Is OMICS Using Knock-off DOIs?
Cheaters be cheating, and the DOI system appears to be vulnerable to fakery
Note: I’ve corrected a mistake in the first release of this post — that is, DOIs can have 5-6 digits after the initial 10. at the beginning. Sorry for that error, and thanks to my side-corrector. I’ve also expanded on why this problem isn’t going away, and may lead to OMICS becoming a predatory DOI handler.
At this week’s US STM Annual Meeting, I repeated an assertion I’ve made before — that DOIs (and the CrossRef logo) are used by predatory publishers to connote legitimacy and acceptance to fool potential authors.
There are numerous examples of DOIs (and ISSNs) being promoted in this manner, so when CrossRef revoked OMICS’ membership and ceased issuing them DOIs, this seemed to, a) confirm that the DOI does connote some level of legitimization, and b) seemed to have the potential to stop OMICS from making further headway as a predatory publisher foisting DOIs on users as a marker of legitimacy.
Well, never underestimate the cleverness of a cheat. As Lisa Hinchliffe discovered after watching a Twitter thread on the topic unfurl after my session, OMICS is using DOI-like identifiers to mark articles now. Generating strings like this — DOI: 10.37421/iem.2020.9.284 or DOI: 10.36648/0976-8505.11.1.1— seems to not only be technologically trivial, but also leads to downstream acceptance of the DOI, as I was able to find these in ResearchGate and other aggregation systems of dubious quality.
Think of it as comparable to a VIN — vehicle identification number — that’s not registered with the state and differs in format just enough to make it hard to distinguish. A knock-off. If a car manufacturer were to fake a VIN to sell a lemon, they’d be cheating in a big way.
Because we have no central source of DOI authentication — and because CrossRef is just one of many DOI handlers — these knock-off DOIs may be viewed as legitimate through the lens of pure technology. More interestingly, as one person noted to me via email, OMICS could itself become a DOI handler, distributing DOIs to itself and other predatory publishers (or legitimate publishers) for a fee.
I just wanted to share this quick note before the weekend, to flag this issue as new, to thank Lisa for catching it, and to see what others think. Comments are open to subscribers below.
Thanks.