Disambiguation/provenance of claimed scientific opinion and research.

No Gravatar

My name is displayed pretty prominently on this blog, but it is not always easy to find out who the real person is behind many a blog. In science, I am troubled by such anonymity. Well, a new era is about to hit us. When you come across an Internet resource, or an opinion/review of some scientific topic, I argue here that you should immediately ask: “what is its provenance?”

In the 350 year history of scientific dissemination[1], provenance has almost always been provided by publishers. Arguably, that was their most important role (and arranging anonymous peer review). Not that they ever met with their authors or always established that a real person or a real group actually existed! But with the explosion of vanity publication and a host of horror stories about articles for sale to authors keen to have a publication to their name, perhaps the role of provenance needs rethinking.

ORCiD is a project that seems to be gaining serious momentum in achieving a mechanism for disambiguation and provenance of researchers. Thus Brian Kelly (who has played an important role in the modern internet in the UK since 1993 or earlier) encourages all researchers to sign up (although I cannot help noting, rather cheekily, that he does not add his own ORCiD as provenance for his blog). ResearcherID was in fact an earlier organisation to offer such a service, but it is run by a commercial publisher and it is hosted at a “.com“. ORCiD at least claims to be an open (.org)anisation, and carries an open source license. It seems that some UK Universities (home to some researchers) have decided to sign up to ORCiD and most I suspect are planning to deploy these resources amongst their researchers, and quite possibly their students as well (postgraduate initially, maybe even undergraduate eventually).

I jumped the gun somewhat, getting mine more than a year ago. Better the devil you know, etc etc! It is orcid.org/0000-0002-8635-8390. What happened next? Well, I publish data@Figshare, who themselves signed up to be an early member of ORCiD. This gives them access to the API (application programmer interface), and so by supplying my ORCiD to Figshare, I can gain access by proxy to the ORCiD features on offer. The most immediate impact is that ORCiD lists all the data-objects I have published at Figshare, thus establishing a trust between them and my ORCiD identity. Mind you, no-one at ORCiD has ever met me, or checked on who I am. I think that task is going to be delegated eventually to e.g. my university (I am not absolutely certain how the linkage between my ORCiD and my employer, who clearly know me since they pay my salary, will be formalised). Because my employer has also now become an ORCiD member, we will be adding ORCiD API access to our own SPECTRa-DSpace data repository shortly, so that the data held there will also be added to my ORCiD lists.

And as the major journal publishers start to do the same, a formal linkage between my identity (perhaps as verified by my employer), journal-published articles (narratives) and my data publications (via the identifiers known as DOIs) will come into being.

How, you might reasonably ask, is this in the least useful? In truth, I am not sure anyone really knows exactly where this is heading. For example, impactstory.org/about is one added value site which attempts to gather altmetrics about the impact your research is having. But hey, the although the preceding link tells you who founded this organisation, you do not get the kind of provenance I am describing above; none of the founders cite their ORCiDs! You do get their @Twitter accounts though; I wonder what that tells us about the modern interpretation of provenance? Well, my impact can be seen here; in truth it’s not quite the impact I imagined my scientific career was having, but I suppose this is early days. What I am pleased to tell you is that ImpactStory does tell you not only about the impact articles I have published has had, but also the data. Two data sets are described as both discussed and highly viewed. Although as usual, you do not get to learn why the data is being discussed!

Where next? Well, to go back to the start of this post; blogs. It would be nice to formally link this blog to my ORCiD ID (this is not done simply by quoting it here, but via the ORCiD API). If/when I work out how to do this, I will no doubt post the event!

References

  1. H. Oldenburg, "Epistle Dedicatory", Philosophical Transactions of the Royal Society of London, vol. 1, pp. 0-0, 1665. http://dx.doi.org/10.1098/rstl.1665.0001

Tags: , , , , , ,

Leave a Reply