[Kde-games-devel] libkdegames

[Jaison Lee]

On 8/12/06, Martin Heni <martin at heni-online.de> wrote:
> > this is the perfect time to try and tackle things. Basically, I was
> > looking at the KDE 4.1 timeframe for an overhauled kdegames, unless we
> >...
> > until 4.0 started slowing down a bit to bring it up and see what
> > Kulow's plans are for kdegames into the future.
>
> This is fine for the games but we should actually start discussing libkdegames
> as we have to keep compatibility in mind. I assume there will be again a
> source or even binary compatibility police in KDE 4.x which will make it much
> harder to make design changes in libkdegames.

Well, I suppose this is true. I will hopwever point out most of the
classes in kdelibs that I see are "Not guaranteed to be binary
compatible across releases." :)

Mostly however, I don't see a choice here. kdegames as a module is
lagging behind the others and it has more work that needs done than
the others (at least more things that it can benefit from (such as
much better graphics support)). Most developers are working on kdelibs
and kdebase right now instead of secondary modules, and I simply don't
see kdegames making vast improvements unless we hold up the release of
KDE4 as a whole, which I don't see happening.

I don't see why in this case we couldn't break something for 4.1 and
then impose the freeze. It would be interesting to know how many
out-of-source games link with libkdegames tho. Anyone have any ideas?

It would also be helpful for me to stop being so vague I suppose. Here
are some thoughts I've had in my head, all of which are still in the
early, unformed stage so comments are welcome, but don't flame too
badly if I've overlooked something.

LIBKDEGAMES STUFF:

I think a lot of work can be done with the network support,
specifically taking advantage of the Internet. I would like the
existing network support enhanced to add support for Internet gaming
so you can play games with other people around the world. There are
numerous ways of handling this and even existing technologies that can
be utilized (GSZ Gaming Zone) but of course they all have benefits and
drawbacks so this will require a lot of preparation to determine what
exactly we want to do.

I also think it would be cool to have the HighScore classes send the
data to a server somewhere so you can see how you do against others
around the world. Care would need to be taken to minimize the impact
of fraudulant entries, which I don't think we will be able to
completely eliminate. Again, there are also pre-existing systems we
can take advantage of.

Make KCardDeck utilize GetNewStuff, especially since cards should be
made to use vector graphics, and cool new designs could become
available.

I think more card-based stuff could be moved down to kdelibs instead
of each program utilizing their own homebrew. Rough examples:

Deck::type() can set it to Standard, Pinochle, etc.
Deck::shuffle() will do what you'd think.
Deck::draw(count) will create /count/ signals with random drawn cards.
Deck::jokers(bool) Deck::cardsLeft() etc. etc. etc.

Is 3D support of any kind needed? There's no 3D anything right now.

GENERAL KDEGAMES STUFF:

More multiplayer games (I would like a Hearts game), making sure to
utilize the above enhancements.

Use GetNewStuff for downloading new levels, skins, etc.

The kdeedu guys had a good idea sometime before about a unified
front-end, although I'm not sure what happened to the idea. It struck
me as something that would be cool for kdegames, too.

Graphics updates. Check out blinKen. Now THAT'S a good lookin' game.  :)