First pre-event meeting: Summary and plans

[PeterKraus]

Dear all,

Thank you very much for attending the First pre-event meeting. The minutes, recorded by Peter, are here. The discussion was already very interesting, highlighting several possible topics and focus areas for our in-person workshop in April. As Giovanni Pizzi highlighted, it's important to use these pre-event meetings to narrow down the tasks, and come up with a plan, so that we can achieve a deliverable or a practical outcome during the 3 days of face-to-face meetings in Berlin:

Focus area: Infrastructure

• RDF seems to be, conceptually, the way to facilitate the exchange of data.
• The details of the implementation of the RDF models (i.e. XML vs JSON-LD) need to be agreed upon.
• We should be building on what's already implemented in the RDM tools which support flavours of RDF.
• Not everyone should have to be an RDF expert: we need to build an "API" that allows end-users not to worry about RDF.

Focus area: RDM tool interoperability for samples

• Focus on exchanging sample information (metadata) between RDM softwares (i.e. machine-machine, not user-machine)
• Building on the agreed infrastructure solution above.
• Annotating samples - mutable/immutable, consumed/available, computational/experimental
• Dealing with exchange of sample history

Case study: Electrochemistry

• Several members present have experience and interest with electrochemistry (incl. battery research & electrocatalysis)
• Wide variety of sample types, including "battery cells" (sample entry corresponds to one item) and "materials" (sample entry corresponds to a divisible amount of material)
• Proliferation of binary, vendor-specific data formats
• Large number of individual samples - battery cycling is often done on 100s of individual cells

Case study: Catalysis

• Several members present come from an experimental catalysis background
• Strong dependence of sample behaviour on sample history
• Lack of adoption of standardised vocabularies/ontologies
• Comparably poor level of programming / IT expertise in the field: developed tools have to be simple

One possible outcome is, as suggested by Eibar Flores, building a knowledge base of datasets, formats, and metadata from the tools we are using. We can get started with gathering the pre-existing knowledge already. We have opened four separate discussion boards for each of the above areas here on github on the links in the headings above. We would like to invite everyone to contribute annotated links to example datasets, and discussing the common practices of how we currently deal with samples in our labs.

Plan:

• A Second pre-event meeting will be agreed upon using a doodle poll, in Feb. 2024

  • Discuss submitted datasets and knowledge-base entries and find common grounds
  • Propose work packages for the workshop

• A Third pre-event meeting will be agreed upon, in Mar. 2024

  • Schedule work packages in the workshop