Re: Report on activities

Dear James,

I was at the ACA meeting in Chicago last week and I thought it useful to bring you up to date on my discusions and plans.  I discussed three different dictionary projects with various people.

REstraints/constraints
Ilia and I picked up earlier on Herbert's suggestion and have already drafted an addition to our proposed restraints dictionary.  In addition to incorporating Herbert's suggestion, we decided that we would use the opportunity to include the possibility of a rigid body restraint as well as constraint as this was not possible under our original proposal.  This turns out not to be trivial as listing the weighting of the restraint depends on how people define the rigid body geometry in their program.  However, in discussing problem with various people, including George Sheldrick, it appears that a restrained rigid body is generally not considered rigid.  SHELX makes no provision for anything other than a constraint.  Ilia and I will be working through our draft in the light of this discussion and submitting our proposal to the core CIF dictionary mainetance group for approval.

Magnetic CIF dictinary
I had discussions with Branton Campbell.  It appears that there has not been much progress on the Symmetry-2 dictionary, but Branton was interested in adding CIF items for magnetic structure description.  There are problems with this that have defeated earlier attempts to define such a dictionary, but we may be able to come up with a basic version that could later be extended. We are currently exploring this possibility.

DDLm
I discussed the state of the DDLm project with both Herbert and John Westbrook.  The project seems to have gotten bogged down in somewhat (from a dictionary point of view) irrelevant detail and Herbert was urging that we get a basic dictionary out there, warts and all, and see where the weaknesses are when we tackle real problems.  John agreed, but is of the opinion that the real strength of CIF lies in the definitions in the dictionaries.  We can play around with the syntax as we like, since the software can always be changed.  As a result of these discussions I will focus my efforts on getting a basic version of the coreCIF dictionary in DDLm to Simon and encourage him to experiment with it.  This strategy means that I do not need to worry about, e.g., looping methods beyond the level that Syd inroduced, and this should speed up getting DDLm cifs into their real-time software trials.  Such trials will no doubt reveal problems we never dreamed of, and people can continue using regular C(Fs until we are ready with a foolproof version of DDLm CIF dictionaries for distribution.

Best wishes

David

_______________________________________________
comcifs mailing list
comcifs@iucr.org
http://scripts.iucr.org/mailman/listinfo/comcifs