Re: Approval of CIF core changes for 2.3

["I. David Brown" <idbrown@mcmail.cis.mcmaster.ca>]

On Tue, 15 Jul 2003, Howard Flack wrote:

> BMcM> I haven't deleted the Z_prime entry, though you have indicated
> BMcM> that you will pull it from consideration at this stage (though I
> remark that
> BMcM> there have been requests from other sources for a record of Z', so
> finding a
> BMcM> sound definition would be a useful priority for the next version).
>
>   I sort of agree.

I have now removed this item and placed it on the list for further
discussion.  CCDC has been alerted to give the matter some thought.

> BMcM> Nor have
> BMcM> I included either of Howard's alternative definitions for
> BMcM> _space_group_symop_operation_xyz
>
>   I'm not surprised. The problem is to know which one is correct. I did
> not look but perhaps there is some small print in the symmetry cif dic
> which makes it clear. I do not have time myself to look.

I have incorporated HDF's second description since the lattice centerings
should be explicitly included in the list.  If they were not, it would be
necessary to include an item specifying the lattice centering, and the
order of application of these centering translations would also have to be
defined in order to avoid chaos in the symmetry relationships given in the
geom category.  Information about the lattice centerings is given the item
_space_group_lattice_type which I did not transfer from the cif_sym
dictionary since this information is not defined in the current core
dictionary and therefore presumably not used.  Incidentally, all CIFs
currently submitted to Acta Cryst. include the lattice centerings
explicity and this never seems to have been an issue.

I had to read HDF's definitions through several times before I was sure
which definition said what. In an effort to be precise Howard has used
descriptions that will not be familiar to many crystallographers (e.g.
primitive and non-primitive translations - many practitioners, myself
included, do not normally think of the primitive (or is it non-primitive?
I am not sure which is which) translations as being part of the space
group since they are not listed in the IT-A tables.  Howard's the
description, while the kind of precise and correct description that we aim
for, is somewhat inscrutable, so I have added at the end of Howard's
definition:

' That is to say, it is necessary to list explicity all the symmetry
operations required to generate all the atoms in the unit cell defined by
the setting used.'

D




*****************************************************
Dr.I.David Brown,  Professor Emeritus
Brockhouse Institute for Materials Research,
McMaster University, Hamilton, Ontario, Canada
Tel: 1-(905)-525-9140 ext 24710
Fax: 1-(905)-521-2773
idbrown@mcmaster.ca
*****************************************************