search

COMCIFS / Powder_Dictionary

CIF definitions for powder diffraction
4 stars 4 forks source link

Update definitions of phase names #70

Closed rowlesmr closed 1 year ago

rowlesmr commented 1 year ago

Some definitions of data items recording phase names were of type Encode and Code; this doesn't represent what they do/are, which is Describe and Text*.

Also altered source for phase names of standards to be Recorded, otherwise Assigned.

* This is one case where it would be nice to have a _type.contents value defined as "case-insensitive sequence of CIF2 characters"

jamesrhester commented 1 year ago

I think I have to raise a question concerning Recorded. Generally names and identifiers are Assigned, as they are arbitrary and not measured. However, if the contents of a data name refer to a name or identifier that is assigned externally to the data set or measurement process (e.g. a NIST standard identifier or an experimenter's name) then that would be indeed Recorded. Have we been sufficiently clear that these data names take values that have been assigned elsewhere, and if so, where those names came from?

rowlesmr commented 1 year ago

To my way of thinking, if you're using a standard, you probably already know what it is and it probably already has a name (whatever that may be) before you use it. It could be the zincite that your mate down the road recommended, or the alumina your bought from NIST; you knew what it was before you started the diffraction measurement. Does it matter if you say "NIST676a Alumina bottle # 1234" or "676a", or "NIST corundum", or "Jim Bob's zinc oxide"; you knew what it was going in to the experiment.

Having said that, I'm not wedded to using Recorded, but I thought it was a pertinant change to make while I was in the phase_name-changing mood.

Have we been sufficiently clear that these data names take values that have been assigned elsewhere, and if so, where those names came from?

Probably not, for example, _pd_qpa_ext_std.phase_name says

Identifies the name of the material used as an external standard for quantitative phase analysis. This should match the value given in _pd_phase.name in the data block containing the calibration diffractogram.

rowlesmr commented 1 year ago

OK. I've changed my mind. I've reread the definitions of Recorded and Assigned, and now think that phase names are Assigned, as they:

represent a decision made that determines the course of the experiment (and therefore may be deemed PRIMITIVE)

I'll change the PR accordingly.