Re: GNOME, .Net and Mono

[Miguel de Icaza <miguel ximian com>]

Hello Havoc!

> Miguel's comment was "he would like to see GNOME 3 based on Mono" -
> he's just saying what he would like. The headline of the article is
> misleading, no one knows what GNOME 3 will be like.
> 
> Whatever the GNOME community decides about Mono, I'm very sure we'll
> take this warning seriously:
>   http://www.joelonsoftware.com/articles/fog0000000069.html

Right.  

I was not trying to force Mono on anyone.  The only reason why I thought
that GNOME would move to Mono on the long term is because our goals are
aligned:

	* Mono will integrate natively with Gtk+ and GNOME.  

	  The team has seasoned GNOME hackers on board, and one of our
	  objectives is to reduce the cost and time of development of
	  future GNOME applications.

	* We will do our best to make sure that Ximian can use this for
	  its future products.  But still, this is up in the air, for 
	  example, Ettore is not yet convinced about using Mono.

	* Mono will have to stand on its own merits, and people will be
	  free to pick it up or not.

GNOME 4 looked to me like far along that people by that time would
realize `this is a neat platform, lets use it'.

That being said, I do not want to suggest that people rewrite any code
to inter-operate with Mono.  It is the other way around.  Mono will have
to provide all kinds of facilities to work well with GNOME.  For
example, since we have a dynamic code generator, exposing GObjects to
the Mono world is important.

But the real reason is that .NET (and hence Mono) was designed to be an
evolutionary path for applications.  Rewriting applications (lets say,
Microsoft Office) for a new platform is something that nobody is really
thrilled about.  For most projects, a new version is not about cleaning
up code, but about addressing user needs.  And a user does not need the
code to be rewritten.

So the way you use .NET is to "extend" your application with chunks of
.NET.  Take IE6 for example, it is still largely a C++/COM application,
but a few pieces have been already written in .NET (my guess is that
their SOAP support is written using .NET).  But that is the general
idea.

For example, it would be silly to rewrite Gnumeric or StarOffice with
C#, that would be a complete waste of time.  But one really useful
feature for Gnumeric would be to have its VB interpreter running on
Mono. 

If we manage to deliver Mono, it should serve to extend existing
applications, or as a platform for new applications.

best wishes,
Miguel.