Re-thinking Mendeley

I’ve got a blog post I’d like to fin­ish here, but it needs time for me to sit down and write it prop­erly. One of the things that has eaten my time instead is look­ing over Mendeley. In the past I haven’t used it because I haven’t had a need for it. I already have accounts on Zotero, CiteULike and I have a copy of Papers for my PDFs. I think this could change as I’ve been work­ing with Mendeley accounts for AoBBlog.

If you visit AoBBlog you’ll see a Bibliographies option on the menu bar, and drop­ping down from that four options. Three of these are shared col­lec­tions and Pollination is cur­ated by David Frost, the man­aging editor of the Annals of Botany. These are all admin­istered at Mendeley. I’ve set up Arabidopsis, Ecology and Nutrition as shared col­lec­tions so that when people who know more about Arabidopsis etc. than me sign up, they can keep the bib­li­o­graphy up-to-date. The reason you can see it on the web­site and not have to guess which AoB staff mem­ber is keep­ing the col­lec­tion on Mendeley is that Mendeley now has an API, allow­ing me to pull data out of the site. There’s also a WordPress plu­gin for Mendeley and that’s how I’ve been able to put some­thing quickly on the web rather than try to delve into JSON myself.

So why care?

About a year ago I put together a list of archae­ology art­icles in PLoS ONE for someone. Recently Martin Rundkvist has also had a look at Archaeology in PLoS ONE. Now if I’d thought to make the PLoS papers I’d found a col­lec­tion, then I could cre­ate a page for the bib­li­o­graphy, or else just drop it in below. Being able to add papers to Mendeley with one click and using the API would mean that you’d be get­ting the latest ver­sion whenever you vis­ited the page, rather than the mid-2010 ver­sion. Handling shared col­lec­tions is a bit more prob­lem­atic at the moment, but in the near future a small group of people could make their ref­er­ences pub­lic on a col­lab­or­at­ive project.

There are some prob­lems. The API is new, and it’s not pos­sible to pull all the data out of Mendeley yet. Page num­bers are the most obvi­ous miss­ing data. There’s also a lim­ited query rate, 150 quer­ies an hour. This may rise in the future but for now it means you’ll have to cache res­ults. It’s awk­ward but it’s not inher­ently a bad thing. Cached data are usu­ally much faster, and it’s not like ref­er­ences will need much updat­ing once they’re in a cache. A paper isn’t going to cease being pub­lished in Antiquity in 2003 and be pub­lished in AJA 2005, unless you’ve entered the data incor­rectly. Tags and related papers are the attrib­utes that are most likely to change and even these could often be updated on a weekly or monthly basis.

I’ll con­cede merely list­ing papers, even if it’s an auto­mat­ic­ally updated list, might not sound that excit­ing, so try this.

WordPress allows you to set up cus­tom con­tent types, and one of those types could be ref­er­ences. When the API is fully work­ing you’ll be able to use it to pop­u­late you data­base with ref­er­ences from Mendeley. Mendeley in turn works with Zotero and CiteULike, so you could use these ser­vices to update your WordPress bib­li­o­graphic data­base through Mendeley. Additionally these ref­er­ences need not be yours, they could be your con­tacts so you’re get­ting help from your per­sonal research net­work. You can then write a plu­gin that will work using short­codes so you can include ref­er­ences in posts or pages in a format like [ref-JonesH2009a]. The plu­gin formats the short code into the ref­er­ence style that you want and includes the bib­li­o­graphy format­ted in the right style at the bot­tom of the post. Then you can use Anthologize to render it into the format you want. It could turn WordPress into a social aca­demic research envir­on­ment. To take it fur­ther, you can make the blog private for mem­bers for col­lab­or­at­ive research, or prac­tice open research, how­ever you want to do it. All of this would be access­ible from any­where with an inter­net connection.

It’s not pos­sible yet, but it could be less than a year away. Personally, I think a tool where I can write drafts and which can ask me “Have you con­sidered these addi­tional papers?” when I add a cita­tion would be extremely useful.

As a real­ity check I don’t yet have the skills with JSON or AJAX to write all of such a plu­gin myself but it’s not bey­ond my abil­ity to learn given enough time.


When he's not tired, fixing his car or caught in train delays, Alun Salt works part-time for the Annals of Botany weblog. His PhD was in ancient science at the University of Leicester, but he doesn't know Richard III.

8 Responses

  1. robert says:

    you’ve got an addi­tional “b” in the domain of the link for the aobblog, res­ult­ing in the link being broken.

  2. Mr. Gunn says:

    I live the idea you men­tion there at the end. I won­der if someone, now that Mendeley has an API, would like to write a Zemanta-like plu­gin for WordPress?

  3. Daniel says:

    I have just star­ted a word­press blog and have installed the Mendeley plu­gin. However it doesn’t seem to show related papers, because my blog is new (and doesn’t have much pos­ted) I am assum­ing that this is for a couple of reas­ons.
    1) There is no research related to my tags,
    2)or, In order to post research you must first upload your research to your Mendeley account.

    Have you heard/seen some­thing similar?

    • Alun says:

      I think it’s most likely that there may not be much research related to your tags. A search on Mendeley would hope­fully show if that’s the case.

  1. August 16, 2010

    […] This post was men­tioned on Twitter by Alun, Leicester Blogs. Leicester Blogs said: Re-thinking Mendeley http://​bit​.ly/​9​w​H​3Tq […]

  2. September 1, 2010

    […] soft­ware for man­aging and shar­ing research papers [1]. This pop­u­lar tool has been get­ting a lot of atten­tion lately, and with some impress­ive stat­ist­ics it’s not dif­fi­cult why. At the time of […]