Re: Where is Gnome heading?

[Daniel Kasak <dkasak nusconsulting com au>]

sardaukar siet wrote:

> Is there a 3.x coming anytime soon?

Changing the major version of a product usually implies a break in 
backwards compatibility. Perhaps a change from 2.x to 3.x for Gnome 
would go along with a change from gtk-2.x to gtk-3.x. That said, I'm not 
looking forward to 3.x, as I'm quite happy with the way our custom-built 
gtk2 software works.

> Where are the revolutionary ideas on the desktop?

Compiz? Enlightenment? I think these projects are pushing the 
revolutionary stuff on the desktop. I don't see gnome so much as a 
desktop 'product' as a collection of desktop software - probably because 
I don't use Gnome / Metacity directly ... I use Enlightenment, with 
gnome software. For me, the software means things like nautilus, totem, 
gimp, evince. I'm quite happy with the way these work, and I can't 
really see any revolutionary changes ahead for these products.

> Is maintaining such a large C-codebase becoming a nightmare to manage?

?

> If so, why not dump C?

I write all my apps in Perl ( gtk2-perl ), but when I get around to 
putting some finishing touches on my Perl projects, I'll be heading 
*towards* C, not away from it. C seems to be the choice of language for 
Linux developers. Maybe for simple desktop stuff like photo management 
etc it doesn't exactly make sense to write the *GUI* in C, but it 
certainly makes sense to write the *libraries* in C, and once you've got 
developers writing libraries in C, then they're free to write apps in C 
too. People are free to write competing software in other languages.

> Why not start a GNOME-3 project and start adding experimental code and 
> features to it?

What's with the pre-occupation with the number 3? :) There is enough 
experimental code slipping into so-called stable releases. For example, 
I'm having to rebuild quite a bit of software on a number of desktop 
machines ( ie downgrade to gtk-2.8.x ) at the moment because of some 
interesting bugs in gtk-2.10.x's treeview stuff. What sort of 
experimental stuff are you looking for anyway?

> Please consider embracing C#

I don't think so. The minute people start rewriting bits of gnome in C# 
is the minute I jump ship. I think people are more than aware of the 
legal minefield Gnome would be wading into by embracing a Microsoft 
technology such as C#. At the *very* least, assuming they don't use 
their trademark dirty tactics to crush the competition, we would be 
playing an eternal game of catch-up. That's required for a project like 
OpenOffice, where MS Office compatibility is an absolute requirement to 
get people to use their product. But this requirement - C# compatibility 
- is NOT a requirement for gnome or for desktop linux. Why do we have to 
lend credibility and developers to their environment anyway? What's 
wrong with C for libraries and scripting languages such as Perl and 
Python for lazy people? Sure there are good reasons for not using C for 
everything. But there's a BIG difference between advising people against 
using C, and advising people to use C#. For example, Hell will freeze 
over before I use C#.

--
Daniel Kasak
IT Developer
NUS Consulting Group
Level 5, 77 Pacific Highway
North Sydney, NSW, Australia 2060
T: (+61) 2 9922-7676 / F: (+61) 2 9922 7989
email: dkasak nusconsulting com au
website: http://www.nusconsulting.com.au