Developing Identifiers for Heritage Collections: Q&A with Rebecca Grant (F1000) and Frances Madden (British Library), authors of the Open Educational Resource

Persistent Identifiers (DOIs, handles, ORCID IDs, Wikidata IDs etc.) became essential resources in our increasingly noisy digital workplaces. Beyond enabling linking instead of losing our resources, PIDs usually come with secure storing, versioning and long-term archiving solutions too. Although assigning PIDs to our scholarly outputs became a no-brainer on the level of project and data management guidelines, in reality, the selection and implementation of PID systems are anything but straightforward – especially in working conditions where the use of Zenodo or other research data repositories is simply not the best solution. In this post, we are talking with Frances Madden (British Library) and Rebecca Grant (F1000) on occasion of releasing a new Open Educational Resource which provides guidance on developing and using PIDs in the context of heritage collections. 

Hi Rebecca, hi Frances, and thanks for joining us! Could you start off by telling us a bit about what motivated the creation of your training resource, Developing Identifiers for Heritage Collections? Who are the target audiences and why? 

Frances: As part of the Heritage PIDs project we wanted to help organisations understand what they needed from PIDs as the various types of PIDs offer different functionalities. The resource is aimed at colleagues working in cultural heritage organisations. Due to the huge variation in size and staffing models across those types of organisation, we were deliberately not too prescriptive in the target audience. When we originally designed the resource, we specified ‘technical decision makers’ within those organisations, but during the consultation it became clear that this was not necessarily applicable to all audiences for the resource so we have kept it more general in the revised version. 

What is the biggest challenge you see regarding the implementation and use of PID systems in the Cultural Heritage sector? 

Frances: From the findings of Heritage PIDs, the biggest challenge to implementing PIDs is in the diversity of the options available and the lack of cohesion across the sector. The cultural heritage sector is diverse in its technical capability and collections. There are pockets, such as natural history collections, where PIDs have been developed to address a specific use case, e.g. CETAF Stable Identifiers, which are not true PIDs in that they are not globally resolvable (i.e. they can only be accessed by the URL provided by the institution which creates the ID) but they do provide long-term accessible links to collection items which can be used to cite collection items. 

For smaller collections particularly, it is going to be difficult to make much progress without the buy in of the suppliers of the collection management systems cultural heritage organisations use. There was some pilot work in the Netherlands which developed an integration between a handle server and several major suppliers of collections. 

And what is the biggest challenge regarding their use in citations?

Rebecca: I think publishers are quite reliant on citation standards to guide authors towards correct formatting. With the introduction of data sharing policies on journals, it may be less clear to authors what exactly should be cited, how and where, for instance, citing heritage materials, serving as primary resources alongside data that researchers produce on top of them. Projects like Heritage PIDs represent community-supported standards for the citation of heritage collections; this in turn helps publishers to supply appropriate guidance in our author guidelines. 

Frances: For encouraging the use of PIDs to end users it is vital that the citation is made easy and becomes a convention. This requires all stakeholders from journal policies, editors, curators, museum websites etc. to all encourage the use of PIDs in citation. 

Having access to high-quality digital or digitized cultural heritage collections is in many cases a primary precondition of making arts and humanities data open and FAIR. Still, cultural heritage data and research data operate in two different silos when it comes to policies, institutions, standards etc. How can we bridge this gap? 

Rebecca: I see publishers’ research data sharing policies as being the ideal “test-bed” for the crossover between digital cultural heritage and research data. When publishers require authors to share their data openly, as on the Routledge Open Research platform, it may be the first time that authors have encountered the FAIR Principles or thought about aspects like open licensing and persistent identifiers. We have integrated the guidance from Heritage PIDs into our research data policy for Routledge Open Research, and we hope that it will guide our authors towards using the most appropriate citations for the cultural heritage collections they use in their research. 

Frances: For me colleagues like Rebecca who see the similarities are vital. Her work to draw policies together and make them cohesive at least across one publisher could really help move the conversation forward. Steps like those of Routledge Open Research really illustrate the benefits of PIDs and drive their adoption, making a contribution towards understanding collections as data, and more importantly as an open data source. 

A practical question: The training material has been developed on a GitHub repository. Can you tell us a bit about your experiences with that, pros, cons? Do you have a sustainability plan in mind for the resource? 

Frances: We developed it using Github based on the simple site framework which one of the project co-investigators, Joe Padfield had developed. It is a very quick and easy way to set up webpages using Github pages. We chose Github with a view to transparency because it allows us to work openly. It also allowed us to capture feedback in a simple way and we took advantage of the integration between Github and Zenodo. We also knew it would be a useful format for when the project ended and we wished to hand it back to the community. Now that the project has finished, we have guaranteed to maintain the resource until the end of the Towards a National Collection programme in 2025, but only on a best efforts basis. We hope other members of the community, perhaps those working on the next phase of TaNC projects, will be willing to take on its maintenance in the future. 

Thank you for your time! It has been great to learn from your insight and experience!

You can access the OER here:

Kotarski, Rachael, Kirby, Jack, Madden, Frances, Mitchell, Lorna, Padfield, Joseph, Page, Roderic, Palmer, Richard, & Woodburn, Matt. (2021). Developing Identifiers for Heritage Collections. Zenodo. https://doi.org/10.5281/zenodo.5205757 (Needless to say, it comes with its own DOI)

Read more about how the Heritage PIDs project is aiding the provision of better guidance to authors on the F1000 blog.

---

---