Sometimes it feels like all we do, as scientists, is read other people’s work. In which case, it’s not surprising that the first software that was impressed upon me as a new postgraduate student was for reference management. At my university, we are encouraged to use EndNote, so this is what I started on. A year or so back, on a personal mission to move to free software, I trialled Mendeley and never looked back.
Setting a “watch folder” has probably been the most useful aspect of Mendeley over my previous reference software. Within Mendeley, I can mark the folder I save my PDFs to (or several folders, if you have a better system than me) and Mendeley will magically add any new files into my reference list.
With EndNote, I always ended up behind in my references due to the nature of searching. You know how it goes: you read one paper, from that you find three references you want to look up, while you’re searching for those three, you see a couple of titles that look interesting, or work from someone you’ve cited before. Click click click, and you’ve downloaded a dozen PDFs in one sitting. But because you’re not necessarily working through a search engine and adding citations as you go, these PDFs can languish without a corresponding citation entry.
Now when I have a frenzy of searching like that, I know as long as I save the PDFs to my watched folder, the next time I open Mendeley, some computer elf will automatically pull up the citations for the new PDFs. It’ll flag any that need attention for me to check. I can’t say everyone works this way, but this feels far more intuitive to me than needing to export a citation every time I download a paper.
If Mendeley does flag a citation as needing checking, it’s often a scanned PDF. Mendeley will have done its best to find the correct citation, but it knows it might be wrong and asks you to check. You get two options: yes this is fine, or search the title in Google Scholar. In this case I’ve generally found that if I correct the “title” field (often it will have inserted a bunch of junk) and search in Google Scholar, Mendeley will then find the correct citation.
This leads me to my next favourite feature, the ability to open PDFs within Mendeley. Great for when you just want to check something, even better if you want to check the citation provided: with the PDF open, and the citation sidebar, it’s easy to cross check and make sure everything looks good. Should you want to open the full file, just right-click and “open externally”. This is far better than my previous method of chasing through a full folder of PDFs for the one I want.
If you use LaTeX, Mendeley has great functionality with BibTeX files. I have all my citations in one Mendeley library; every time I update Mendeley it updates the corresponding .bibtex file, then the next time I LaTeX my thesis document, it reaches in and pulls up the latest version of the .bibtex file. I switched to Mendeley at the same time I started using LaTeX, so I can’t compare this functionality with other software, but I have been so impressed with how easy it is.
Other features include PDF annotation, and free online storage space. If you use more than one computer, you can move your citation library into the online storage space and keep your library “synced”, then you can access it from other computers (although the software is free, you can upgrade your account and pay to get more online storage space). Given that I don’t use Word for academic writing anymore, I don’t know how the Word plugin compares, but after a quick glance, it did all the things I’d expect it to. I particularly like that within Word, you can start typing the name of a citation and the plugin will bring up suggested citations.
All that said, Mendeley does have some issues. A major one for biologists is the difficulty in italicising within a citation. There are workarounds for this issue, but this seems to bother some users.* I find the programme slow to open, which may be due to the programme checking the “watch folder”; but once open, it runs smoothly. Finally, I don’t believe the range of citation styles is very good, but you can make your own or adjust a pre-existing style. None of these issues have stopped me using the programme, but they might be more important to some people.
It’d be great to hear how other people manage the masses of citations required in science. What combination of reference software and word processor do you use? How well do they work together, and have you tried any others?
For more information on Mendeley, check out this previous article.
*If you’re wondering, Word users can insert html tags (<i>Species name</i>) within their Mendeley citation, while BibTeX/LaTeX users need to turn off a setting (Tools – Options – BibTeX – turn off “Escape LaTeX special characters”) and then use \textit{Species name} as you would within text.
Have you ever tried Jabref? It seems to share all the features of Mendeley, and seems very powerful. Unlike Mendeley, it’s open source, which is nice. I use it with LyX, and it integrates well. The Jabref devs are very open to feedback as well.
For creating a library ReadCube is far better than Mendeley. It has the folder watch and export options that Mendeley has along with meta data extraction. In addition you can search Pubmed and Google scholar, and download referenced and citing papers within the program. Papers are even cross-referenced.
Probably the most impressive feature though is that it scans your library and recommends articles.
Downsides are it doesn’t have the word plugin Mendeley has and as its in development it is a little buggy.
Thanks for the comments; it’s great to hear what other people use so fellow readers can know their options.
Mendeley is currently free, but it’s not open source. And Mendeley is also lacking many journal styles. I only save the PDF:s that I can’t access freely on the net, no point in saving local copy’s of ALL PDF:s. My experience is also that it never works when a reference management program tries to collect references from many difference sources, in this case differently formatted PDF:s, it’s usually a lot safer to try to retrieve ALL references from the same source e.g. PubMed, with a reference management program that can search PubMed. When I tested Mendeley I also had big problems with the Microsoft Word plug-in, it constantly crashed Microsoft Word. Another problem I had was import of references in different formats into Mendeley, the import filters was not working as they should.
I’ll stick to JabRef, it’s open source and will stay free, and as a user I have much more control over the software, compared to a commercial product like Mendeley, where it’s practically hopeless to get support.