discuss-gnustep
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [GSWHackers] Re: OGo/GNUstep cooperation Re: Re[2]: Frameworks integ


From: Helge Hess
Subject: Re: [GSWHackers] Re: OGo/GNUstep cooperation Re: Re[2]: Frameworks integration
Date: Wed, 3 Mar 2004 00:11:17 +0100

On 02.03.2004, at 23:49, Fred Kiefer wrote:
I think you are wrong here, Helge.

OK, good to hear some other voice on the issue ;-)

A common development of a WebObjects clone would be beneficial to all sides,

That is what I would assume but which doesn't seem to be a general agreement on - at least if you take into account that this for obvious reasons implies that on several things only one of the two base implementations can be used (given that there is a 95%+ match in terms of API).

but the questions that Manuel raised on the organisational form are all valid and must be solved, no matter who takes part in that project. So I would call them productive suggestions. I would expect, that if you accept them as this, a solution to them wont be too far away.

OK.

Lets analyse what we have: two LGPL frameworks. So do we have major organizational issues? No. In the worst case we can maintain two variants of the same codebase, license is not an issue at all, both frameworks are free software in all respects relevant to FSF (which also nullifies any rants regarding company or marketing issues).

We have the issue of copyright assignment. There will be no FSF copyright assignment of SOPE code. If this is a must for gnustep-web people, we are already stuck. Hopefully its not. And after all copyright assignments are *not possible* in Germany, and I guess the same is true in France (and the whole EU).

This is IMHO thought too far ahead, but even after a merge there could be two repositories - one for gnustep-web and one for SOPE. The important thing is that code is shared, at least a 95% sharing should be possible given that the APIs should match.

So after all I still (as written in other mails) recommend that we should start cooperation slowly. Currently I think the best way is too see what gstep-web and SOPE applications are available and try to port them to either environment by adding missing functionalities and fixing API incompatibilities. More useful and reliable comparisons regarding stability and performance of respective code sections can be made if we have the same, real-world, application running on both servers.

For some weird reason I haven't been able to locate a single, LGPL/GPL application based on gnustep-web - but there must be some, I already asked and wait for input. I would also encourage gstep-web to try to port OGo or other applications part of OGo CVS to gnustep-web (eg SimpleScheduler in Misc might be a starting point). This would require merges of several parts of SOPE with gnustep-web to import the missing functionalities.

I think thats a good plan to get started. To actually make that plan happen, we need: a) work on the gnustep-web side to see what is required to make SOPE applications
   run on that
b) input regarding existing gnustep-web application, so that SOPE developers can
   start working on that from the other side

best regards,
  Helge
--
http://docs.opengroupware.org/Members/helge
OpenGroupware.org





reply via email to

[Prev in Thread] Current Thread [Next in Thread]