Re: GNUstep Live on OSnews

[Xavier Brochard]

Le 31 juillet 12:11:51, vous avez écrit :
> On 31 Jul 2017, at 11:38, Xavier Brochard <xavier@alternatif.org> wrote:
> > As I've always wanted to work on a light desktop using GNUstep, I propose
> > to work on this :
> > Even if the project is not about building a desktop, a lots of components
> > are already present. My idea is to write a short how-to that can be
> > polished along the time. It should answer to such basic questions that
> > are evident for the team but not for the others :
> > - which Window Manager can I choose ? which one for that task ?
> > - how should I configure GWorkspace ?
> > - where can I find themes ?
> > - can I take some Etoilé components ?
> > etc.
> > 
> > Hence, starting with something light and extending it later. Nothing
> > related to development but to building a working environment. I think it
> > can help to attract devs because one can see what small apps / components
> > are missing, and start to develop using GNUstep framework and tools.
> 
> I see this as a losing battle. A GNUstep desktop will not be competitive
> with a Qt or GTK desktop unless GNUstep attracts a lot more developers and
> a desktop that is obviously not competitive will not attract mode
> developers.  We tried this with Étoilé and it didn’t work.  The best way to
> attract new developers is to show them that GNUstep *doesn’t* require them
> to throw away all of their existing investment.  Show them great apps
> written with GNUstep that integrate closely with their KDE or GNOME
> desktop.
> 
> There’s some great work in this direction.  For example, DBusKit allows us
> to integrate with other desktop services.  In an ideal world, we’d use DBUS
> to replace gdomap and gdnc on systems that are running DBUS already, rather
> than reinvent the wheel (it doesn’t matter that our wheel came first).
> 
> Better integration with other fd.o technologies would make the risk of using
> GNUstep a lot lower: if no one can tell that your app is written with
> GNUstep and it integrates seamlessly with their GNOME or KDE desktops, then
> you don’t lose market share for your app by going with GNUstep (and you get
> good a Mac port basically for free).
> 
> David

I don't think it's a loosing battle as long as it is kept light. LXDE, LXQt, 
and XFCE are successful while they offer far less "fun" than the big ones. 
Even EDE (http://equinox-project.org/) has success in the small FLTK world. 
Without forgetting Enlightenment... Also, remember that many distribs offer 
GNUstep as a desktop install option.

My purpose is only to to write some recipes to have various light usable 
desktops. For example WindowMaker + PCMan + TextEdit + ... or Fluxbox + 
Gworkspace + ... It doesn't have to be "full GNUstep" at the beginning.

To attract more developers, you can also be more attractive : offer something 
that is easy to try, release often the desktop, offer small tasks to work 
with, ... 
I don't see the desktop as a requirement to atract devs, but as something that 
can help, because people will try it and talk about it.

I don't know why Etoilé failed, but IMHO it was may be too ambitious, it 
couldn't release often and it was not easy to try (lack of packaging). 

Also, as a sysadmin I have many friends developers who ask to macOs 
compatibility. Then I talk about GNUstep but they ask "show me, show me 
something that works" and they mean show me a complete environment because 
they want to be convinced that everything will work, from file selector to 
theming.

Regards,
Xavier