[Kde-games-devel] Kolf -> Playgroung

[Paul Broadbent]

Hi,

I'm technically the kolf maintainer, although I have very little time to put 
into it, and really should have resigned long ago and made the status of the 
game clear. Sorry about that.

On Monday 17 November 2008 22:39:46 Stefan Majewsky wrote:
> On Montag 17 November 2008 22:12:15 Albert Astals Cid wrote:
> > A Dilluns 17 Novembre 2008, Mauricio Piacentini va escriure:
> > > I do not think we can do this in the middle of a stable series (4.x),
> > > can we?
> >
> > Yeah why not, if it does not work and nobody is going to fix it we can
> > drop it imho.
>
> I'm not fully aware of Kolf's current state (last time I checked it was not
> so good), so I cannot comment on that. Apart from that, a move out of the
> module is a bold action. Is there something that was done or can still be
> done in advance?
>
> It would be nice if someone with some insight into the game could provide
> us with an overview. From what I know, the state could be anything from
> "has some major bugs which could still be ironed out in this or the next
> release cycle" to "absolutely mislead design, needs to be rewritten".

As other people have said the kolf code is a bit of a jungle. It was like that 
when I started, and when I did the port to Qt4 I wasn't a very experienced C++ 
programmer, so I undoubtably made the problem even worse. If I'd know better 
I'd have probably done a major re-write back then.

As Stefan pointed out, the ball collision code is particularly bad. One of the 
last things I did was a partial rewrite and tidy of that, which improved it 
somewhat, but it still has serious design flaws, and bugs. I tried for far too 
long to hack that code into a working condition, but never quite managed to 
get there. 

As for the re-write vs fix debate, I'm not sure. I don't think it is quite so 
bad that a skilled developer couldn't make significant improvements to the 
current code to iron out the bugs and make the game acceptable to the players. 
This would be an unpleasant, but probably not too difficult a process if they, 
unlike me, knew what they were doing. But at the end of the day anyone who 
worked on it in the future would still have to suffer the painful process of 
learning their way round the class design, so if someone has the motivation 
then a radical re-write is probably be the best move.

Paul
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.kde.org/pipermail/kde-games-devel/attachments/20081118/5c37f216/attachment.htm