The Persistence of Infrastructure

Letter from the Executive Director, February 2020

Last week, I participated in the fourth PIDapalooza meeting in Lisbon, along with Alice Meadows, NISO’s Director of Community Engagement, and 175 other enthusiasts for all things identification. This festival of persistent identifiers is an annual gathering of people who create, manage, and use identifier systems. There were great conversations about the implementation of a connected network of identification systems in Portugal, updates on a variety of PID system developments and interoperability issues, and even an interpretive dance of the interactions between scientists and the PID supporting organizations. Alice drafted a quick summary of the session and her thoughts on the rest of that conference. 

I was there to report on the new Draft ISO Principles of Identification (accessible only for NISO Members), a technical report being drafted by members of the ISO Subcommittee on Identification and Description (ISO TC 46/SC 9). Instead of your typical presentation, I styled the “PID Party” session after a pub quiz. That draft is currently out for ballot before the NISO membership in their role as the US-Technical Advisory Group (TAG) for TC46/SC9 — as it is also before other Participating Members of TC46/SC9. The purpose of this document is to set forth definitions and a set of ideal attributes that identification should embrace. These twelve attributes include such things as public availability of minimal metadata, persistence, and resolution. The Balloting closes on February 20, 2020, so please submit comments on the Principles. We hope that this document will provide a guide for the creation of new identifier systems, their revision, the ongoing maintenance, and the case for supporting their work in the community.

One of the principles in the Technical Recommendation is focused on identifier system sustainability. Indeed, the topic of financial sustainability and support came up repeatedly during PIDapalooza. Feeling somewhat like a broken record, I have written repeatedly over the years about the need for support of the infrastructure of scholarly systems. 

On the positive front on this topic, Laurel Haak, ORCID Executive Director, announced during the closing session of the meeting that ORCID had achieved break-even financial status. (Note: ORCID is a NISO Voting Member.) While ORCID has to continue to improve their financial status as it repays its initial funding loans and builds out a pool of retained surpluses to be permanently stable in its finances, ORCID is well on its path to doing so. This was certainly a happy moment for ORCID and is an important milestone for a system that was launched just seven years ago. Congratulations to Laurel and her entire team on achieving this milestone and sustaining the growth, with hopes for continued success.

This positive news was in counterpoint to the many other efforts in the community that continue to struggle. While there are a top-tier of successful identifier initiatives, such as ORCID, CrossRef, ISBN, ISSN, and a few others, many other infrastructure projects are constantly scrambling for resources, attempting to achieve what ORCID has accomplished. For example, the inventive FREYA project will be coming to a close later this year as its three-year EU funding period is coming to a close. Hopefully, many of the project’s outputs, such as the PID Graph, the PID Forum, their Ambassador Program, and several important identifier reports, will be maintained in some fashion by the project’s partners moving forward.

Other identifier projects lack the funding to advance their missions. Others are run as side projects in support of larger organizational missions, such as the identifiers.org site run by EMBL-EBI. It is troubling how precarious are some critical components of our scientific and cultural infrastructure. Others, despite having the force of need or demand, struggle to gather the resources needed to get themselves airborne. The Research Organization Registry (ROR) project, which held an update session the day before the start of PIDapalooza, is seeking development capital to build out the system. The backers of ROR are seeking two years of funding of $400,000, but as yet has only attracted one quarter of that amount.

There are many other examples, and while certainly not every idea needs to succeed, many have succeeded in a variety of ways, save in the matter of finances. During the last morning of the conference, there was a gathering of PID systems providers. The meeting’s purpose was to seek to build the capacity of the PID Community through coordination and, possibly the creation of an “Intergalactic Federation of PID Providers.” This (envisioned) coordinating body would organize the parties to advance common interests, such as certification, sustainability, interoperability, preservation guidance, and advocacy for resources. While many of these goals are laudable, as I participated in the conversation, I was less than enthusiastic. There is a significant need for momentum around many of the topics raised, but there are existing organizations and forums where these types of coordination can take place (including NISO, but also several others).

For example, there is a need for advocacy to generate additional resources for PID systems. What the PID infrastructure community lacks are the resources to support the organizational infrastructure necessary to support a supra-organization comprised of PID providers in advancing this. Funding, as currently configured, such as it is, does not support “infrastructure” generally; rather, it funds specific efforts in the community. Each provider is too busy advocating for itself and searching for resources to support itself to driving funding for others. Most PID systems are fully committed to managing the assignment, data curation, systems support, resolution, and other public services. To the extent that there are “extra” resources, those are directed toward education, outreach, and systems development/retooling/expansion. General advocacy, education, and infrastructure coordination are important, but instead of generating the interests of potential funding bodies, it could just as likely inhibit the engagement of those organizations, particularly in the USA but potentially worldwide. These organizations provide either direct — or more likely indirect — support for PID infrastructure in specific cases, but not for infrastructure generally.

In 2018 at the 11th RDA Conference in Berlin, Patricia Flatley Brennan, Director of the US National Library of Medicine, presented on a keynote panel during which she outlined her desire to see greater funding for infrastructure support. She described how critical the infrastructure is for the dissemination and preservation of scholarly communications. Yet, she went on to describe how poorly that infrastructure fits into the funding structures for science. Infrastructure isn’t normally novel, except in the case of entirely new systems. The launch of a new identifier system might receive three to five years of initial support, perhaps a few years beyond that, but with the expectation that the system will become “self-sustaining” after the period of initial support. As with the development of ORCID, it is clear that this process takes far longer than that to gain wide support and the resources that might flow from that. While these projects grow and become ever more embedded in the community, the more critical the information exchange, the less likely these efforts are to get sustained funding from those same funding bodies. Organizations like NIH/NLM and DOE often support their own infrastructure tools, but ongoing support and maintenance is something that is generally not funded on an ongoing basis, such as an endowed chair or institute might be.

Generally, the expectation is that scholarly supporting organizations (i.e., publishers, libraries, or library consortia) will continue to support these scholarly communication infrastructure tools. Setting aside the existing challenges of supporting this infrastructure, there are additional problems moving forward for two reasons. The first is that the drive toward open access publication is driven by a desire of many in the open access advocacy community to drive down the resources (i.e., income) for the publishing community. One recipient of “publisher surpluses” are the infrastructure support organizations, such as CrossRef, ORCID, NISO, and the new RoR registry, all of which draw resources from publishers, among other sources. Another potential unintended consequence of the move toward open access could be the continued deterioration of library budgets, as funding moves from traditional libraries toward research offices at many institutions. This is not to argue against the move toward greater open access, but rather to put forward the precarious situation of many of these infrastructure systems should there be a rapid shift in resources as open access gains in popularity and adoption. The community needs to foster the growth and continued success of these infrastructure systems and plans to support such systems should not be made only as resources from traditional sources begin to dry up.

Before I draw this article to a close, I would be remiss if I didn’t touch on the role that NISO plays in this story, since the many things that we do to support this infrastructure might not be obvious. NISO serves in a variety of important roles in this persistent identifier ecosystem. First, as a consensus standards body, NISO has initiated and advanced several identifier initiatives in our community, including what became ISSN, DOI, and the numerous schemas and vocabularies that integrate these identifiers into the ecosystem. NISO continues to be involved in the maintenance, regular review, and revision of many identifier standards, both nationally and internationally. NISO serves as the ISO Secretariat for the committee that standardized and maintains more than a dozen international identification standards. While NISO represents US interests on these groups and establishes US positions on these initiatives through membership ballots, we also serve as neutral managers of the committee and its overall work. As Secretariat (operating officially on behalf of ANSI), NISO supports—directly through staff time and other administrative funding — the working group management, the editorial, the balloting, and project support for all of ISO TC 46/SC 9. Right now, there are eight active working groups that NISO is coordinating in that sub-committee, working on revisions, systematic reviews of standards, on new project development, and other administrative issues.

None of this work is accomplished without resources. The first step in advocacy is raising awareness. Perhaps, this initial introduction to PIDapalooza, the world of persistent identifiers, and the role NISO plays in this world is our first step in advocating for resources. If you are a NISO member, or if your organization is not, and are in a position to help fund some of this critical infrastructure work, we should talk. Perhaps one place to do that is the upcoming NISO Plus conference. But unfortunately, by the time you read this, it likely will have sold out! I hope to see many of you at that exciting event.

Sincerely,


Todd Carpenter
Executive Director, NISO