Re: GNUstep and distributions...

[Riccardo]

Hey,

On Friday, November 12, 2004, at 04:19 AM, Chad Hardin wrote:

> I still think that what is really needed is our "own" distribution.  I 
> go back and forth all the time on whether I should start working on 
> simplyGNUstep again.  It's not an easy thing to do though, I have a 
> hard enough time finding the energy to work on the printing stuff in 
> GNUstep after work work and school take their drain on me.

although you may be right that it could be fun and useful, the best 
thing still is to give gnustep a way in existing environments, say in 
some linux or BSD distro, say in solaris or Aix or hp-ux or whatever.
Think about: you have your Suse box working and just for curiosity want 
to try out how to develop something fast. Or you have your New solaris 
Blade workstation and want to try out GNUMail or the filters of PRICE. 
GNUstep should coexist easily with the rest.

The hope is of course this will spur interest in GNUstep!

Also for some people there is no choice! I know some computers in the 
Lab MUST run redhat 7.2 or some highly paid apps won't work (ok, this is 
because linux world is crap, but no comments here). Or do you think 
someone that uses a Sun E350 as research workstation could be convinced 
to remove solaris and install simply GNUstep ?

As I often see in some FOSS projects, we forget that "my own laptop" 
where I can dual-boot, reformat and do waht I want is a very different 
environment than what is a real-world work environment where the 
computer administrator maybe isn't even the user.

We need to work for both.

> Anyhow, I do think that a GNUstep distro with a solid development 
> environment and some good apps would be very popular.  It might even be 
> be the basis for a profitable business.

Yes, that is a nice view for the future, however consider the thoughts 
above and also consider that there are few applications for gnustep, 
there is no critical mass. I even KDE 1 was richer !

IMHO in the shorter term we should focus on stabilizing gnustep and also 
give a uniform tree of applications. They do not need to be killer app 
(like an Improv clone) but they should give the idea of a more complete 
environment to the user. And be easily collected and installed on the 
various distros.
Also I think we have 3 very strong points
        1. portability (Linux, BSD, Solaris and I hope more in the future)
        2. easy interoperability with Macos-X
        3. good and efficient development environment

1. is a point that could prove interesting if joined with 2. if we can 
add windows in a smooth manner that would gain us many interested people.

2. is a strong point to exploit. If we have apps like GNUMail that work 
well in both environments, this is a good point

3. this is more for developers, but once they have seen and tried points 
1. and 2. they might get interested.

for point 3. I think one of the show stoppers is ProjectCenter. I can't 
compile and work with CVS versions since week on none of the computers I 
have! Add this to the changed format... it is a mess.
Gorm is on the very good way (compliments Gregory) and probably emacs 
from Adrian will be a nice gift to many (where is VIM for gnustep ???).


I think I put down a lot of work in only a few lines...

-R