Skip navigation.

Gnome, Mozilla and the Challenge of Longhorn

Gnome 2.x
Gnome 2.x

With lots of talk recently about mono/java/python, it seems like a response to Longhorn and XAML is brewing. People at mozilla are talking about how to create an alliance against the hegemon. Co-operation with gnome is discussed on mozillazine. I do hope that gnome can use the XUL language or something like it so that gui programming is easy for us mere mortals. Lets not re-invent the wheel reproducing XAML when we've already got XUL!

Re: Want to fight? Package a SOLUTION

1. Find a way to end the GNEOM/KDE war for good. Maybe the market just lets one win. Maybe we build an interoperability layer between them. I'm not sure, but the environments are now only superficially different but the energy wasted is incredible.

As someone that is pretty much KDE/Gnome agnostic - either one is pretty much fine for me even though I'm using Dropline right now(anxiously awaiting 2.6) and loving it - I wish one of the desktops would've "won" years ago, but that doesn't seem to appear very likely anytime soon. I think the best we can hope for is better interop between KDE/Gnome gtk+/qt.

2. Ignore some viable languages. There are a dozen excellent languages with robust libraries out there, clustered into groups: systems languages (C/C++), large application languages (Java/Mono), scripting languages (perl,python,ruby). Distros should install one toolkit per group (exception: C++ comes free with gcc's C support).

C is the gnome systems standard and as you pointed out it C++ comes with GCC anyway. As far as Mono/Java and the scripting languages it would be a lot easier to just write compiler for all of those(in fact I think there are or are in development) that target IL(Mono) bytecode. Java is bit trickier because of the nature of it's bytecode, but the point is you want one runtime to rule them all.

3. Pick a browser. Hint: Konqueror - you lost. Konqueror has 10% of the market share of Mozilla and none of the momentum. I don't care about Apple, their direction is their own and they have "picked" already.

Although I think it makes sense to have mozilla as a core and just develop derivates from it, I do believe it's just a browser and not something to get religious about. The KHTML guys will have to follow Mozilla anyway if some XUL/XAML open standard is ever set.

4. DOCUMENTATION. Ed is right when he says we need something like MSDN. Finding good programming guides for some of this stuff is like panhandling. Maybe you get something, maybe not.

MSDN might be a mess, but it's a big mess with lots of info. KDE clearly wins over Gnome when it comes to developer docs.