Why I Request ROR IDs for New Crossref Members

Collin Knopp-Schwyn of Crossref (one of ROR’s governing organizations) shares how they began contributing to the ROR dataset and discusses the value of ROR users providing curation requests to the registry.

This story begins, as so many stories do, with a mistake.

In my role as a Member Support Specialist, I spend a good chunk of my time reviewing new applications to join Crossref. These are submitted by universities, scholarly societies, commercial publishers, and a variety of other organizations interested in registering digital object identifier (DOI) records for their published materials.

One day, I noticed that a colleague had set up a membership for a Vietnamese university, Ho Chi Minh City University of Industry and Trade. The university’s name looked familiar to me, but when I queried our internal membership database I couldn’t find a separate account with the same name. I headed to ROR but couldn’t find an existing ROR ID for the university. Still, I couldn’t shake the feeling that something was missing.

After much poking around on the university’s website, I found a news release indicating that the university had changed its name from Ho Chi Minh City University of Food Industry to Ho Chi Minh City University of Industry and Trade the year before. Under its old name, the university did already have a ROR ID … and an existing record within Crossref’s membership system.

How ROR helps Crossref’s membership team

Capturing previous and alternative organization names in Crossref’s membership records helps us avoid situations like the one I encountered with this university, where we inadvertently create a duplicate record for the same entity under two different names. Generally speaking, each organization should only have a single Crossref membership, so additional information like past names, locations, and web domains helps us disambiguate between organizations with similar names or identify when a single organization may have been known under different names over its lifespan. ROR records are tremendously useful for identifying abbreviated versions of publishers’ names or the names of universities in languages other than the ones submitted by membership applicants.

After resolving the matter of this university’s duplicate Crossref accounts, I submitted a curation request to the ROR team to update the university’s ROR record with a more complete list of their past names in English and in Vietnamese. This was easy to do; after all, I’d already located the university’s press release in which they outlined the full history of their past organizational identities, so all I needed to do was share this information with the ROR curation team via their handy curation request form.

In the two years since this case of duplicate identity, I’ve submitted hundreds of additional requests to ROR. Many of these arise directly from this process of evaluating new Crossref membership applications. I begin by querying ROR when a new university, society, funder, or government entity applies to join Crossref, then I’ll submit a ROR curation request if the applicant organization is eligible for inclusion in ROR and not yet present in the registry. Likewise, if an existing Crossref member with a ROR ID gets in contact to let us know that their organization has changed its name, I try to pass this update along to the ROR curation team after updating their Crossref membership details.

The value of community curation

I grew up editing Wikipedia, which was the first place I encountered persistent identifiers like DOIs. Wikipedia editors, in their infinite wisdom and mild good humor, concocted the WikiGnome, a characterization for any editor making incremental and unostentatious edits to the encyclopedia.

The work of helping curate a clean, current, and comprehensive ROR dataset often feels Sisyphean, but I can’t pretend I don’t enjoy it. I cannot say why exactly, only that making these contributions scratches an itch in my brain. Like the WikiGnome, I relish the gradual, well-considered addition of new records to the ROR dataset. The regular ROR version release schedule and the nimble work of ROR’s curation team helps keep the registry fresh and reflective of real-world conditions. When I add a curation request for a new Crossref applicant, they will often have a ROR record and ID ready and waiting for them by the time they’re ready to begin registering DOIs for their published materials.

ROR’s community-led curation model allows for gnomelike contributions like mine as well as for more systematic contributions from funding agencies, governments, and other interested parties worldwide. ROR offers a great deal of value to me as a Crossref employee for how it augments and improves the organizational information we maintain. It feels only fair that I’m able to return a fraction of that value to the registry through record updates and additions. After all, the data is usually already right in front of me when Crossref members submit it; all I need to do is pass it along in a format the ROR team can use. In turn, this helps the exemplary members of Crossref who provide ROR IDs as part of their affiliation and funding metadata more completely and accurately represent the totality of organizations supporting the research that they register Crossref metadata records for.

My hope is that anyone who also finds ROR useful within their workflows is similarly able to give back to the registry when they identify gaps and inaccuracies. My journey with ROR began when I encountered outdated data that led to a mistake, and it continues with ensuring that the same mistake doesn’t happen again for anyone else. A perfectly “complete” registry is a beautiful, impossible dream, but the more that ROR users are able to keep the dataset up to date and enriched with relevant new additions, the more useful ROR will be for everyone – and the fewer mistakes we’ll all make.