Wine, WIDE & Unix (was: Support for pkgconfig)

[Dimitrie O. Paun]

On April 25, 2003 04:53 pm, Francois Gouget wrote:
> > So, can someone please tell me a *single* advantage of having KDE & QT
> > as the GUI platform for Linux vs. having WIDE?
>
> I mentioned many but you seem to be conveniently ignoring them.

No, I'm not ignoring anything. You referred to KDE/QT/GNOME/GTK, I argued
precisely against KDE/QT. Once again, I think it is unacceptable to have
our _platform_ (not apps) covered by the GPL. It sets an uneven playing
field. I'll ask again: is this the kind of free OS we want to build?
Yes, QT is licensed under QPL as well -- go read it, you can't develop
commercial stuff on it unless you pay the Trolls money, its puts them in
a monopolistic position. I hate to say this, I like KDE, I've used it for
years, but if it were to kill GNOME, it will kill Linux with it. I don't
want to change one monopoly (MS) with a far worse one (Trolls) (and BTW,
it's worse not because MS is a better company than the Trolls, but rather
because it puts the Trolls in a much more powerful position to control the
platform).

> At the same time you dismiss out of hand solutions that are being worked
> on, exist and are usable now, just because they are not perfect. 

Look, I'm not dismissing any effort. I gave the L&F as an example, maybe
it wasn't the best. I agree that we don't need WIDE to solve that problem. 
But those are the easy ones to solve. What about OLE? Forget about all the 
Win32 software out there, what about our Linux apps? Right now you can't 
match GNOME and KDE apps. After much work and flamming, they start to look 
alike thanks to BlueCurve. But that is skin deep :) There's no forseeable 
solution to the hard problems, and I don't see how we'll be in any better 
shape 5 years down the road. For the user of the system, this is just a 
big minus. Period. They don't care about religious issues like GTK vs. QT.

> But you are essentially proposing to add yet another toolkit/desktop 
> environment and IMO this requires having very strong arguments showing
> the existing stuff *cannot* be fixed and that the new stuff will not 
> suffer the same issues and will be significantly superior.

You see, that's my problem -- it's not yet another toolkit, it's the
toolkit witch is used by the vast majority of the software out there.
OLE and COM may suck, but it's the standard used by most software.
We can't make our apps interoperate in our small universe. But if you
look at the larger picture, how do you make all software interoperate? 
Tell me it wouldn't be cool to be able to run Word under Wine, and embed
Gnumeric in it, instead of Excel. Now that's what I call interoperability.

-- 
Dimi.