Re: A GNOME Bindings release set?

[James Henstridge <james daa com au>]

On 25/11/2003 3:37 PM, Murray Cumming Comneon com wrote:

> Mike Kestner wrote:
>  
>
> > On Mon, 2003-11-24 at 04:40, Murray Cumming Comneon com wrote:
> >    
> >
> > > That does not include non-developer platform API such as 
> > > libgnomeprint, or libgda. It's great if you wrap those, but 
> > >      
> > it needs 
> >    
> >
> > > to be done separately. So if you currently bundle lots of stuff 
> > > together in one package, you need to put the extra stuff in 
> > >      
> > a separate 
> >    
> >
> > > package.
> > >      
> > Why?  I don't understand why we would need to force people to 
> > fragment their bindings.  Can you explain the reasoning 
> > behind this requirement?
> >    
>
> So that people can more clearly identify what is a binding for the GNOME
> developer platform, and so that we can more clearly identify whether they
> are following the API/ABI rules. Modularity is good.
>
> The number of tarballs used to distribute the source code of a binding
> should have very little effect on developers. I encourage people to
> encourage people to create umbrella packages on distros to make it easier
> for people to get their bindings.
>  
I would have to agree with Mike here.  How exactly does the inclusion of 
gnome-print binding in the same tarball as other gnome developer 
platform bindings make a difference?

If Mike's C# bindings are anything like the Python ones, then the 
gnome-print bindings would be exposed to the programmer as a separate 
module, so they would know if they were using gnome-print or not.  
Surely it is fairly obvious that no part of a language binding is any 
more stable than the library it wraps.

I am sure that GTK# and gnome-python would not be the only bindings 
inconvenienced by this sort of (unnecessary) requirement.

James.

--
Email: james daa com au
WWW:   http://www.daa.com.au/~james/