Tentative agenda for COMCIFS closed meetings Aug 28th and Aug 29th

[Herbert J. Bernstein]

Dear Colleagues,

   Might is be possible to see a document on the new DDLm prior to the 
meeting?

   Shouldn't we be having three separate discussions and possible votes: 
One on DDLm as currently posted, one on DDLm as newly presented, and one 
on CIF2?  The first of these could be done on the 28th.

    Regards,
      Herbert
=====================================================
  Herbert J. Bernstein, Professor of Computer Science
    Dowling College, Kramer Science Center, KSC 121
         Idle Hour Blvd, Oakdale, NY, 11769

                  +1-631-244-3035
                  yaya at dowling.edu
=====================================================

On Tue, 16 Aug 2011, James Hester wrote:

> Dear COMCIFS members:
> 
> Below is a tentative agenda for the two COMCIFS closed meetings.  Note that the Aug 29th meeting will
> occur directly after MS89, at which Nick Spadaccini and Syd Hall will be presenting a revised version of
> DDLm.  I expect that presentation to provoke considerable discussion, which is why the Aug 29th meeting
> will be entirely devoted to CIF2 and DDLm discussion.  Please advise if you would like anything added to
> the agenda on Aug 28th. 
> 
> COMCIFS meeting agenda: Aug 28th
> 
> 1. Brief report from chair on developments in previous triennium
> 2. COMCIFS directions over next triennium:
>  (i) reforming dictionary management (see also Brian MacMahon's email of July 6th)
>  (ii) developing a solid software base (APIs, canonical parser libraries in C/C++)
>  (iii) other concerns to be raised by members
> 
> COMCIFS meeting agenda: Aug 29th
> 
> 3. Discussion of DDLm as presented in MS89
> 4. Subject to outcome, if any, of (3), *final* vote on CIF2 syntax draft as distributed by email last week
> 5. Any other business
> 
> ================
> Explanation of item 2(i):
> 
> I propose partitioning dictionaries that are managed by COMCIFS into three groups:
> (i) the 'core' group which are small enough and/or general enough for COMCIFS members to be able to
> approve both content and technical conformance;
> (ii) a 'delegated' group where COMCIFS delegates management of dictionary content to an outside body or
> dictionary management group; once COMCIFS has voted to delegate this management, no further votes would be
> required to approve dictionary content or updates, but technical correctness would be checked;
> (iii) a 'contributed' group where COMCIFS makes no statement as to scientific content and simply checks
> technical conformance.  This group would correspond to those dictionaries that Brian was discussing in his
> email, and technical conformance would mean a simple syntax and DDL conformance check, not requiring a
> vote at COMCIFS level.
> 
> Explanation of item 2(ii):
> 
> Over the years COMCIFS has had various initiatives to produce and promote CIF software, but we are still
> at the stage where it is difficult to find a standalone CIF parser written in C that could serve as a
> starting point for integration into other languages, scripting or compiled, or could be dropped into a
> larger software project.  Part of such an effort would be development of an API.  I propose that COMCIFS
> set itself the entirely modest goal of developing and approving an API and a modular C parser before the
> next meeting.  We will discuss the best way forward.
>  
> --
> T +61 (02) 9717 9907
> F +61 (02) 9717 3145
> M +61 (04) 0249 4148
> 
>