[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Reply to: [list | sender only]
Re: [magCIF] magCIF dictionary v1
- To: firstname.lastname@example.org
- Subject: Re: [magCIF] magCIF dictionary v1
- From: David Brown <email@example.com>
- Date: Fri, 04 Jul 2014 11:45:20 -0400
- In-Reply-To: <372EFF30-BE48-46D7-8298-76E5EBC8DB3A@digis.net>
- References: <372EFF30-BE48-46D7-8298-76E5EBC8DB3A@digis.net>
Comments on the circulated copy of magCIF v1.|
1. This is a good start. The most important item at this stage is the 'description'. It might also be helpful in some cases to include other items, such as 'units', that complement the description. The first and most important step is to focus on the concepts that are being defined and making sure that 1) they are precisely defined and 2) the concepts are the natural concepts of the discipline and that they work together logically. On first glance it appears that the v1 of the dictionary is doing a good job here.
Other items are less important at this stage. 'Analogies' and 'related items' do not appear (at least in that form) in the DDLm dictionary, but they may be useful at this stage in drawing attention to the relationships between different items, though ultimately if this information is important, it should appear in the 'description', or else in the 'methods' which are the formal algebraic statements that allow the item to be calculated from other items appearing in the CIF (see the example from the coreDDLm dictionary below). Methods are not required to be present and can be added later if desired. Aliases do not have to be included at this stage as every item with a name _blah_blah.blah will have an alias _blah_blah_blah. In general, the less information that is included in the early stages of dictionary preparation the easier it will be to focus on the important items.
2. In the following item note that all allowed combinations of S, M and A will have to be listed individually in the dictionary, adding another 12 items unless the order of the letters is prescribed.
#aliases _atom_site_refinement_flags_magnetic #values "."->"no constraint on magnetic moment", "S"->"special position constrain on magnetic moment", "M"->"modulus restraint on magnetic moment", "A"->"direction restraints on magnetic moment", "combinations of S, M, and a"
3. I note that cartesian coordinates are being defined. These are also used in the coreDDLm because the geometry 'methods' in this dictionary are all based on cartesian coordinates. I show below the item from coreDDLm that defines the cartesian coordinate system and note that it appears to be different from convention proposed for magCIF. Unless there are important reasons for doing otherwise, I would recommend using the convention already defined, otherwise it will be necessary to define a second set of cartesian atomic coordinates for use in magCIF. As a personal aside, I do not understand why we still use cartesian coordinates when computers can work more efficiently using crystal coordinates. The use of different cartesian conventions is a recipe for confusion, but old habits die hard. Those who insist on using cartesian coordinates must take responsibility for avoiding the confusion caused by using different conventions in a discipline-wide standard.
Extract from coreDDLm:
Matrix used to transform fractional coordinates in the ATOM_SITE
category to Cartesian coordinates. The axial alignments of this
transformation are described in _atom_sites_Cartn_transform.axes.
The 3 x 1 translation is defined in _atom_sites_Cartn_transform.vector.
x' |11 12 13| x | 1 |
( y' ) Cartesian = |21 22 23| * ( y ) fractional + v| 2 |
z' |31 32 33| z | 3 |
The default transformation matrix uses Rollet's axial
assignments with cell vectors a,b,c aligned with orthogonal
axes X,Y,Z so that c||Z and b in plane YZ.
With a as atom_sites_Cartn_transform
_atom_sites_Cartn_transform.matrix = [[a.mat_11, a.mat_12, a.mat_13],
[a.mat_21, a.mat_22, a.mat_23],
[a.mat_31, a.mat_32, a.mat_33]]
begin:vcard fn:I.David Brown n:Brown;I.David org:McMaster University;Brockhouse Institute for Materials Research adr:;;King St. W;Hamilton;Ontario;L8S 4M1;Canada email;internet:firstname.lastname@example.org title:Professor Emeritus tel;work:+905 525 9140 x 24710 tel;fax:+905 521 2773 version:2.1 end:vcard
_______________________________________________ magCIF mailing list magCIF@iucr.org http://mailman.iucr.org/mailman/listinfo/magcif
Reply to: [list | sender only]
- [magCIF] magCIF dictionary v1 (Branton Campbell)