[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: GNUstep Window Manager (was RE: Idea)

From: Nicola Pero
Subject: Re: GNUstep Window Manager (was RE: Idea)
Date: Sun, 7 Jan 2001 02:53:09 +0100 (CET)

> > > well, I have thought about that a little bit in the past... It would 
> > > certainly
> > > a big plus to have a WM written in ObjC which could offer tight 
> > > integration
> > > into GS with less effort. 
> > 
> > Ahm - I'm not sure having a WM written in ObjC would make such a big
> > difference.  WindowMaker is an excellent piece of software and I'm quite
> > proud it's part of gnustep.  Certainly, it doesn't make sense to take the
> > effort of rewriting all of it just because we want it in ObjC.  If someone
> > has all that time to devote to coding, why not write a brand new gnustep
> > application in ObjC - that would be much more useful.
> Well, you did not get my point, then...;-) I do certainly not want to 
> reinvent the wheel! What I would like to see is a window manager which
> offers tight integration with GNUstep. This window manager should:
> - be fast
> - have a small mem print
> - be responsible only for drawing and handling windows
> - offer virtual desktops
> - *not* include extensions like a dock, config tools and so forth!

If you configure window maker appropriately, it already does all this. 

My main point is (I am repeating myself here) that it is much more useful
to write new gnustep applications rather than spend time attempting to
write a new window manager in Objective-C.  I'm under the suspicion that
you don't realize that writing a good window manager is a completely non
trivial task.  If you invest the same effort into producing an end user
application, you will benefit the project much more and you will have much
more success.

This pushes me into expanding my thoughts about the importance/priority of
writing applications...

At this point, I think we have the following top priorities on the gui

 * work on support for binary packaging (RPMs, DEBs).  When we release
   gnustep base 1.0, we want it to be easily available in binary form on
   every gnu system on earth.  The gui library - and some gui applications
   - are also quite ready to be packaged in binary form and shipped
   around.  In general, we want every gnustep-related project controlled 
   by the make package to be easily packaged and shipped everywhere.

 * finish Gorm and Project Center.

 * improve the gui text system - rewrite the layout manager and more work 
   on that.

 * write applications, applications and applications.

 * more work on the gui library, interaction with window managers or
   whatever - work mainly driven by the needs of applications. <exception:
   bug fixing, in particular a certain set of very delicate bugs related 
   to the scrollview/clipview which has resisted all our debugging efforts
   up to now; some missing parts such as the color panel etc> 

 * write more tutorials on writing applications.
As you guess, IMO the real top priority now is simply producing end-user
applications/products available in binary form; most of the tasks, if not
all, are simply subprojects related to this.

We have spent *years* to build our marvellous Objective-C libraries. 
*Years* of work of many of the nicest hackers around, and contributions
from all over the world, are all piled in the source code of our gnustep
libraries.  To finish these libraries, we have consumed an enormous amount
of hacking work, and produced mostly nothing usable, from a user point of
view, for years.

No surprise that we have lost support and mind/market share in all these
years when the users had to wait without end.

Now we have the best libraries around.  Again, no surprise considering the
amount of work invested in them. 

After years on work concentrated on the framework, we really need to stop
(for a while only :-) dreaming about rewriting all the system in
Objective-C :-) and face the task of using our marvellous libraries to
write end-user products running now on what is available now. 

Our golem (call it frankestein if you like) is finally alive and
operative.  Let's give to the world examples of its power then. 

We have the advantage of the best environment; we have the disadvantage
that in the meanwhile we have lost most of our original support and
mind/market share, and other free projects are now much bigger than us. 

And still, we are a lot and have a lot of supporters and contributors, and
the project is actually growing now faster as a result of the fact that
the gui libraries now work.  And the advantage of the better/best design
makes a real difference in this field (free software) so I see the future
for gnustep as very happy. 

Of course, we need the contribution of everyone.  Writing free gnustep
applications or tools is - now - the best way to help gnustep and make it

reply via email to

[Prev in Thread] Current Thread [Next in Thread]