From Sci-Mate to Mendeley – a brief history of reference managers

This is an edited version of an article I wrote for the April issue of eLucidate, the UKEIG Newsletter.
The bibliographic reference is the foundation of scholarship. A reference is a surrogate for knowledge, a surrogate for research results that have been condensed into a journal article and then stored in the minds of scholars. Reference management therefore is all about knowledge, though we librarians sometimes forget that.
In the early days the main functions of personal bibliographic database programs were to provide a readily-accessible and searchable store of knowledge, and to act as an index to a reprint collection. Until the early 1990s [1, 2] these remained the main reasons for wanting to use such programs. Annotating and subject tagging references were important functions, but bibliography production was relatively unsophisticated. This was the era before end-user online searching so having access to your own personal database on your desktop was an advantage. The tools available at this time emphasised information storage and retrieval.
During the 1990s networked desktop access to major bibliographic databases became common and personal word processor packages were widely adopted. These two developments led to changes in personal bibliographic software usage. Facilitating the production of a bibliography in an appropriate format became the strongest reason for maintaining a personal database. Why bother creating your own comprehensive personal database to search when the whole of PubMed is available to search from your web browser? Hence, the knowledge component of reference management was pushed aside and we focused more on the notation of the reference – the arcane codes that we use to refer to documents – and on the mechanics of bibliography assembly and production, integrated with the process of writing a manuscript. A large number of programs competed in this market but two programs became dominant – Endnote and Reference Manager – both owned by the same company. In UK Higher Education site licences covering both products became available and we thought we had the situation covered. But nothing stands still, and a new trend emerged towards online bibliographic management tools.
Endnote produced a web-based version called Endnoteweb. It was originally a cutdown version of the main program, and is still less functional than the desktop version of Endnote. A new product called Refworks was designed from scratch as a standalone online tool. Endnoteweb and Refworks have been competing with each other in the last year or two, launching new versions at a frightening pace. Both are still primarily about bibliography assembly, I feel, but a new generation of bibliographic social bookmarking tools have put knowledge back at centre-stage, Bibliographic social bookmarking is the next wave of bibliographic management.
Eva Amsen a couple of years back pointed out how the adoption of bibliographic social bookmarking could work to the benefit of scientists, comparing it to the way that Flickr has become a large repository of images. Programs like CiteULike, Connotea, Mendeley and Zotero are at the heart of this trend. They vary in functionality but are all free and have grown into the bibliographic world from the web world, whereas the more established products (Endnote, RefWorks) have gone the other way. Connotea and CiteULike are both supported by major publishers; Zotero is an open-source product supported by a US university; Mendeley is a start-up company that has been successful in raising venture capital. These tools have many users now – Mendeley claims to have 8 million references shared, with 100,000 users.
These new tools have also been marketed at scientists and researchers directly, bypassing library support. Librarians need to become familiar with these tools, embrace them if they are useful, and ensure that the tools’ developers are aware of the role of libraries in providing institutional support for bibliographic management for researchers and students.
Libraries have started to take note of these tools. In 2008 in the USA the Northwestern University Library organsed a workshop called CiteFest to compare established tools like EndNote and RefWorks with newcomers such as Zotero, CiteULike and Connotea. Attendees worked through a series of exercises to test the functions of each product. Citefest declared Zotero the winner of their challenge. In 2009 Martin Fenner and I helped to organise a workshop held at University College London, Bibliographic Management meets Web 2.0. It focused on the needs of researchers, who are heavy users of these tools. Representatives from six of the online bibliographic management tools put their products through their paces and attendees had a chance to try some exercises (different from but inspired by the CiteFest exercises). Martin summarised the day on his blog, diplomatically suggesting that since “all the reference managers demonstrated were up to the challenge (though Connotea and CiteULike couldn’t put references into a Word document) it would be wrong to declare a winner”. My own feeling was that Mendeley made the best showing. Martin also pointed out that “the market is developing so fast, that a feature comparison will look very different in 12 months time”. Martin’s comparison chart shows the features of all the products (and others too) in a simple visual way.
Early in 2010 Innovations in Reference Management was held. This event, IRM10 for short, was organised by the TELSTAR project “to showcase and discuss innovative ideas and developments in the use of Reference/Bibliographic Management software.” This featured several of the same products again, but also included some usage case studies. The TELSTAR project is subtitled “Integrating References and Citations into Learning Environments” so it is mainly concerned with student needs. Recently TELSTAR released software that brings the functionality of RefWorks into the Moodle virtual learning environment
TELSTAR are running another event later this month, and Martin Fenner will speak at this, bringing a researcher’s perspective. He gives a taster of his talk in a recent blog post.
Portability is now a key requirement for bibliographic management tools, so we are all looking for online tools that are flexible and easy-to-use. We also want tools that support interaction with research collaborators. Easy sharing of references and the ability to tap into community knowledge is an interesting development but it has not yet become a ‘must-have’ feature. It may not be easy for the new generation of online tools to wean us away from existing tools, but since the new tools are mostly free-of-charge they come with a low barrier to adoption.
We like to use tools that save time and money. I have found that there is great interest in Mendeley at our Institute. People try it out and like it. Ten or fifteen years ago there was a steady drift from other reference tools towards Endnote. Endnote was the tool of choice for biologists. I think we are seeing the start of a similar drift away from Endnote towards Mendeley.

1 Thomas E. Wolff, (April 1992), Database, pp 34-39. Personal Bibliographic Databases: An Industrial Scientist’s Perspective.

2 JHRD Correia, (1990) CABIOS, 6(2), 126-7. Management of personal bibliographic reference using a simple database program

About Frank Norman

I am a librarian in a biomedical research institute. I've been around a few years, long enough to know that exciting new things fall into the same familiar patterns. I'm interested in navigating a path for libraries as we slip from print through to electronic information resources.
This entry was posted in Bibliographic management. Bookmark the permalink.

5 Responses to From Sci-Mate to Mendeley – a brief history of reference managers

  1. Martin Fenner says:

    Thanks for the nice writeup. Comparison charts of reference managers are a good start, but it is much more interesting to look at the bigger picture. And the big picture is changing at a much faster pace than just a few years ago. As for the historical persepctive, here is another interesting reference from the early 1990s.
    B Warling. EndNote Plus: Enhanced Reference Database and Bibliography Maker. Journal of Chemical Information and Computer Sciences (1992) vol. 32 (6) pp. 755-756 doi:10.1021/ci00010a601
    Easy sharing of references and the ability to tap into community knowledge is an interesting development but it has not yet become a ‘must-have’ feature.
    I am really excited about this, but I agree that it is something that is still only used by a small minority.
    P.S. I can’t find the eLucidate article at the link you provided.

  2. Frank Norman says:

    Martin – that’s a nice reference, marking the emergence of Endnote as the major player. I have trouble recalling exactly when things happened back then so articles like that are useful. I think it was another couple of years before the Windows version came along, giving it the cross-platform advantage.My favourite PC reference manager until then had been Papyrus, but they were too slow to get a Mac version and never did bring out a Windows version, so it lost out to Endnote.
    I agree re. the excitement over sharing and community knowledge. We just have to be patient.
    p.s. Sorry re. eLucidate. I got vonfused as they had two issues in quick succession. My article was in the April issue, but I can’t find it on the website now. In any case, most of the article is in this blog post.

  3. Martin Fenner says:

    Frank, one interesting development is that reference management (with first citations and later links to papers, books, etc.) is moving to digital asset management, e.g. managing the (digital) object itself. The iPad and similar eBook readers with an A4 screen size are another step in that direction. Even though I store all papers as PDF, I still print them out for reading. Since last week I own an iPad, and the reading experience is so positive that I might skip that printing step.
    Because the iPad doesn’t support Flash, I unfortunately can’t read the Cell papers in the “Article of the Future” format.
    Thanks for the link Rintze. And looking forward to talk to you at the Innovations in Reference Management Workshop in a week.

  4. Frank Norman says:

    Thanks Rintze.
    Martin – I am sure you are right. I have thought for some time that when we have a decent e-book reader it would change not just book reading habits but also journal article reading habits. Logically the only reason for our current behaviour of printing articles is because we want to read them more easily. A device that provides a oood reading experience obviates this need to print.
    The only caveat is that a single screen is fine for reading one item at a time, or reading sequentially. If you want to compare different articles or book sections, paper makes this easier (unless you have 2 or 3 iPads to hand).