Building an academic information highway block by block

Five years ago we had the idea of interconnecting all Canadian universities, research centres, associations and funding agencies using software, which led us to build UNIWeb. From the beginning, we understood that each connection had to advance the goals of both the individual institutions and their members. Here we reflect on what we’ve learned along the way, and on what we are currently doing to build a novel kind of academic information highway.

Lesson 1: the network already exists

While it may be obvious to many, it took us a while to realize that a network of Canadian researchers already existed. Researchers belong to multiple research associations, centres and other types of networking hubs. They also network with one another at conferences and other events. So, clearly, we didn’t need to “build” a Canadian research network because it had already been built organically over many, many decades, and it works rather well.

Lesson 2: global networks are better… or not

The most common feedback we got in the early days was that global networks such as ResearchGate and Academia.edu were the next big thing. This fit into the popular business concept of going global from day one. But this view contradicted Lesson 1 because the network already exists, which means that there is no dichotomy between a national network and a global network. All of these different scales are important: from the local campus and city, to provincial, national, and global levels. The modern “software” flavours of global research networks are just that: flavours of a tool that you might find useful, or not. In particular, the aforementioned examples are centralized websites, each controlled by a single commercial owner.

Lesson 3: the network needs time-saving tools to thrive even more

The real, living network of researchers is under significant stress levels due to scarcity of time. So many things demand so much time that many good initiatives and projects are either dropped or never finished. In 2012 we conducted a focus group with researchers to discuss an earlier version of our software. The message we received was loud and clear: their main stress point is the preparation of funding applications that takes a lot of their valuable time. Whatever time is left, is spent on research, teaching and administration.

Lesson 4: there is a need for an academic information highway

For a long time we thought that a “social network” for researchers was about building web profiles, showing connections, publications, and so on. And of course, these are useful, but we were missing the point that people need fast and effective ways to move information across organizations. For a network formed by institutions, funding agencies, associations, centres, and conferences, this means that information must flow across different kinds of organizations with different data needs. Not a trivial task at all!

Lesson 5: ORCID is a step in the right direction, but…

Several years ago, ORCID introduced the idea of a unique identifier for every scientist with the objective of properly linking publications to their authors. The benefits of reliably linking authors to their work by a number instead of names or email addresses were clear to most people. They also introduced the concept of a reusable profile for each scientist, whose benefits were not so clear. The ability to reuse academic data seemed a step in the right direction, but the details of how this is done raises many questions. ORCID has sole control of what data is captured by their profile and who can use it (with the owner’s consent), and how much it costs to gain access to it. This approach to reusing academic information makes the exchange of data vulnerable to changes in the direction and agenda of ORCID. And the fact that there are various paid access levels to the personal information of scientists makes the proposal a bit murky. So, the ID sounds good, but we are skeptical about the rest.

Lesson 6: software makers have no business owning other people’s information

Well, actually, they have a lucrative business doing just that. But we don’t like it. We are the type of software maker that makes and sells the software. We don’t like owning the information of other people, and we don’t like others owning our information. So, what’s the lesson here? I don’t know, I just wanted to clarify that we don’t follow that kind of business model.

Putting all the lessons into practice

To sum up, this is what we know: the real, physical research network is a beautiful system that exists at multiple scales and spans several types of research organizations. Part of its beauty comes from the independence and autonomy that each participant has in it. Today, the primary challenge is to optimize the time management within the network; reducing unnecessary frictions and making everything speedier. And a big part of the challenge is to achieve these goals without taking ownership of any personal data or commercializing access to it.

A lego-like approach to building an information highway

One reason LEGOs are so empowering is because they let individuals combine building blocks in any way they like, and still get a solid and dependable working structure at the end. That principle guided us during the design phase of our software solution to simplify the flow of information across academic organizations. But before I tell you how we are building that information highway, let me provide some context about what our software already does.

UNIWeb is a web-based research network that functions at the scale of a single organization, such as a university, academic association or research centre. The software serves to manage academic CVs, export CV information to funding agencies, compute academic metrics, etc. It also features web profiles for professors and graduate students, and the ability to find connections based on common research interests, among other things. Currently, 10 independent Canadian institutions are using the software, and they each have full ownership of their data. This approach has simplified the process of exchanging information between academic institutions and funding agencies since 2012.

Our current work is focused on building an academic information highway in Canada that will let researchers in any institution (with the appropriate software) register to other networks, and share their information across them. For example, as a member of an institution, you will be shown a list of other organizations with compatible software, and a single click will let you create a new account on an independent system, and link the new account to your existing account. Right after, you are offered the option to select what information to share from your web profile and/or academic CV (e.g., list of publications, projects, profile picture). What’s really amazing about the technology is that the systems can synchronize the information for you as it changes over time. This ability spans all the information captured in a web profile, or private academic CV. You make changes in one place, and those changes get propagated across the organizations that you want, instantly. Like Dropbox, but for academic data.

Why is this good? Imagine…

Here are some scenarios that will ignite your imagination. Imagine never having to submit your CV to a funding agency because they already have your latest CV information. Imagine registering to a research conference with a single click, and your profile information becoming immediately available and linked to those of other prospective attendees. And imagine that any academic association or research centre to which you belong will be able to have your up-to-date information without you having to send it to them each year.

This solution is all about minimizing the time consumed by the exchange of academic information. What’s more, it’s faithful to the principle that each individual organization must have full control of what information is captured in their web profiles, academic CVs, and other types of structured documents. And, as is already the case with our software, the data captured by each individual network is owned by the host organization. We don’t control what information is captured or shared, we don’t own the information, and we don’t make money selling access to it.

Currently, the only software that is compatible with this “highway” architecture is our own, but we are working to convince others to agree on an open standard for academic data syncing, so that institutions can choose which software is best for them. We will be publishing new articles soon with the details of the initiative and how we are dealing with some important technical challenges.

Get in touch with us

Drop us a line if you think that we are on the right track, or if you think that we are missing something. We are planning to release this new functionality in beta by mid year (July 2017), so your comments can have a big impact on how this turns out.