Further exchange with the DFG and NFDI directorate on future community integration and onboarding

To advance the understanding and developing its role DataPLANT exchanged ideas with Ms. Kerremans from the DFG and Mr. Sure-Vetter from the NFDI Association on future community integration and onboarding of new participants. The topics we had already talked about e.g. at the NFDI Infratalks were discussed further: long term development and sustainability of the NFDI and the onboarding process. DataPLANT was able to place some points and present ideas that need to be developed further step by step, e.g. by putting it into a strategy paper. In addition to it, there were aspects that still needed some thought/formulation and that should certainly be discussed:

Blatt_blau.pngActual workflow/process when setting up a research proposal (coming from the various NFDI communities): Implementing a short interaction and review process to produce a NFDI endorsement on it (e.g. in the form of half page for reviewers to acknowledge previous interaction).

Blatt_blau.pngIn this (application) process the type of further involvement with appropriate scoping will be defined: embedding (tight integration), consulting (in some aspects), not relevant (no suitable common ground; referring to local support).

For this data stewards were seen as important building blocks of such future workflows. Overall, new approaches to the DFG proposal process were considered favorably, and reviewers of the various types of grant applications should be briefed accordingly to improve the common understanding of the process. Nevertheless, applicants (and supporting consortia) are still obliged to justify the individual steps (access to resource pools, consulting, services, data stewards). Additionally, it was seen important to distinguish between the different requirements and expectations of individual SFBs, GRKs, (project proposals):

Blatt_blau.pngSubject-specific (standardization, infrastructure, workflow support, ...) via NFDI/subject consortium.

Blatt_blau.pngDirect local support for university infrastructures and resources (more general RDM, training, local helpdesk).

Blatt_blau.pngRequirements arising outside these two areas (e.g. necessary special developments) should still follow the classical INF model.

To further the discussion, the topic will be discussed again in more detail in the Baden-Wuerttemberg Science Data Center working group on "Business Models" on behalf of the sibling project BioDATEN.

Further on the suggestions of DataPLANT's onboarding procedure got agreed upon and refined:

Blatt_blau.pngSince there are no (re)distributable funds involved, no formal process via DFG is required.

Blatt_blau.pngDataPLANT will continue to follow the already established path formalizing the connection through Letter-of-Interest (by the applicants) and Letter-of-Support (by the consortium) or similar formats.

To support this electronically, DataPLANT plans to integrate a "Become a member" button/page into the Science Gateway. A further advancement could be, that the information on participation is pushed into the ORCID profile of the researchers. This should help us to better communicate with the community.


DataPLANT participating in the first NFDI consortium assembly

Today the first meeting of the consortium assembly of the NFDI association took place online. In this meeting the nine initial consortia kicked off another integral part of the NFDI governance. A couple of important personnel decisions were made. The Chair (Christoph Steinbeck of NFDI4Chem) and Vice Chair (Dirk von Suchodoletz of DataPLANT) of the Consortium Assembly have been newly established. These two offices were not previously provided for in the articles of the NFDI Association, but were deemed useful by the Consortium Assembly. They are to be defined in the rules of procedure of the body.

The following persons were then elected as experts of the consortium assembly for the scientific senate:

  • Barbara Ebert (NFDI4BioDiversity)
  • Robert Schmitt (NFDI4Ing)
  • Torsten Schrade (NFDI4Culture)
  • Christof Wolf (KonsortSWD)

A key point of the gathering was the discussion of section concepts, especially after the announcement of the new guidelines of base services consortia by the DFG. Based on a joint strategy process, four cross-cutting topics were highly prioritized and working groups submitted a concept for each. These concepts are to be finalized in the coming weeks and then submitted to the Scientific Senate for establishment. We will keep you further informed in this regard in a timely manner on the ongoing process and the consequences for DataPLANT. The next assembly is scheduled for the beginning of next year.


Choosing and deploying a data repository for DataPLANT

A service developer and administrator in DataPLANT Jonathan Bauer from the university of Freiburg gave a presentation on the current state of the deployment of InvenioRDM for the Science Data Center coordination working group on infrastructure. The Invenio repository software is getting production ready and will be deployed as a DataPLANT service for data publication. KeyCloak is providing authentication services and bwSFS ("storage-for-science" in Baden-Württemberg) offering the storage backend (in the form of an object storage). At the moment the service installed as single instance but a Kubernetes cluster installation is envisioned in the long run to serve larger amount of data and users. Authentication in Invenio is possible via OpenID connect and OAuth, it offers built-in ORCID login.


Invenio is evaluated for a while already within the context of BioDATEN project and got chosen for its fitness for the purpose of open data publication. Records in the repository could be individually controlled for data and meta data. An embargo (holding period) can be applied to a record which is not publicly available until the embargo has ended. Special viewing and edit links can be generated and shared. A range of persistent identifiers can be applied, primarily support for the DataCite DOIs. It is possible to specify existing DOIs or mint new ones. Additional related identifiers can also be specified like e.g. ARK, arXiv, handle. The integration with 3rd-party services is exposed through REST APIs. This enables a lot of potential integrations with other services like auto-update profile on new publications, integration with the Science Gateway of BioDATEN for e.g. discovering or the publication of templates. For Github or GitLab webhooks could be used to publish code releases in a fixed version. We hope to publish the ARC specification in its first version on it in the near future.