Telecon (2019-08-07)

Return to Archive

Attendees

• Stevan Earl
• Colin Smith

Agenda & Notes

This agenda is largely informed by our 2019 ESIP Session (notes are here).

Agenda notes are italicized.

Refine IMCR scope

Uncertainty about project scope largely stems from an inability to prioritize use cases.

• Who are the primary constituents we are trying to serve and what are their use cases? - Information managers (or anyone playing an information management role) in the environmental and ecological sciences. We are focusing on this domain first since it's where the expertise of this clusters contributors are from. Once methods are formalized and robust, we will expand to other domains.
• Who are ancillary constituents and what do they need? - Ancillary constituents are machines that need detailed metadata for each software library function with annotation to ontologies.
• Prioritize this list, constrain scope to the current priority, and add future scope to a project road map. - Humans, machines.
• What science domain(s) are we serving? - Environmental and ecological first, then others.
• What level of software maturity does the IMCR contain? - Mostly production ready software packages, but pre-preproduction packages and code snippets are welcome. Our curation activities are focusing on production and pre-preproduction software packages first, then on code snippets.
• What do the registered software items represent (e.g. single scripts, packages, both)? - Both.
• Explicitly communicate this in a project scope. - Will do.

Improve software metadata

... so IMCR is useful to platforms like the Data Discovery Studio

• What level of detail needs to be added? - Function level inputs and outputs.
• How can we perform this task? - This may differ by implementation language and community of practice. This is nearly impossible to access programmatically for the R language. Perhaps this is simpler for others?
• How much effort will it require? - A lot.
• What priority does this task have? - Low. See prioritized list of action items below.

Reconsider controlled vocabulary structure and terms

The current controlled vocabulary is a customized construct designed to serve the non-expert data manager to the detriment of excluding others. Aligning the IMCR vocabulary with established vocabularies (e.g. SWEET, GCMD) would enable machine actionability and inclusion in platforms like the Data Discovery Studio.

• Do we want to continue with the custom vocabulary or align with established vocabularies (can we do both)? - We will keep the current scheme as it is one commonly held in the minds of our primary constituents. We'll map the IMCR terms to other vocabs when possible and consider integration at a later time.

Scraping and sorting the web for IM software

This is being done by others, whom we can collaborate with. - We'll check in with Ilya when this prioritized item comes up.

• Where does this fit into project priorities? - Low. Such an effort will require substantial effort on the part of current cluster participants for which such an activity is outside their current professional responsibilities. Additionally, this would dramatically change the complexion and direction of this clusters current scope and priorities. However, such a resource would be very useful.
• How much effort will this require? - This will likely require a fact finding expedition. Who will lead it? - A lot. We'll delegate a fact finder when the priority arises.

Action Items & Notes

Listed in decreasing priority:

Refine scope

Done. Should we link from the home page (for primary user group) to technical information (for secondary user group)?

Wrap IMCR in a simple but attractive website in preparation for production release

Done, website is here.

Update the IMCR vocabulary

... and consider aligning or mapping to the Software Ontology, SWEET, and GCMD

• SWEET - The SWEET Ontology may be useful for finer concept granularity, which is currently of low priority. Will revisit this later.
• SWO - The Software Ontology has terms useful to both immediate and low priorities. Immediate focus is on aligning "Operation", "Process", and "Data management" classes of the IMCR with SWO. Alignment notes are being kept in this document.. SWO alignment is complete.
• GCMD - Several concepts within NASAs Global Change Master Directory "Science Keywords" class are relevant to IMCR. Alignment with GCMD is complete.
• What vocabulary is DataONE using for their Software Tools Catalog? Can we align with it? A request for this information has been placed.
• The Kepler Project was contacted for information on the ontology used to categorize resources within that software workflow system, but the Kepler Project doesn't appear to have a maintainer and inquiries sent via the project's contact interface fails.

Other SWO classes to be used in IMCR:

• Use "programming language" for OntoSoft "Implementation language" fields. This has been added to IMCR best practices.
• Use "Data/Format" for OntoSoft "File format" fields. This has been added to IMCR best practices. This has been added to IMCR best practices.

Identify domain vocabularies for use with IMCR software

• The LTER Controlled Vocabulary "disciplines" category has been added to IMCR best practices to keyword science domains.

Back up IMCR to GitHub

Done. A GitHub repository has been created for backing up IMCR software metadata. The toolkit function "backup()" facilitates this task.