Re: [RFC] [PATCH]: Icon themability patch for -GUI images/icons

[Nicolas Roard]

Le 31 oct. 04, à 20:41, Fred Kiefer a écrit :

> Hi Quentin,
>
> Quentin Mathé wrote:
> > Le 30 oct. 04, à 17:16, Fred Kiefer a écrit :
> > > Sorry, I am actually getting bored of this and most other 
> > > pseudo-discussions going on in the GNUstep mailing lists. Perhaps I 
> > > should think about dropping out of them.
> > Could you explain what you mean here ? … or tell us what are the 
> > pseudo-discussions you are talking about ? I would like to know to 
> > avoid such position on your side and have more friendly and 
> > progressive discussions.
>
> nice of you to ask. I was refering to the fact that we already had a 
> similar discussion on icon themes before and to the discussion on 
> anti-aliased fonts for xlib and to bits of the discussion on the 
> windows backend and ...

Well, the thing is, I think most of the discussions here where people 
argue produce little results -- in fact, if somebody code one of the 
options debated, generally this is accepted in the end... so.. if 
people want something, they just need to DO them. ;-)

> You see, with the change I made for AA as the default in xlib most 
> people that will see any difference are now worse off than before. (I 
> explained why in a mail already) Still some argued very hard to get 
> this default setting. I just don't understand them, and to be honest I 
> don't want to understand them any more. This is not about making 
> GNUstep a better environment.

while I'm not even using xlib, I can understand why people want AA by 
default. Alone it does not make gnustep a better environment, but I 
understand that it could be seen as a nice touch. Anyway, as long as 
things are easily modifiable with defaults...

> Or the statement in the discussion about what isn't working for the 
> windows backend. Someone, and it may be the same someone, claimed that 
> the pasteboard may not be working. He didn't even state, that it isn't 
> working, which just means that he doesn't really care. But than, why 
> bring up the topic? For me the windows backend hasn't been working for 
> various reasons over the past few months, still I could always tell, 
> what was actually broken and sometimes even how to fix it. BTW, it is 
> still broken for me for all applications which only start a menu and 
> not a window.

hm, I don't know if it was me saying that :-)
But if that's the case, my recent mail about the windows backend was 
more to encourage people to have a look on it, as it works better than 
generally assumed (even if things are missing -- and some very annoying 
ones .. ;-)
I know that if I have the opportunity, I will try to contribute to it.

> One basic pattern of these pseudo discussions seems to be that 
> somebody states, that if a single feature would be added to GNUstep, 
> than it would suddenly attract loads of users.

Well, that's generally a dellusion, but it's still true that some of 
theses features will help attracting users..

> Theming is one of these features that pop up from time to time. I am 
> not against themes and theme support in GNUstep.

...and I agree that theming is such a feature. In fact, proper theming 
is imho anyway necessary for GNUstep if we really aim proper 
crossplatform (read: blending as much as possible in the host system), 
regardless of attracting users that want to have yellow-stripped with 
red plots windows..

> GNUstep is already more themable than anyone, to my knowledge, 
> exploits. You may change all the system colours by supplying a 
> different system colour file. You may change most icons, by just 
> editing the nsmapping.strings file. You may change many aspects of 
> controls by overwriting the GSDrawFunctions class in a bundle. None of 
> this ever gets used.

I disagree a little :)
First, some of theses changes were from me, and explicitely to allow 
easier theming, yes. And you're right that currently, doing a theme on 
GNUstep is a LOT easier
than it used to be -- I know it first hand. But I disagree with you 
that anyone is using it -- as I am using it :-P
In my opinion, it's now fairly easy to write a "programmed" theme on 
gnustep (which was my first goal).
If you want to test a theme like 
http://www.roard.com/screenshots/screenshot_theme38.png , you can try 
http://www.roard.com/gnustep/SimpleUI.tgz (just compile it and add it 
to the default GSAppKitUserBundles array..).
Now my goal is to move Camaelon (which was for me more a testcase than 
a theme) to a proper theme engine, using pixmaps. It's in the work and 
things should be ready soon, hopefully. I think people don't use the 
themes "possibilites" of gnustep because they are basically too 
complex/hidden to the user. I think that a simple pixmap themes engine 
will fulfill the job in a simpler and better for the average user.
The other reason we don't have other "programmed" themes is because, as 
always, we lack developpers ;-) and in general, "programmed" themes are 
not that much popular, if I count the number of them on KDE/GNOME 
compared to the pixmaps one.

> It may be that I have been hanging around in this mailing list for to 
> long. The same arguments come up over and over again. And there is so 
> much serious work still untouched....

it's always the same thing: lots of discussion but nothing much happend 
until someone provide a code. Don't forget we're all volunteers and I 
guess everybody is doing what he can to improve gnustep. I wish I would 
do a lot more than the little things I did.

> For my own contribution to GNUstep, I am not that much proud on the 
> new features, that I implemented (although there have been a few), but 
> on the cleanup of code I did over the years. Now sometimes I even have 
> the feeling that GNUstep may be degrading again. I know this is not 
> actually true, still I get that feeling.

--
Nicolas Roard
"Any sufficiently advanced technology is indistinguishable from magic."
 -Arthur C. Clarke