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

List:       kde-core-devel
Subject:    Re: API problems (Was: future versions)
From:       Maksim Orlovich <mo002j () mail ! rochester ! edu>
Date:       2004-02-06 15:49:45
Message-ID: Pine.GSO.4.58.0402061019020.331 () mail1 ! ats ! rochester ! edu
[Download RAW message or body]

On Fri, 6 Feb 2004, Waldo Bastian wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On Fri February 6 2004 12:03, Martijn Klingens wrote:
> > On Thursday 05 February 2004 15:54, Waldo Bastian wrote:
> > > I'm not in favor of too much revamping for KDE 4.0 because that tends to
> > > give application developers only more work to do without getting much in
> > > return. It will take longer for development books to catch up, etc. etc.
> > > Having to make changes for Qt4 sounds already bad enough.
> >
> > On the other hand, carrying along bad API for a prolonged timeframe will
> > hurt us just as well. I think we shouldn't refrain from refactoring known
> > weak spots (e.g. stuff already marked deprecated). We should just not
> > refactor for the sake of refactoring...
>
> So then the question becomes which API parts are so bad that they cause real
> hurt?

Well, from questions on #kde-devel, it seems there is at least some
confusion about the number of socket classes around, as people aren't
quite sure about what to use. However, that can probably be addressed by
moving the less-favored classes into a compatibility lib, and making
documentation/includes separate as well.

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