Rioxx: The Research Outputs Metadata Schema

Governance Group minutes 2022-06-01

Wednesday 01 June, 10:00

Attendees: Petr Knoth (PK), Mick Eadie (ME), Paul Walk (PW), Nicola Dowson (ND), Beverley Jones (BV), George Macgregor (GM - Chair)

Minutes of last meeting

GM welcomed RGG members. GM noted that several informal meetings had taken place since the last RGG meeting in January 2022. These meetings had been held to discuss practical issues surrounding the implementation of specific RIOXX updates and, as such, no formal minutes had been taken.

PW highlighted that the minutes of 27 January 2022 had yet to be published on the RIOXX website and that he had no record of GM emailing his usual .md file of the minutes. GM reported that minutes had definitely been produced but that he may have forgotten to send them to PW, as is customary.

Action: GM to locate minutes of 27 January 2022 and email to PW for immediate publication on the RIOXX website.

Brief update

GM reported on recent meetings with UKRI representatives about RIOXX, its current state of development, and its potential for satisfying many of the UKRI Open Access Policy technical requirements. The meeting (with Rachel Bruce, Anisha Ahmed and Tahia Zaidi) was productive and was an opportunity for a RGG representative to articulate the discovery and aggregation benefits of RIOXX. UKRI indicated that finalising the OA Policy had been a mammoth task which had distracted them from formalising the Policy's technical requirements but that they expected to publish a paper during the summer of 2022 which would seek to provide research institutions with further detail about the Policy's technical expectations. UKRI and GM agreed to keep in touch, with UKRI asking to be informed as soon as RIOXX v3.0 (final) was published.

GM also provided a report on a separate meeting with Jisc (with Azhar Hussain - AH). AH was positive about RIOXX v3.0 (beta) but was keen to see the final version published as soon as possible so that Jisc could evaluate if/how they could support institutions with adopting it. GM reported that there was significant discussion about the possible repository plugin requirements, which repository platforms should be targeted for such plugins, and how best to support commercial vendors. It was noted that both EPrints and DSpace had plugins for RIOXX v2.0 and, to this extent, a v3.0 plugin would not necessarily be beginning from scratch; however, that it would remain a significant piece of work. GM suggested that he would be willing to liaise with commercial vendors to assist them in their interpretation and implementation of the profile, e.g. guidance, testing, counsel developers, etc. AH requested that he be kept up-to-date of RIOXX developments, in particular the publication of v3.0 final.

PK expressed positivity about recent UKRI and Jisc meetings and envisaged an important role for CORE as a RIOXX validator service, perhaps as an extension of the CORE Repository Dashboard. As RIOXX would improve the quality and relational potential of harvested for CORE it was important to ensure repositories were deploying the profile as optimally as possible.

PW queried the timelines insofar as both UKRI and Jisc were large organisations and therefore required long lead times in order to respond to sectoral developments. PW therefore suggested that the RGG should be seeking to publish v3.0 final as soon as possible. This position was agreed by all members of the RGG. GM reported that he had, in fact, already informed UKRI and Jisc that the RGG would publish v3.0 as soon as possible.

RIOXX review

GM opened this agenda item by summarising that all recently discussed updates and changes to the profile, arising from recent informal meetings and email discussion, had been implemented. Changes had been implemented on GM's local GitHub repo for pushing to the official AntLeaf repository. For priority discussion GM highlighted several elements which had recently undergone most change. These included the RIOXX treatment of dc:relation but also rioxxterms:grant and rioxxterms:project.

dc:relation

RGG discussed recent changes to dc:relation. GM indicated that he was concerned about the growing number of element attributes as a result of the decision to encode JAV types. Such as:

<dc:relation type="https://schema.org/ScholarlyArticle" version="SMUR" deposit_date="2021-07-06" resource_exposed_date="2021-07-20">
    https://www.repository.org/article_123456_preprint.pdf
</dc:relation>

<dc:relation type="https://schema.org/ScholarlyArticle" version="AM" deposit_date="2021-07-28" resource_exposed_date="2021-07-28">
    https://www.repository.org/article_1234567.pdf
</dc:relation>

PK suggested that this was not necessarily an issue but expressed a view that, from an aggregation perspective, it is useful to know the location of the relations (i.e. are some relations deposited locally or simply referencing an external repository). A technical discussion of this issue ensued but no decisions were taken. After reflection, RGG members reversed the decision to encode JAV types as a dc:relation attribute to communicate manuscript/journal version types, on the basis that it had the potential to introduce conceptual confusion in the profile since rioxxterms:version uses JAV.

Action: GM to remove JAV attribute and associated profile guidance from v3.0 final.

rioxxterms:grant & rioxxterms:project

Recent changes to rioxxterms:grant and rioxxterms:project were also considered to have introduced conceptual inconsistencies, possible confusion, and data repetition, as a result of permitting reciprocal URI referencing between grants and projects. For example:

<rioxxterms:grant
    funder_name="Arts and Humanities Research Council"
    funder_id="https://ror.org/0505m1554"
    project_id="https://handle.net/10378.1/1590366">
    https://doi.org/10.35802/218671
</rioxxterms:grant>
<rioxxterms:project
    grant="https://doi.org/10.35802/218671">
    https://handle.net/10378.1/1590366
</rioxxterms:project>

There was some discussion surrounding the importance of RAiD in better communicating project and grant relations. Following discussion, the RGG agreed that the Group had been 'overthinking' these elements and that they should be simplified with the reciprocity removed, as per:

<rioxxterms:grant
    funder_name="Arts and Humanities Research Council"
    funder_id="https://ror.org/0505m1554">
    https://doi.org/10.35802/218671
</rioxxterms:grant>
<rioxxterms:project>
    https://handle.net/10378.1/1590366
</rioxxterms:project>

PW and PK noted some typos in the beta profile that required attention. PW also recommended that the RFC 2119 terms used needed to be doublechecked as, in some cases, OPTIONAL was possibly being used inappropriately.

Action: GM to fix typos. Action: GM to make the above noted simplifications to the profile and its documentation.

Dissemination

The need to promote and publicise the publication of RIOXX v3.0 final will be necessary. GM listed a blog post and email lists as the obvious dissemination channels but opined that a longer form paper or report might be beneficial. PW suggested UKSG Insights as a suitable destination, with the potential of reaching the right audience. RGG members agreed to work towards an Insights submission following the publication of v3.0 final.

The prospect of dissemination at a UKCORR Member Day was discussed. ND noted a recent UKCORR Committee meeting concluded that there was insufficient community appetite for a Member Day event but that the Committee was contemplating one-off webinars on specific topics as an alternative. RGG members agreed that such a webinar could be a useful dissemination mechanism, especially after more is known about UKRI's future OA Policy technical paper. It was therefore agreed that the RGG retain this idea for later in 2022.

Change of name

GM introduced this agenda item. He related several instances of individuals within the repository/scholarly communications domain who had questioned the name of RIOXX, either because they professed not to understand it or because they associated it with the now defunct RCUK. PK and GM had also recently attended a meeting in which one of the attendees indicated that they had an uncertain idea of what a metadata profile was and that the name of RIOXX did little to communicate its purpose. These comments demanded discussion from the RGG about whether a name change should be considered.

ND noted that a name change carried with it many pros and cons, particularly as RIOXX is already well-known under its current name. GM opined that, for him, RIOXX had transcended any association with RCUK but accepted that its description could be improved to better communicate its purpose. RGG agreed that a decision on a name change was not something that could be taken in the meeting but was one to be discussed 'offline', with potential candidates circulated to RGG members by email.

AOB

No other business was raised.

DONM

To be confirmed after completion of the above noted actions.

GM 10/06/2022