Skip to content

Meeting Notes 01.02.2021

Jump to bottom
Casper Welzel Andersen edited this page Feb 1, 2021 · 1 revision

Present:

  • Emanuele Ghedini
  • Jesper Friis
  • Casper Andersen
  • Joana Morgado
  • Rickard Armiento
  • Saulius Grazulis
  • Stuart Clark

Starting with a comment about the newly created PRs by Jesper (#13 and #14), specifically #14 regarding the incorporation of CIF labels in ontological concepts.

Following this, Saulius, Jesper and Rickard discussed how to define the concepts of properties describing a unit cell in ontologies, symbolic versus descriptive parts.
Saulius: It’s a symbol, it’s an approximation. Need to single out the approximations that are useful for crystallographers and define them rigorously.

The discussion turned towards naming conventions.
Saulius: Use something different than A, B, C for FractionalPosition due to convention-confusion.

Emanuele: One can use EMMO to express equivalent concepts. This can be used to import all CIF concepts and match it up. One can make a sub-class of language (and Crystallography) as a CIF Crystallography Language, fill it up with the CIF conventions and match things up internally in the ontology.

Rickard: We should define the naming conventions we want to use, prior to moving forward and delve into more minute and detailed conceptual aspects.
Saulius: We should avoid confusions at all costs, sticking as closely to IUCr (CIF) as possible. Simply starting there and then moving on to what is missing that we intend to describe.
Jesper: Can we describe “a single He atom alone in the universe”. Saulius: No, this is not what CIF is intended for.

Chat comments from Emanuele:
Agree 100% with Saulius. Facilitating the user to draw the connection between EMMO/CIF concepts to real data is extremely important. My suggestion is to use CIF names as skos:prefLabel for Language/Crystallographic/CIF subclasses, since it is exactly what they are. There are already duplicated concept between CIF and EMMO, like _chemical_formula_iupac and the IUPACNomenclature EMMO class. My suggestion is to declare _chemical_formula_iupac and then we will deal to draw a relation between those two classes. Agree also with Rickard, that we need understand the implementation degrees of freedom to put CIF into EMMO and then to commit to one.

Clone this wiki locally