Martin Paul Eve bio photo

Martin Paul Eve

Professor of Literature, Technology and Publishing at Birkbeck, University of London

Email Books Twitter Github Stackoverflow MLA CORE Institutional Repo ORCID ID  ORCID iD Wikipedia Pictures for Re-Use

One of the best things about Mendeley is that, the second you mention their name on Twitter, a horde of helpful and informative community liaison team members descend upon you. This means that, if you have a query, you can be sure that someone knowledgeable is never far away.

That said, it seems that my current issue -- being able to convert either an OpenOffice or Word Zotero document to Mendeley -- is not possible. Having communicated with @mrgunn he suggested two approaches:

  1. Save the document without Zotero markup and re-insert the citations
  2. Run the document through a LaTeX converter (untested)

As the document in question has over 500 citations, I didn't find this to be of much use (which is not a criticism of @mrgunn, but rather the sad state of interoperability). However, I decided to do a little digging myself to work out what the actual issues are...

Now, in OpenOffice (at least), a Zotero field code looks like this: Reference: ZOTERO_ITEM {"citationItems":[{"locator":"17","uri":[""]}]} RND5qVC9BQOk1. Upon visiting the URL in question, it is quite clearly a Zotero catalogue entry. A Mendeley citation, on the other hand, looks like: Reference: Mendeley Citation{f19dafe4-bd54-4b6c-9585-7acfd39428c9} RNDSDLGneoN3n. The task of a converter would be to match these two styles up.

However, it doesn't seem to be that simple. Zotero reference fields can also contain a plethora of other information that Mendeley seemingly doesn't support, the foremost (and to my mind a deal-breaker) being page numbers (!) This means that any conversion to Mendeley's format would be lossy; it would be impossible to preserve all the data from Zotero citations because Mendeley simply doesn't support including all this information.

While Mendeley seems, from my Twitter conversations, to be winning the hearts and minds of the research community, time and time again I query whether this is because they have a superior product to, say, Zotero or because they are able to market it better owing to the monetarization of their service. Things they win on: pretty interface; social aspect of sharing; online profiles generated automatically are great. Sadly, however, some basic stuff is missing. If they could implement this and then work with Zotero to allow document cross-compatibility (alongside not asking me to enter my email password to garner contacts -- this is incredibly bad practice -- OAuth?), they'd be on to a winner with a superior product.