Allow me to introduce myself

My university, like many others, is racing to embrace an open future. We are putting stuff into our repository as fast as we can. Each item has a unique identifier, like an International Standard Book Number (ISBN) or a Digital Object Identifier (DOI), so that we know exactly which book or paper we are talking about.

We are also encouraging staff to share their research data, where they can. We are working with the Australian National Data Service (ANDS), through their Cite My Data service, to make sure that these data sets also have Digital Object Identifiers.

Excitingly, these identifiers will link the papers, chapters, artworks, and (insert your favourite research output here) with the data sets. How cool is that? When I write my groundbreaking libretto, drawing on my amazing new data set, everybody will know exactly which dataset was used in exactly which libretto.

And everybody will know exactly which ‘me’ did it, because I’ll have included my ORCID ID, Scopus Author ID, Google Scholar ID, or my (insert your favourite researcher ID scheme here).

Everyone will know, that is, except for my university. My university will just have to guess.

Please allow me to introduce myself. I’m Jonathan O’Donnell. I’m not this Jonathan O’Donnell (although it would be really cool to work on the Arctic for the US National Parks Service). I’m certainly not this J. O’Donnell (I wish! He writes beautifully about digital humanities).

You might know me by my ORCID ID (0000-0001-5435-235X), or by my Scopus Author ID (23005925700), or even my Google scholar ID (3pvY_LgAAAAJ). If you know who that is, then you know who I am. Categorically. Unambiguously. Forever.

These three identifiers are examples of unique identifiers provided for free to academics. Admittedly, it is probably unlikely that you use identifiers like these day to day:

Hi, 3pvY_LgAAAAJ. How are you?

Not bad, thanks, nla.party-626227. Have you seen 0000-0001-5875-8744 around?

We don’t talk like that. Computers do. They do it so that we can disambiguate scholars of the same name. These sorts of identifiers are vital if you have variations to your name or change your name, lose your job, or move to a different institution (or country) or move between academic and #altac careers. I’m only a tiny researcher, so they are really important to me.

They are so important that I’m going to wait right here while you go and sign up for one right now. Go on – I’ll wait.

• ORCID and ORCID identifiers.
• Scopus Author ID – I think that Scopus have folded their author IDs into ORCID.
• Google Scholar Citations.

I don’t know what it is like at your university, but where I work, we don’t actually know who we are. We know what we publish, and we proudly tell the world about it. We know what data we collect, and are increasingly keen to share it with the world. But we don’t have a clue who we are. Or, to be more exact, my university doesn’t know who I am.

Unless you work at my university, you probably don’t know me as RMIT employee number 24323. That’s what my university knows me as. That’s all they know me as. They don’t know me as any of those other identifiers. At the moment, there is no easy way to link my external identifier (ORCID, Scopus, or Google Scholar) to my internal identifier, my employee number (e-number).

So, I’m having an identity crisis. My external identity is blossoming. It is becoming more and more intertwined as computers pick up these identifiers and I build cross-links between them. Meanwhile, my RMIT identity, the identity that pays my wage, is stagnant. External me is reaching out while internal me is stuck forever in its feeble e-number – limited, lost, dead. Go towards the light, e-number! Go towards the light.

It will take considerable work for my university to see the light. They will need to:

• Decide that they should adopt an external identifier for all research-active staff.
• Decide what identifier they should adopt.
• Explicitly link that identifier to the internal identifier, preferably through our Lightweight Directory Access Protocol (LDAP) server or similar technology.

Making policy is hard. On the face of it, this one should be a no-brainer. By linking internal employee numbers to an external identifier, my university would gain significant advantages:

• We would encourage all our researchers to adopt an external identifier, which would be a good thing.
• This would improve the profile of our researchers, in the same way that open repositories improve the visibility of papers and other outputs.
• It would make it easier for our researchers to measure their performance using alt-metrics.
• Most importantly for the organisation: it should make the collection of research statistics much easier. Given that we spend an enormous amount of staff time doing this now, that is a clear cost saving for the university.

If it is so smart to do this, why haven’t we done it already? Perhaps we are shy. I don’t think so.

Is it because we are allergic to things that we don’t control? It can’t be that either because we have championed external identifiers for a long time. I remember contacting my university library (probably 20 years ago) to ask for my first International Standard Book Number. I was so excited! In those days, the university library used to be the custodian of blocks of ISBNs and distribute them to staff upon request.

This is what I think it is: we’re allergic to these new technologies that we don’t control, blind to services outside the walls. Also, it is a bit hard to link to different external services, and to keep those links working over time. And it should be noted that identifiers like this are only relevant for staff who may be contributors to research, so they are not a universal solution. They won’t cover all staff. However, they will cover all staff with an academic output, which would be a lot better than the current situation.

Besides that, there needs to be a fight an evaluation of corporate solutions (a la Elsevier and Google) versus open solutions (a la ORCID), and whether the business case is worth the effort. For the record, I think that it is absolutely worth the effort, and that open beats corporate every time.

However it happens, I think linking to an external identifier is inevitable. When it happens, the triangle will be complete. When I write my groundbreaking libretto, which is built upon my wonderful data set, everybody will be happy.

• People will know exactly what data I have drawn upon.
• They will know exactly which research output I have created.
• And they will know exactly who I am.

Everyone will know, including my employer. I will be able to stand up and be counted.

---

Addendum – 9 May 2014

On Twitter, the wonderful @Wragge alerted me to a list of UK projects that are working with ORCID. This is really exciting to see.

Addendum – 17 May 2018

Exactly four years later, I’ve just linked my RMIT employee identifier (e24323) and my RMIT student number (s9501990) with my ORCID identifier (0000-0001-5435-235X). Now, in theory, RMIT can categorically state that the student that is me and the staff member that is me are one and the same person. That’s nice.