Continuing our series of Kitchen Essential interviews with leaders of research infrastructure, today we hear from Chris Shillum, Executive Director of ORCID (Open Researcher and Contributor Identifier). ORCID provides a persistent identifier (PID) — the ORCID ID — that researchers can connect with their professional information, so that it can be shared with other systems.

Please tell us a bit about yourself: your role at ORCID, how you got there, and why you embarked on a career in research infrastructure?

I’m the Executive Director of ORCID, which is US non-profit jargon for CEO. I joined the team full time in this role in 2020, but I have a long history of involvement with ORCID before that. For most of my career, I worked in scholarly publishing at one of the largest commercial publishers. During that time I was involved in several collaborative scholarly infrastructure efforts, including being part of the group that founded ORCID. Once it became an official organization, I joined the inaugural board and remained on it for about nine years. So I’m very fond of ORCID — over the past 12 years I’ve seen the organization establish itself and become one of the core pillars of the scholarly infrastructure; it feels like coming full circle to now be part of the team.

Early in my career, when working on the first wave of the transition to online scholarly publishing in the late 90s, it became clear that to offer the best services to researchers, publishers needed to figure out how to collaborate. That led to the formation of what became Crossref, which set the precedent for publishers, who were otherwise fiercely competitive, to collaborate on shared infrastructure projects. This then helped establish trust during early discussions about an open non-proprietary researcher ID service. When it became clear that this would need support from a broader constituency to succeed, experts from research institutions, libraries and funders also joined those discussions, and ORCID was born!

Today, the largest group of stakeholders in the ORCID community are research organizations themselves. But collaboration is still critical. In order to improve the experience of researchers, and particularly in the case of ORCID, to reduce the time and energy they spend on admin tasks, everyone involved in scholarly communication needs to work together towards the same set of goals. Like many of my colleagues, I ended up working in scholarly infrastructure somewhat by accident, but I love the opportunity it provides to combine technical knowledge with community building and the ability to collaborate with talented experts around the world.

What do you like most and least about working in research infrastructure?

One of the threads that’s carried through my career is getting stakeholders from different sectors and communities working together on shared problems. I am at my core an engineer, so I like to solve problems and build things. The opportunity to do that collaboratively and get all our stakeholders — universities and research institutions, funders, policymakers, publishers, and vendors — working together to make life easier for researchers, is what I like best.

I also love working with a dedicated and motivated team spread around the world. ORCID has close to 40 people working across 12 different countries. One of the things that ties us together is the belief in our mission and the orientation to work hard, work smart, and do a better job at serving our stakeholders every day.

What do I not like? In some ways ORCID is like a 40-person multinational spread across those 12 countries, and that brings with it the complexity that faces larger organizations. We deal with a lot of bureaucracy, from employing people in multiple jurisdictions to the huge variation in procurement requirements of our members in 57 countries. It’s something we have to do, but I rather wish we could do a bit less of it!

Based on your own experiences, what advice would you give someone starting, or thinking of starting, a career in research infrastructure?

Learn how to collaborate. None of the infrastructure efforts that I’ve been involved in could have been delivered by any one organization alone. So understanding who is who in scholarly communications and the different roles that they play is a first step. But then understanding how to build bridges to bring everyone together to solve common problems is key. People at different organizations have different perspectives; sometimes they also have different priorities and different values. Working with the diverse group of folks who helped found ORCID, I learned a lot about the importance of shared goals.

We’ve covered some of this history in a document published last year to celebrate our 10th anniversary. By sharing some of what we learned along the way, we hope others can benefit from our experience. Infrastructure is about different groups collaborating to make something possible that wasn’t possible before, and that turns out to be quite a complex thing to achieve.

I’d add that, when building infrastructure, sustainability and trust are key. So understanding not just how to solve the problem technically, but also how to communicate effectively to and engage with all the different stakeholders and spread the message is really important. Not that I do that just by myself at ORCID — I have a very good team who are experts!

Also, develop financial and business skills. If you want people to trust your infrastructure, then you have to make a convincing case it’s going to be around for the long term. Understanding how to build a business model that is sustainable, and how to run an organization efficiently so that you can make progress with the available resources, are very important skills as well.

What sort of infrastructure does ORCID provide, and who are your users?

We provide a unique identifier for researchers. Anybody in the world can come to orcid.org and sign up for an ID if they find it useful. The initial use case was to solve the ambiguity around authors’ names: many researchers have the same name, and individual researchers may publish under different names over time, either because their name changes or because of different naming conventions for different research outputs.

But over the 10+ years since ORCID launched, many other use cases have emerged. Being able to uniquely identify contributors helps with generating insights about research, particularly outcomes and impacts, which are increasingly of interest to funders, governments, and policymakers. This is only possible if you can connect the dots between researchers, the work that they do, the organizations they belong to, and the funding that they receive. We’re a key pillar of the so-called “PID (persistent identifier) infrastructure,” along with fellow organizations such as Datacite, ROR and Crossref. Although our users are researchers, we also serve a wide range of organizational stakeholders; we’re helping them better understand the world of research and also enabling them to reduce the administrative burden that they levy on researchers.

How is ORCID sustained financially?

We are now sustained 100% by our members. Researchers can, and will always be able to, use ORCID free of charge. So we’re supported through the membership fees paid by around 1,350 organizations globally, for which they get benefits including better technical support, and access to our member API, which allows them to contribute data on behalf of their researchers to ORCID. Our biggest group of organizational members are universities and research institutions — over 1,000 of them. But our members also include funders, government policymakers, vendors who provide services to our stakeholder groups, and publishers. It is only through their combined support that we’re able to continue our mission.

ORCID wasn’t always able to sustain itself from these membership fees. In the early years, a number of the publishers who came together to form ORCID provided startup loans. A few years later, we were fortunate enough to receive a number of very significant grants that helped us expand globally and bridged the gap until we reached sustainability.

Since breaking even in 2019, we’ve generated a reasonable surplus each year, which is really important. We don’t want to be running the organization on an empty tank; we want to be able to invest in refreshing our technology from time to time, and to see ourselves through any financially hard times.

As the leader of a research infrastructure organization, what do you think are the biggest opportunities we’ve not yet realized as a community — and what’s stopping us?

I think we’ve made enormous progress, in particular, as one of a group of PID infrastructure providers working towards the goal of automating the integration of information about the world of research — saving researchers time and effort and allowing the generation of new insights. Today, around 8 million researchers are actively using ORCID every year and over 5,000 systems are integrated with our APIs and using our dataset.

But there’s more to be done. The thing we’re focused on at the moment, and the core of our strategy, is that more, more fully populated ORCID records will help us better fulfill our mission. So we’re working on making it easier for researchers to ensure that their ORCID records are complete and up to date — while doing less manual work. The key here is making it easier for organizations that hold trusted information about researchers and research to contribute it to ORCID so that it’s available to everybody else to reuse — with the researchers’ permission, of course; one of our founding principles is that researchers are always in control.

Communicating why this is important to our diverse stakeholders is a challenge. The message is one of altruism — contributing data that you hold may not be useful to you, because you already know it, but it is useful to everybody else. If we can all work together to create an openly accessible interlinked web of data about the research ecosystem, then we unlock even more potential to both generate better insights and further reduce admin burden for researchers.

What holds us back is the enormous complexity of the scholarly communications ecosystem: the number of different stakeholders involved, each with different priorities and perspectives, and all with constrained resources when it comes to building and integrating systems. So it just takes time. One of the things we’ve spent a lot of energy on in recent years is tailoring our messages to our individual stakeholder groups, because everybody has something slightly different that they can contribute, and everybody has something slightly different that they can get out of participating. It’s important that we talk to our stakeholders in those terms — terms that make sense to them and resonate with them. Fortunately, we have a great community that supports us very well. And they do a great job of getting the message out as well.

It’s also worth saying that scholarly infrastructure is not just, or even mostly, a technology challenge. I would never dismiss the huge amount of work that our tech team has done to build and maintain a great system, but that’s not enough. People are often surprised to hear how much of our effort goes into non-tech activities such as outreach and communication.

Looking at your own organization, what are you most proud of — and what keeps you awake at night?

I’m very proud of everything that everyone involved in ORCID has achieved together over the last 12 years. It’s been great to see that something that started as an idea, a discussion among a relatively small group of people, is now known by millions of people in research and is part of their day-to-day lives. I like to see things grow, and I think we’ve successfully done that with ORCID.

Looking more recently, in the past three years since I joined ORCID, the thing that I am most proud of personally is our work to increase global participation in ORCID. That wasn’t a focus in the early years, when the organization was fighting to get itself off of the ground and become sustainable, so it was apparent when I joined that there was uneven participation around the world; countries with lower-income economies couldn’t participate as fully as the richer nations, and we needed to fix that.

We’re working to improve this situation in a number of ways. Last year, for the first time, we introduced fees for our consortia based on affordability — previously everybody paid the same regardless of geography. And with the help of some of the earliest supporters — our founding lenders — we also got a grant fund off the ground. We’ve now made two rounds of awards to help with both community outreach and technical integration in some of those under-represented communities, and are just about to finalize the third.

As for what keeps me awake at night? You know, not very much. One thing I do worry about is somewhat beyond our ability to influence or control: the potential fracturing of international collaboration in the world of research. We set out to be a global organization that is open to participation by all, and we hope to be able to continue that. But we are also aware that, in the current geopolitical situation, there may come a point where that is increasingly difficult to achieve.

What impact has/does/will AI have on ORCID’s work?

The extreme view would be that AI takes away the need for the kind of structured metadata and identifiers that we provide. I’m a little bit skeptical about that, and in fact we’re pretty sure that the big AI players are ingesting our data already! We can see their crawlers hitting our website, which is fine. Our public data is openly available to anybody who wants to use it. In fact, I think that structured, curated data is going to be quite important as the application of AI to research grows.

Obviously AI has huge implications for the world of research more broadly, but we try to keep our focus on a core foundational role — assigning identifiers and gathering and distributing metadata. I don’t think AI is going to remove the need for that any time soon.

The other big tech innovation of recent years is distributed computing, which may have more of an impact on us because ORCID is today very much a centralized system. That’s something that we are keeping an eye on and thinking about what we might need to change in order to integrate with that world.

What changes do you think we’ll see in terms of the overall research infrastructure over the next five to ten years, and how will they impact the kinds of roles you’ll be hiring for at ORCID?

Increasing transparency is critically important. We are very careful to record and disclose provenance information about all the data in ORCID. We know who made each assertion about each piece of data. And we make these “trust markers” openly available. With the risks from AI, and the incentives that drive some researchers toward bad practices, openness and transparency about who is saying what is key to rebuilding and maintaining trust in research and scholarship.

I also hope we see a lot more efficiency in the administration of research. Researchers want to do research; they don’t want to have to deal with the systems of dissemination and assessment. While I think those systems are necessary, they shouldn’t require as much effort from researchers as they do today. So the contribution we make at ORCID, along with our fellow scholarly infrastructure organizations, is to enable those efficiencies to be realized.

But I’m not sure that these changes will impact our hiring decisions. We already have a diverse team, including highly skilled engineers, product managers, communications specialists, operations experts, and our engagement team who work very closely with our community. Obviously the technology will change, and we continually help our tech team to keep up-to-date with those trends by providing training and development opportunities. We have just hired our first dedicated Data Protection Officer, bringing additional skills in the area of privacy into the team. And I’m sure we’ll need to add more cybersecurity expertise because it is an ever more complex challenge to protect ourselves from those risks. But overall, I don’t see a huge change in the range of skills that we need to successfully operate a significant piece of the scholarly infrastructure.