[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-devel
Subject:    Re: Call for Kastle branch
From:       James Richard Tyrer <tyrerj () acm ! org>
Date:       2003-08-16 19:29:57
[Download RAW message or body]

Rob Kaper wrote:
> In a rather long post on my blog ( http://capzilla.net/blog/2057/ ), I've
> made a call for a Kastle branch.
> 
> In short, I think it makes sense to consider those contributors not being
> able to participate by giving them a chance for more peer review and
> discussion before new features are committed as a result of the hackfest.
> 
> It's quite possible that "non-trivial" changes will be made and I'd like to
> remind everyone that it's a good idea to discuss those changes on the
> appropriate mailinglists first and give contributors not present a chance to
> voice their concerns, requirements and wishes.
> 
> I realize there's nothing in our policies to enforce this kind of behavior,
> but note that it would be respectful.
>  
<<
There may be no I in team, but there's a K in teamwork all right. Projects 
like KDE celebrate distributed development. All changes are available in a 
public CVS repository and practically every single thought or idea is being 
discussed on the KDE mailinglists beforehand. This method of work respects 
every single contributor, regardless of location.
 >>

I don't think that this is the case.  I am on record as strongly favoring 
discussion of how KDE should be designed before code is written, or failing 
that, after code is written and it doesn't work.  I don't see that 
happening in many cases.  I have suggested that this discussion should 
occur on the mailing list, perhaps an additional mailing list is needed, 
but have been told that Bugzilla is the place for this discussion.

Obviously, a discussion on Bugzilla can only occur after the code has been 
written -- in some cases before it was designed.  Obviously, it would have 
saved effort in many cases if the discussion had occurred before the code 
was written.  But, I have been told that the rule is that he who codes 
makes the decisions.

However, it doesn't really make a lot of difference who discusses the 
questions because it is the discussion that leads to new ideas.  So, the 
"hackfest" would be a good thing because it should lead to discussion.  The 
problem I see is that this discussion would not be for a long enough period 
of time.

If that is the case, there will be a need for more bug reports.  And bug 
reports do not, despite what I was told, promote discussion because bug 
reports are short and are subject to misinterpretation -- developers think 
that you are criticizing them when you tell them exactly what their code 
does wrong.

Discussion is a good think, and I would like to promote it.  But, I don't 
see how adding an additional tag to CVS commits from Kastle will help.

Just wait and see how much discussion of my recent post:

[PATCH] Qt PostScript font names

there is.

--
JRT

 
>> Visit http://mail.kde.org/mailman/listinfo/kde-devel#unsub to unsubscribe <<
[prev in list] [next in list] [prev in thread] [next in thread]