[MEP]: Avispa Research Group as maintainer of the constraint system(Camilo Rueda)

[Camilo Rueda]

Dear all,

It seems by the reactions we got that our MEP was less concrete than 
expected. Our working idea
is that since nobody is actually maintaining the constraint system may 
be someone should.

Our short term plan is simple: 1) acquiring a detailed knowledge of the 
internals of the Mozart system and
2) integrating two constraint systems: real intervals and soft 
constraints.
For the first one, Raphael pointed out that

 > Regarding knowlegde transfer, it's a person-to-person thing.  You have
 > to contact the right people, and decide the best way to work together
 >...
 >To my knowledge, Christian Schulte, Kostja Popow (both in Sweden),
 > and Denys Duchier (France) are the persons to talk with.

We definitely agree. We had planned to invite Christian, Kostja or 
Denys here.
For the second one, as Raphael suggests, we will submit a MEP to this 
list for each constraint system.
Thank you all for your helpful comments.

Cheers,

Camilo, Juan


>
> Message: 3
> Date: Tue, 21 Jun 2005 09:26:08 +0200
> From: Raphael Collet <raph at info.ucl.ac.be>
> Subject: Re: [MEP]: Avispa Research Group as maintainer of the
> 	constraint system   (Camilo Rueda)
> To: hackers at mozart-oz.org
> Message-ID: <d98feh$lto$1 at grizzly.ps.uni-sb.de>
> Content-Type: text/plain; charset=us-ascii; format=flowed
>
> Great!  I am happy that you can now officially get involved in the
> maintenance of Mozart.  For several years the constraint system has
> lacked development and maintenance.
>
> However, I don't understand the purpose of your mep proposal.  I need
> some clarification.  I don't see what is your proposal.
>
> Please apologize my direct style.  Don't take it bad, it's only for
> launching some discussion.
>
> Camilo Rueda wrote:
>> Abstract
>> ========
>>
>> ...
>>
>> Clearly, transfering the constraint layer to the ARG is a sensitive 
>> issue
>> concerning the whole community. A knowledge transfer stage is a 
>> crucial
>> requirement for our objective of providing effective maintenance of 
>> the
>> constraint layer. ARG must acquire detailed knowledge about the 
>> low-level
>> structure of the language before attempting any modifications or 
>> improvements.
>> At least in the first stages, we expect a strong collaboration with 
>> the original
>> Mozart developers.  This is important for the knowledge transfer 
>> between the
>> Mozart developers and the Avispa Research Group.
>
> What do you mean by "transfering the constraint layer to the ARG"?  Do
> you mean more than knowledge transfer?
>
> Regarding knowlegde transfer, it's a person-to-person thing.  You have
> to contact the right people, and decide the best way to work together.
> For instance, when I redesigned the by-need synchronization, I went one
> week in Sweden to meet Christian and Kostja.  They taught me what I
> needed to know, and gave me very helpful hints.
>
> Many of the original developers have left Mozart.  To my knowledge,
> Christian Schulte, Kostja Popow (both in Sweden), and Denys Duchier
> (France) are the persons to talk with.
>
>> Rationale
>> =========
>>
>> -Discussion (within the Mozart community) regarding ideas on this MEP.
>>
>> -Knowledge transfer
>>
>> -Adding new constraint systems such as soft-fd and ri.
>
> I suggest you propose concrete MEPs for those systems.  I don't want to
> give acceptance to people, but to projects.  Even veteran developers
> must submit a MEP for new development in Mozart.  But of course, the
> more you get experience, the easier and faster your proposals get 
> accepted.
>
> Cheers,
> raph
>
>
>
> ------------------------------
>
> Message: 4
> Date: Tue, 21 Jun 2005 10:23:37 +0200
> From: "Christian Schulte" <schulte at imit.kth.se>
> Subject: RE: [MEP]: Avispa Research Group as maintainer of the
> 	constraint	system(Camilo Rueda)
> To: "'Camilo Rueda'" <crueda at atlas.puj.edu.co>,
> 	<hackers at mozart-oz.org>
> Message-ID: <001101c5763a$85fdf140$acd4ed82 at beule>
> Content-Type: text/plain;	charset="iso-8859-1"
>
> Dear all,
>
> I second Raphaël: what is the concrete plan? Why does providing 
> additional
> constraint system require a change of Mozart?
>
> Christian
>
> --
> Christian Schulte, http://www.imit.kth.se/~schulte/
>
> -----Original Message-----
> From: mozart-hackers-bounces at ps.uni-sb.de
> [mailto:mozart-hackers-bounces at ps.uni-sb.de] On Behalf Of Camilo Rueda
> Sent: Monday, June 20, 2005 9:57 PM
> To: hackers at mozart-oz.org
> Subject: [MEP]: Avispa Research Group as maintainer of the constraint
> system(Camilo Rueda)
>
>
> Please find attached a draft MEP which proposes
> that the Avispa Research Group takes the responsibility for the 
> maintenance
> and future development of constraint programming in Mozart. 
> Justification
> can be found in the MEP.
>
> Cheers,
> Camilo
>
>
>
>
>
> ------------------------------
>
> _______________________________________________
> mozart-hackers mailing list
> mozart-hackers at ps.uni-sb.de
> http://www.ps.uni-sb.de/mailman/listinfo/mozart-hackers
>
>
> End of mozart-hackers Digest, Vol 18, Issue 24
> **********************************************
>