One on one with Bruce D'Arcus, creator of the community-driven Citation Style Language

Bruce D'ArcusMost of our users know that Mendeley can format citations automatically in most word processors. Some may not realize, however, that this bit of magic wasn’t developed entirely by us. Rather, we use tools that were created by a global community of academics and released for everyone to use. I recently had a conversation with the initiator of this movement, Bruce D’Arcus, on where the project is going, what it means to research, and how you can take part.

Bruce D’Arcus is a professor of human geography, with research interest in citizenship and the politics of dissent. His work involves cultural studies and he recently published a work entitled Boundaries of Dissent: Protest and State Power in the Media Age (2006, New York, Routledge)

Bruce, can you explain to everyone what the Citation Style Language (CSL) is and how you got involved with this project?

CSL is an open language to describe citation styles. I initially developed it to scratch an itch: to be able to format my manuscripts reliably in different output formats (HTML, PDF, etc.). At the time I was writing my first book using the DocBook XML language, and needed a citation solution for it. So I wrote the language, and the first implementation to process it.

More broadly, I wanted a language that …

  1. … could describe some of the more complex formatting required not just in the sciences, but also the social sciences, humanities, and law.
  2. … would be independent of any particular application, so that the same styles could be shared by different user communities.

In essence, I have always conceived of CSL and associated implementations as sort of a 21st century analog to BibTeX: one that takes advantage of new technologies like XML and the internet, and which insists on being usable across a range of applications and document formats.

Why is CSL important to researchers?

In the end, if we’re successful with CSL, my goal is that in the future researchers no longer have to think about citation styles: that they simply click a link on a style repository or a journal site, and are instantly ready to go.

I also want people to have greater freedom to use different bibliographic and editing solutions while still achieving the same results. In my own case, for example, I now have the freedom to author in Word or OpenOffice, using either Mendeley or Zotero, or to use my preferred emacs environment to author in markdown text files using pandoc.

So to boil it down, less hassle and more freedom.

What’s the relationship between Mendeley and the CSL project?

Mendeley is one of a number of implementors of CSL. Not only do they use it as their style language of choice, but they have contributed to CSL and related development.

How can researchers get the styles they need to format their papers according to the style required by their publisher or institution?

In the short term, they can go to In the longer term, we will be migrating style hosting to, and hope to create an enhanced style browsing, commenting, and editing interface there.

If the style isn’t available there, how can they make a style for their use?

Right now, for the technically-proficient, they have to create the XML styles themselves. For those not comfortable doing that, the Zotero community is probably a good model. There, users post a style request on the forum, and typically someone will create one for them. While this isn’t the most efficient approach imaginable, we now have close to 1500 styles.

[WG: Mendeley is working on a easy-to-use CSL editor, but it’s not quite finished. Please check out the code and dive in, if you’re so inclined. Here’s an example implementation].

Where do you see the CSL project going in the future?

We’d really like to keep CSL stable for awhile, so we are not really talking about any big changes or new features. The one exception is adding support for the rather annoyingly novel way the latest APA style handles long author lists. That feature will be coming soon in a minor release.

I’m more interested in the ecosystem around CSL, in promoting adoption of CSL, and in making the style creation, commenting, and editing process I note above much easier.

As I mentioned earlier, for example, I would like to move the styles to the repository at, and to allow people to report problems with styles directly there.

I’d also like to see a nice, easy-to-use style creation wizard where people could, in say two minutes, create a fully-functional style that is instantly available for them, and anyone else, to use. It is my strong belief that we need to build on the social approach that has largely accounted for CSL’s initial success. The most efficient way to cover every journal out there is by crowd-sourcing.

Finally, I’d really like to see more journals start to create and host their own CSL styles for authors to download.

What’s the best way for researchers using Mendeley to support CSL?

I would say it is to:

  1. Realize that you are part of a larger community of users. There are many, many styles out there for you to use, and most of them have been developed independent of Mendeley.
  2. Be patient as we build out the ecosystem I note above.
  3. Either create new CSL styles yourself or interact with the community to improve existing styles.
  4. Be open-minded about my idea of an online repository at that facilitates the crowd-sourcing of styles I mention above; post a comment here about your thoughts on that.
  5. Encourage journals to create and host their own CSL styles.

2 thoughts on “One on one with Bruce D'Arcus, creator of the community-driven Citation Style Language

  1. I’m surprised Bruce didn’t also advise Mendeley to be as open as possible in the development process! That’s been one of the great strengths of Zotero, and it would be great to see Mendeley adopt some of the best aspects of that open-source, open communication approach. For instance, the Mendeley feedback forum, with its voting for features etc. is nice, but many users find it frustrating, as there is no real collaboration. From time to time the Mendeley team checks in on issues, but there’s not much of a sense of where the software is going and whether it will adequately address users’ needs in the future.

  2. Jon – His main interest is in adoption of the CSL standard, but I see your point about the Feedback forum. It is hard to follow what’s going on and to get a general overview of what progress is being made, but we do make an effort to participate in conversations wherever they may be happening on the web, so I hope that counts for something 😉

Comments are closed.