Discussion List Archives

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: Please consider the following new default values.

James,

These defaults should ABSLOLUTELY NOT appear in the dictionary. They were deliberately omitted from the original dictionaries because they are in no way guaranteed to be correct. Anyone looking for room temperature structures must know the difference between structures where the author states that the experiment was done at room temperature and those where the temperature was not given and might have had any value. It is impossible to do science properly if you are being fed information that you cannot rely on. An absorption correction may have been applied; the crystal has a high probability of not being white if the author fails to indicate the colour. There are many crystal structures reported without much of this information, and just because an author fails to say that he or she used Mo radiation does not mean they used copper. I did a study once on anisotropic atomic displacement parameters and had to reject most of the values reported in terms of B (rather than U or beta) because there are two conventions, and most authors did not say which convention they used. If I had been presented with an assumed default I would have had to reject all the results with the assumed value because I could not rely on its authenticity.

If default values are needed, they should be supplied by the application software that reads the CIF. Presumably it has a better idea of what the default should be as well as the effect of that such an assumption would have on the subsequent use of the information. The default value should always be ?, meaning that nothing is known about the value. Even the default . (period) would be wrong, since the dictionary cannot know whether an item is inapplicable or just not reported.

None of the suggestions in the list should be implemented. If something is unknown, it is, surprisingly, not known and should not be guessed at by a dictionary that knows nothing about the measurements, even if the dictionary gets it right much of the time.

David

I. David Brown
Professor Emeritus
Department of Physics and Astronomy
McMaster University
Hamilton, Ontario, Canada

From: coreDMG [coredmg-bounces@iucr.org] on behalf of James Hester [jamesrhester@gmail.com]
Sent: December 13, 2017 00:52
To: Distribution list of the IUCr COMCIFS Core Dictionary Maintenance Group
Subject: Please consider the following new default values.

Dear Core DMG,

A number of new default enumeration values were introduced in the new DDLm dictionary which are not present in the old DDL1 dictionary. Could you please consider whether or not they are reasonable and comment either here or at Github.  A lack of comments will lead to Antanas and I likely erring on the side of caution and removing those that are not obviously correct.

I reproduce Antanas' list and comments below (html format). If this is unreadable, please view the original at https://github.com/COMCIFS/cif_core/issues/58.

James.
===

1. Experimental

ID Data name Default
1.1 _exptl_crystal_density_meas_temp 293.
1.2 _exptl_crystal_density_meas_temp_lt 293.
1.3 _exptl_crystal_density_meas_temp_gt 293.
1.4 _exptl_crystals_number 1
1.5 _exptl_absorpt_correction_type none
1.6 _exptl_crystal_colour_primary white

The _exptl_crystals_number and _exptl_absorpt_correction_type seem reasonable, but should
exptl_crystal_density_meas_temp_* and_exptl_crystal_colour_primary have default values at all?

2. Atom sites solution

ID Data name Default
2.1 _atom_sites_solution_hydrogens difmap
2.2 _atom_sites_solution_primary direct
2.3 _atom_sites_solution_secondary difmap

3. Diffraction radiation

ID Data name Default
3.1 _diffrn_radiation_wavelength_determination fundamental
3.2 _diffrn_radiation_probe x-ray
3.3 _diffrn_radiation_xray_symbol K-L~3~
2.4 _diffrn_radiation_source tube
3.5 _diffrn_source_target Cu

Note: the _diffrn_radiation_source data item has only become an enumerator value in the DDLm version of the dictionary. It is more thoroughly discussed in issue #59.

4. Reflections

ID Data name Default
4.1 _refln_[A|B|F|F_squared]_calc 0
4.2 _refln_[A|B|F|F_squared]_meas 0
4.3 _refln_[F|F_squared]_sigma 0

5.Diffraction reflections

ID Data name Default
5.1 _diffrn_refln_scan_mode ot
5.2 _diffrn_refln_scan_mode_backgd st

6.Chemical

ID Data name Default
6.1 _chemical_enantioexcess_crystal_technique CD
6.2 _chemical_enantioexcess_bulk_technique OA
6.3 _chemical_absolute_configuration .

7.Refinement

ID Data name Default
7.1 _atom_site_refinement_flags .
7.2 _atom_site_refinement_flags_adp .
7.3 _atom_site_refinement_flags_posn .
7.4 _atom_site_refinement_flags_occupancy .

Seems reasonable since the lack of these data items imply that the related techniques were not applied thus resulting in a '.' (inapplicable) special value.

8.ADP

ID Data name Default
8.1 _atom_site_adp_type Uiso

Seems reasonable.

9.Crystal system

ID Data name Default
9.1 _space_group_crystal_system triclinic

This addition seems a bit strange. Might it be related to the removal of the default cell_angle* values (discussed in #57)?

10.Bibliography

ID Data name Default
10.1 _journal_page_first 1
10.2 _publ_body_element section
10.3 _journal_index_type O
10.4 _citation_coordinate_linkage yes

--
T +61 (02) 9717 9907
F +61 (02) 9717 3145
M +61 (04) 0249 4148
_______________________________________________
coreDMG mailing list
coreDMG@iucr.org
http://mailman.iucr.org/cgi-bin/mailman/listinfo/coredmg

[Send comment to list secretary]
[Reply to list (subscribers only)]