CCDC/IUCr journals recommendation 1: Recognition of scientists responsible for data collection

[James Hester]

For details see the original document attached to
https://www.iucr.org/__data/iucr/lists/coredmg/msg00345.html

In summary, IUCr Journals and CCDC propose using audit_contact_author loops
from January 2021 to determine the crystallographer(s) responsible for the
structures presented in those blocks.

My comments:

(1) While I know there is an ad-hoc practice of using data_global to hold
information that is relevant to all data blocks within a single file, I
don't think this practice is actually formalised. So for now I think the
recommendation can only apply to the data block within which the structure
is presented, as pointed out in the CCDC response.  Perhaps we could make
an exception if the CIF makes use of the _audit_link_block mechanism to
link the global block into the individual structure blocks.

(2) A CIF data block can contain information about anything that has CIF
data names defined. So if a CIF holds raw data, but no structure,
audit_author refers to the person collecting the data. If a CIF contains
both raw data and a structure (e.g. a powder diffraction experiment with
the resulting solved structure included) then the file was presumably
produced by the person solving the structure and so audit_author refers to
them.  So the principle is that "the person responsible for the most recent
content in the file is the audit_author."   This will be important to
emphasise for CIF software authors, as it means that they should not by
default carry forward the contents of _audit_author.

(3) I note that IUCr Journals propose to use both audit_author and
audit_contact_author, whereas CCDC will use only audit_contact_author.  Is
there any reason for this discrepancy?  By definition, audit_contact_author
will always be a single person, whereas more than one person could be
involved in preparing the data block contents (audit_author).  If there is
only one, and no audit_contact_author is nominated, then they by default
become the contact author. If there is more than one, then a contact author
must be nominated (who is not necessarily one of the audit_authors).  The
particular chain of logic to be employed here should be clarified for the
benefit of software and data block authors.

(4) I agree that there is a need for a new loop that allows roles to be
assigned to authors.  It will need to be a separate category to allow for
multiple roles for a single author, and would be simple to design but may
take time to choose the appropriate descriptors. Is there anyone interested
in taking the initiative on this?  If so, a new thread on this list can be
started.

Do others have any comments?

James.
-- 
T +61 (02) 9717 9907
F +61 (02) 9717 3145
M +61 (04) 0249 4148
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.iucr.org/pipermail/coredmg/attachments/20200309/fa695512/attachment.html>