Collaboration between CLARIAH Tool discovery and related initiatives #128
Labels
discussion
This is a discussion point; invitation to discussion
FAIR Tool Discovery
FAIR Tool Discovery
planning
Preparation and planning activities
Milestone
In this issue I'd like to track potential and ongoing collaborations related to
our CLARIAH Tool Discovery track.
Codemeta Community
First of all I'm seeking collaboration with the wider codemeta community and
have raised various issues on their issue tracker to express certain software
metadata aspects that could hitherto not be properly expressed in the existing
codemeta/schema.org vocabulary, mostly related to software interface types and
data input/output types. This has led to an extension on top of codemeta, in
collaboration with @dgarijo:
https://github.com/SoftwareUnderstanding/software_types. Codemeta and
schema.org are fairly slow moving targets, so an independent extension was the
quickest way to get results. It could eventually be merged into codemeta
and/or schema.org if proven interesting/popular enough.
Research Software Directory (eScience)
I met with @maaikedj and @jmaassen from the eScience center last week.
We discussed possible collaboration between their Research Software
Directory (source,
instance) and our work. Where we focus fully
on automatic data harvesting and a unified data representation, they have built a
fully-fledged CMS where users can express their software metadata. They do use
some smart automatic harvesting to help the content editor auto-fill various
properties and connect with various standards (including codemeta, citation.cff).
They are currently working on a new version of the RSD (due in september) which they will also
offer "as a service" (for institutions that can't or don't want to self-host).
We agreed that collaboration between our two projects could be beneficial for
both parties. The focus here would be finding common ground on the data
representation and the vocabularies used (LoD, codemeta, schema.org). I invited
them to pitch in in any of the vocabulary discussions we have (e..g
#32). If we have
sufficient agreement on data representation and vocabulary we can envision an import/export
function in the Research Software Directory which enables connectivity with our work:
codemeta.json
file in their CMS (for inclusion in the source code), as opposed to having to edit JSON-LD by hand as is the case now.ODISSEI also expressed an interest in the Research Software Directory. As
ODISSEI and CLARIAH are merging in the proposed continuation project in the
future, this may be an extra incentive for cooperation.
KNAW DANS
In April, I spoke with Wim Hugo (DANS KNAW) who also expressed an need/interest
in expressing tool metadata and data types. They are looking into possibilities for an
operational tools and type registry that will be compatible with work done to
date by CLARIAH. Part of this we might accommodate with our software types extension, but their requirements probably go further than this.
DANS also works on Dataverse (@4tikhonov), where they are also working on supporting codemeta.
CLARIN Switchboard
I already worked with the CLARIN Switchboard in the past and have it on my
radar to realize a degree of convertability of our software metadata
descriptions to the format their tool registry requires.
This is tracked separately in issue #36.
@antalvdb suggested to arrange a meeting with @dietervu to discuss further collaboration here.
CLARIN VLO
Inclusion of CLARIAH software in the CLARIN VLO would be facilitated by export to CMDI (@JanOdijk). This is tracked separately in issue #37.
Ineo
Progress to connect with Ineo (CLARIAH-internal) is tracked in issue #35 (@Seb-CLARIAH)
Conclusion
In all these collaborations I would like to emphasise the most important goal is to come to
agreements on common data representations and vocabularies, facilitating
exchange between different tools (and providing the necessary implementations).
Let's track any progress on these and other collaborations in this issue. Feel free to comment on anything or suggest additional possibilities!
The text was updated successfully, but these errors were encountered: