ResearchGate is academia’s most popular social network, with good reason. While some decry the platform for questionable user recruitment tactics, others love to use it to freely share their articles, write post-publication peer reviews, and pose questions to other researchers in their area.
ResearchGate quietly launched a feature recently, one that we think could be a big deal. It may have huge upsides for research–especially for tracking altmetrics for work–but it also highlights how some of the problems of scholarly communication aren’t easily solved, especially when digital persistence is involved.
The feature in question? ResearchGate is now generating DOIs for content. And that’s started to generate interesting conversations among those in the know.
Here’s why: DOIs are unique, persistent identifiers that publishers and repositories issue for their content, with the understanding that URLs break all the time. A preservation strategy is expected when one starts issuing DOIs, and yet ResearchGate hasn’t announced one, nor has DataCite (which issues ResearchGate’s DOIs).
Some other interesting questions: what happens when users decide to delete content, or leave the site altogether? Will ResearchGate force content to remain online, or allow DOIs to redirect to broken URLs?
And what if a publication already has a DOI? ResearchGate does prompt users to provide a DOI if one is available, but there are no automated checks (as far as we can tell). That may leave room for omission or error. And a DOI that potentially can resolve to more than one place will introduce confusion for those searching for an article.
As a librarian, I’m also curious about the implications for repositories. IRs’ main selling point is digital persistence and preservation. So, if ResearchGate does indeed have a preservation policy in place, repositories may have lost their edge.
We’ll be watching future developments with interest. There’s great potential here, and how ResearchGate grows and matures this feature in the future will likely have an influence on how researchers share their work and, quite possibly, what it means to be a “publisher.”
ResearchGate also does not check whether DOI’s listed by users are real. This “publication record” for example seems careless at best and fraudulent at worst; the claimed paper has not been published by the cited journal (or any journal as far as I can tell): http://www.researchgate.net/publication/258325997_Out_of_Body_Experience_induced_by_hypnotic_suggestion._Part_1_phenomenology_and_perceptual_characteristics
I think (and hope) that the disregard for truth and ethics displayed by ResearchGate will serve to emphasize the value of properly curated web resources.
Great point, Janne! Automated checks would be a great value add (making it much easier for users to catch typos) and would help RG keep their data clean. You point to a much larger issue of how RG will uphold the “social contract” (to borrow from Carl Boettiger’s great post on DOIs [1]) that comes with minting DOIs. How will they ensure data quality if not checking inputted DOIs for correctness? How will they ensure persistence if no preservation plan exists (or if they do have a preservation plan, how can users trust them with their data if they don’t know about it?).
I’m curious to see where RG takes their DOI functionality–and all the related features they could offer (automated checks, preservation, etc)–in the future.
[1] http://www.carlboettiger.info/2013/06/03/DOI-citable.html
I agree with most of this, but …
“As a librarian, I’m also curious about the implications for repositories. IRs’ main selling point is digital persistence and preservation. So, if ResearchGate does indeed have a preservation policy in place, repositories may have lost their edge.”
That has the flavour of the Shirky Principle: “Institutions will try to preserve the problem to which they are the solution”. What we want is surely for the preservation problem to be solved; not, particularly, for library to solve it.
Agreed! Having worked with IRs for most of my career, I’ve always been confounded by the insistence that the ‘grand principle’ of long-term preservation should win over researchers more than usability and features of our systems. Why would people want to deposit in IRs when they’ve got other places to do it that are easier to use, have a larger degree of uptake than our IR, can more easily talk to other systems via (gasp!) open APIs, and so on?
If ResearchGate can guarantee preservation, I’d personally welcome it. (Though I doubt I’d use it any more than I do now, since it’s an information silo.) Better systems are better systems, no matter who’s created them. But it will give libraries a run for their money, just as Figshare has done.
I can’t see how to get email alerts when a new comment is left. Am I missing something?
Actually, no! Our comment plugin doesn’t offer that. I’ll do some digging to get that turned on in our system.
Stacy – without a preservation plan and with no duplicate checking, isn’t this just going to pollute your index and result in lots of crappy data? It would be good to know a little more about what RG’s preservation plans are here.
Yup! Exactly. A proliferation of DOIs will make it even harder to accurately track and associate metrics for the same scholarly work that has multiple versions IDs. Fingers crossed, RG will be more forthcoming about preservation in the near future.