> doesn't really work in building a house.  I don't see how we should
> expect it to work here.

It's not anarchy. It's meritocracy. The only anarchic about it is that the 
volunteers decide what they work on and how much they work on it.

> > That said, it's wrong to conclude from one bug not being handled in a
> >  timely manner that bugreporting is useless. There are a lot of bugs
> > fixed every day (http://www.commit-digest.org/ has more details in
> > the statistics section).
>
> But, we do have a bug problem.  Perhaps we need to refactor our software
> engineering methods.

Go find methods, that
1) are appreciated by the contributors (the ones who do the work), i.e. no 
unecessary paperwork or other additional overhead.
2) work together with the current "workflow", the motivation of contributor's, 
the social structures and whatnot
2) scale for a big FLOSS project like KDE
3) actually improve something instead of causing just work
 
> This is really a lame excuse.  If we are too busy developing new stuff
> to fix bugs, we have a serious problem.

People working voluntarily on KDE4 don't need to apologize to nobody for doing 
that. FLOSS developers aren't resources you can manage like employees. For 
instance, if they decide to write a new multimedia framework (which is 
needed) they do that, instead of doing bug hunting in Arts (pointless 
long-term). If you want the same person to fix arts for example, you'd 
probably need to pay him/her loads of money.
Also, often it turns out to rewrite something from scratch instead of fighting 
to get some unmaintable piece of legacy code stable.

> Do we have a list called: howtohelp?  Actually, it seems that what is
> needed is a list for people to post requests for help.

Actually this is what this list is about, it just got an unfortunate name.

In short: If you want to influence an FLOSS project like KDE, join it and do 
work on it. If you do good work, people might start to listen to your 
suggestions. Little steps and improvements here and there will be probably 
more successful instead of trying to tell people that they all should switch 
to methodology X tomorrow
Judging as a spectator how bad things are managed won't get you nowhere.

Regards,

Frank
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://mail.kde.org/pipermail/kde-quality/attachments/20070617/babbb1fc/attachment.pgp 


More information about the kde-quality mailing list