Organizing your Library, Zotero and Mendeley, an overview and ideas, and because the title was already long I decided to write more the put it a bittler longer.

Some days ago we started here a discussion about which program could be the best to organise a researchers library, papers after papers, issues after issues.. Let’s try to summarise,

for MAC user you have Papers, I don’t have MAC therefore I won’t speak about it.

On the other hand, Zotero and Mendeley have presented themselves as two very useful tools for library organisation. Both are free and run on every system, so to speak. Summarising their characteristics;

Zotero is an add-on for Firefox, while this works very nice when you are searching and adding papers directly it might became something not so comfortable when you are just surfing on the web or when you just want to read with documents and leave the browser alone. Firefox still consumes some valuable resources of the computer, for those using old computers that might be a problem. Mendeley instead, works as a separate tool, independent from any other program. And however, both have the possibility to add to the library directly with one click.

Mendeley is able to, if you add the PDF directly to the Library, retrieve all the information about it, as long it is text readable PDF and not just a PDF made up from pictures, i.e., old scanned papers. You can do the same just specifying the DOI or the PMID.

Personally speaking, once in your library I found much more comfortable and easy to find what I am looking for in Mendeley than in Zotero.

Both programs have large community online, and both allow you to synchronise with the server. Comparing, Zotero gives you only 100 MB free at the beginning and Mendeley 500 MB + 500 MB shared. However, Zotero is cheaper in the upgrade, in case you want to have extra space.

In terms of group synchronisation, Zotero has some several conflicts when dealing with version of different languages. I never tested this option in Mendeley.

Zotero can’t open the PDFs by it self, depending always on your PDF reader. And because is dependent on Firefox it will often be necessary to change the way Firefox open PDFs… On the other hand, Mendeley cames with an outstanding characteristic, in my point of view. Not only it allows you to open PDFs directly on the program itself, being completely independent from external program, but also it allows you to underline and add notes. For linux users that is undoubtedly gold! Finally we can forget about trying to run Adobe Professional in our systems and use a native program. One can argue that Okular already did the job. Yes, but no so well done as Mendeley. Actually, Okular was imperfect enough to become unusable, at least last time I used, around 6 month ago. This characteristic for me puts Mendeley far above Zotero.

Telling a bit more, Zotero stores the data in your computer in a non-human readable way. That is, you cannot find you paper if you go just surfing in or File Browser. Mendeley do much better the job by creating a tree of files according to your preferences.

Citation and Word Plug-ins: I have already used both programs to do the citation and references task. As far as my knowledge (yes, as a good researcher I didn’t want to investigate more into the issue :D) Zotero plug-ins can reach more programs than Mendeley, p.e., Mendeley plug-in do not work with Microsoft Office under 2003, while Zotero does.

Zotero is much more powerful in citation and references because it allows you to surf freely on your library while adding each paper at a time, when adding multiple references. Mendeley still doesn’t allow you such thing, you have to select all the papers without changing the Library folders, that is, you have to found your paper in the list of “All Documents”, which when dealing with hundreds or thousands of papers might be a little painful. Moreover, Mendeley does not allow you to edit multiple references, say (1-4), you have always to redo it, you can’t just add the (1-6) later without selecting again the 1-4.

Regarding references styles, Zotero is a bit more evolved, however, Mendeley can retrieve the citation styles from Zotero.

So, how to decide, should one use Zotero or Mendeley?

If you don’t care about PDF annotation and you are okay with the search engine of Zotero and the fact that it is and add-on, I would recommend Zotero, without any doubts.

However, if you really appreciate PDF annotation and easy search, Mendeley is your solution. However, by now, one cannot rely only on Mendeley, it is still a bit painful to do the citation work.

Therefore I come across with the solution of using both simultaneously :). By now I am keeping in Zotero just the information about the papers (without the PDFs), synchronised with the server, 100 MB are enough for the task. And I will use Zotero plug-ins for the citation work. While keeping Mendeley with the PDFs along with all the other information, and by now 500 MB are enough for my Library.

So, you can argue that at the end I am using more resources and more confusing strategy. Yes, but my computer is powerful enough to deal with the task and for my strategy of work it is painful, in one way or another, to use just one of the programs. However, in my opinion, Mendeley has much more possibilities to develop and to become much better than Zotero, simply because their bases differ. Do not forget that Zotero is already at version 2 and Mendeley stills in 0.9.

Whatever the case, I really appreciate the community for developing such tools! Both are excellent, and without them.. life as a researcher would be even more painful :D

Remember that this post is just my opinion about the issue, was my intention to share with you this idea, might it also be useful for you.

Looking forward for your answer
Best Regards
João Teixeira
:)

16 thoughts on “Organizing your Library, Zotero and Mendeley, an overview and ideas, and because the title was already long I decided to write more the put it a bittler longer.

  1. I am one of the founders of Docear, which is a new software for organizing, creating, and discovering academic literature. Today, we released version 1.0 of Docear after a ~2 year beta phase. If you are interested in reference management, you might want to have a look at Docear. The three most distinct features of Docear are:

    1. A single-section user-interface that differs significantly from the interfaces you know from Zotero, JabRef, Mendeley, Endnote, … and that allows a more comprehensive organization of your electronic literature (PDFs) and the annotations you created (i.e highlighted text, comments, and bookmarks).

    2. A ‘literature suite concept’ that allows you to draft and write your own assignments, papers, theses, books, etc. based on the annotations you previously created.

    3. A research paper recommender system that allows you to discover new academic literature.

    And Docear is free and open source and available for Windows, Mac, and Linux. More information can be found in our Blog, including a detailed explanation of what makes Docear superior to Mendeley, Zotero, etc. (at least in our opinion :-) ). http://www.docear.org/2013/10/17/docear-1-0-stable-a-new-video-new-manual-new-homepage-new-details-page/

    If you don’t like reading, there is also a 6 minute introduction video on our homepage http://www.docear.org ;-)

    In case you are using a BibTeX based reference manager such as JabRef (and you don’t want to switch to Docear), you might at least be interested in Docear4Word http://www.docear.org/software/add-ons/docear4word/overview/. Docear4Word allows you to insert references and bibliographies from BibTeX files to MS-Word documents. Hence, it makes writing papers much easier, since e.g. JabRef has no own MS Word add-on.

    Finally, I would like to point you to a recent Blog post I wrote about what makes an evil reference manager. Maybe the post helps you deciding which reference manager to use (even if it’s not Docear). http://www.docear.org/2013/10/14/what-makes-a-really-really-bad-reference-manager/

  2. Hi there, good write-up. I just came to the conclusion, before reading your blog, that i would use the two tools. So, this has confirmed a lot of my own conclusions.

    Just one question though, how do effectively sync or import your Mendeley collections into your Zotero…?

    I’m guessing Export the collection in Mendeley using RIS format and the import it in Zotero? Is that the most effective?

    Shazana.

  3. Olá,
    Sei que não é um blog ou thread sobre Mendeley, mas algumas coisas complicam demasiadamente o software. O Ricardo (Vidal) aformou que o Mendeley procura metadados nos PDFs. Senão, procura no próprio PDF por informações relevantes. Tudo automático. Sóque as informações “caçadas” no PDF são, MUITAS vezes, incorretas ou incoerentes, o que dificulta a indexação do PDF. Porque não uma caixa perguntando “OK? ou Não importar” para não haver este problema?
    Um abraço e parabéns pelo review (João) e ao Ricardo pelo software.

    José Ricardo Borba
    Porto Alegre – RS – Brasil

  4. Hi,

    Yes, you are right. I had already thought about using Tags with that purpose. Looks nice really.

    I like a lot the fact that you can retrieve all the information giving the DOI or PMID.

    Cumpz,
    joao

  5. Hi again,

    Regarding the PDF importing and data extraction, it’s a quite a bit more complicated than just a few regular expressions. You can just imagine the various types of PDFs. Not only the formatting of the documents but also the way the documents were created with restrictions, watermarks, etc. It’s something we’ve worked on since day one. Mendeley was among the few reference managers that could auto-extract the metadata from the PDFs and populate the reference info without accessing the web.
    Although Mendeley works best when you use the Desktop and Web components, you could work solely with the Desktop application without ever connecting to the web with some degree of success. Obviously, many features such as collaboration and sync (among others) would not work/be available.

    I wanted to address a question made by an earlier comment regarding subfolders. At the moment we don’t support these because we believe that you can achieve the same purpose through the use of tags. With our search and filtering options, you can easily pull up all papers with a particular tag or set of tags. Look at the tags as metafolders :)

    Cheers,
    Ricardo

  6. A bit of technical blabber.

    Parsing PDFs is a major pain in the rear. I’ve tried it myself before. I wrote a Python script that parsed the PDF file and extracted the DOI wiht regular expressions. But it’s very complicated because there’s no defined location/format for this number. I could eventually get it with a couple of conditional expressions, but only for a limited number of journals.

    How do these programs extract that metadata?

    I assume that they probably do the same. An idea I had when writing this script was to build a database with the regular expressions needed to parse the DOI. For example, if “Nature” was found in a suitable location (I remember I was always looking for a specific “piece” of the first page of the PDF, the footer) it would fetch Nature’s DOI regular expression and use it. It worked neatly but the hassle of building and maintaining this database was just too much.. Besides, my goal was to get the info on the PDF so I could index it and thus make my whole database “searchable”. And then I found Papers :) Shame I only came across Mendeley a few weeks later!

    Regarding old PDF files, the only option really is using OCR to first translate image to text and then perhaps treat it as a “regular” PDF. But this is a long shot because neither step (OCR or PDF Parsing) is reliable enough. I’d say this would improve the success rate by only a meager percentage.

    Any comments on this Ricardo?

    Cheers!

  7. “we stand on the shoulders of giants”!

    No words, it is just perfect! It defines perfectly what I was thinking. XD
    Off course I was kidding when I wrote: “…that we are in the XXI century, and we do not need such things from the last century…” XD It was just to poke you.

    Well, as I thought, there is a simple solution but not as quick and easy as with Zotero. But still, there’s a solution and something to upgrade in the previous versions of Mendeley. I have asked you that, because normally the older PDF files does not have embedded metadata, and that way, is very difficult to get something from them.

    About the Zotero group collaborations, it is probably not as developed as in Mendeley (another reason to test the program), but you still can share the papers and you can join a community to collaborate and work more accurately.
    You control the functions with the add-on but you share and create the forums within the site (www.zotero.com).

    Good work and keep in touch.

    Cheers,
    Andre

  8. Hi Martins, Thanks for the great feedback. We really appreciate the effort you put in expressing your thoughts and this in turn drives us in the right direction.
    Let’s get some technical bits and bolts out of the way:
    Regarding citation styles, I can tell you that we use the latest standards for citation style language (CSL 1.0), as does Zotero. We are currently working on improving our citation style editor that will then be included within Mendeley and allow users to tweak and generate their own citation styles. This is extremely useful for schools or countries that have their own citation styles, such is the case of Portugal. You can get a preview of the citation style editor here, however we’ll have a built-in version.

    As for collaboration, I was unaware that Zotero allowed for private groups with commenting, notes and highlight sharing with members of your group while working within the application. I’ll give it a look :) Anyhow, the important thing is that you’re getting your research done and collaborating with colleagues.

    For older papers, I would never undermine previously published work no matter how old. I actually cited papers from the 1800’s in my Masters thesis! Remember that “we stand on the shoulders of giants”!

    To better understand what goes on with metadata (document details), when a PDF is added to Mendeley, the software attempts to extract the metadata if available. It also extracts the full-text for search purposes. If the metadata is unavailable or incomplete, the application looks through the text and attempts to find the appropriate data (this can be quite difficult for obscure journals). If all this fails but a DOI or PubmedID is found, it will attempt to complete/correct the data with a web look up.
    So, if your paper is from 1930, is a scanned image, has no embedded metadata, I suspect that the only option would be to run some sort of OCR on the document (and that is not always successful!). So, for older papers you might need to find the DOI or some other identifier or if it’s really old and unlisted, you’ll just have to add the info by hand. But I’d suspect that to be a rare occasion.

    Ok, and one final thing. I’m not a developer of the application but am part of the team that has helped define it with the great help of researchers such as yourself and your other blog buddies. :)

    Cheers,
    Ricardo

  9. Well, it was nice to read the direct opinion and the advise from a Mendeley programmer! XD
    It is true that with the suggestions and options that you have exposed, I am more convinced. But… as everything in life and for the advance of Mendley, I will do the role of the devil’s lawyer. I am using, for my citations and to organize all my papers, Zotero from Firefox, and for the moment it fits just right with my needs and expectations. Even the fact that we are working with a simple add-on, from my point of view is an advantage. You have said that differently from Zotero, Mendley “offers more than just on-line storage but also collaborative options to set up private groups within which you can work on research projects, share PDF annotations (notes, highlights, etc) and also the online aspect of being able to work with other Mendeley users.”! But, apart from the pdf annotations and the aspect, Zotero can do the same! I am using it for those proposes as well.
    Another thing, for the older pdfs, 1930 for example, how can you import with Mendley the bibliographic information? You will probably say, that we are in the XXI century, and we do not need such things from the last century, but it is really, really, very important from the scientific point of view. And I want to do it with one click as well! XD
    Anyway, as you suggest and I did not know before, you can import the information automatically from Zotero with Mendley, and that is really something. XD

    Just for now, you have convinced me, to use Mendley and to test it. You have been doing, really a great job for the community and you deserve to have, at least, a positive feedback from your work. As a member of this blog, I will give my contribution.

    Talvez um dia possa vir a ser uma boa escolha para representar academicamente o Mendley! Bom trabalho Ricardo e obrigado pelo teu comentario. Abraço

  10. Antes de mais! Uau e Olá! Não esperava mesmo um comentário de um ‘developer’, e ainda para mais português. eheh, como se custuma dizer, “onde mija um português, mijam dois ou três” :D (já agora como é que encontraste o post?)

    Now, to business. I really don’t mind anything you can say, in fact, I appreciate a lot your suggestions. I apologise for my ignorance in some points.

    I was now checking better the citation issue, it is true that there is a button “Merge Citations”, to be honest I didn’t tried it the first time, I was now checking it, and works very well!! It is different from Zotero, but however it works quite comfortably.

    The integration with previous Word versions, I agree with you, we are almost at 2011, but I mention it because a friend of mine at work got an old version of Word and Mendeley plug-in didn’t work. So he went on with Zotero.

    Just about the Zotero-Mendeley sync, this option, as far as I understood allows you to add in Zotero and automatically it will also be added to Mendeley. Just the paper profile or also the PDF? Does it work the other way around? (I ask but I am wondering if it makes sense at all to work the other way around.) I didn’t tried it because I don’t want to mess up something now :P

    Something would like to ask you (sorry for making you a tutorial :P), by now we can’t create subfolders, right?

    Another ‘something’ I would like to suggest, the “Delete” button is to much close to the “Add Document” button :P already happen to me that I delete a file without almost noticing.

    For sure I will join the advisors section! I am really excited with Mendeley!

    Keep on the good work, (looking forward for the updates :D)
    we will speak again soon,

    Abraços,
    João Teixeira
    joaomcteixeira@gmail.com
    @Ubuntu 10.10

  11. Ola’ Joao,

    Excelente ideia. Gostaria de saber qual o resultado do vosso trabalho colaborativo e como e’ que o processo decorreu.
    Ja agora, conhecem ja’ o programa de representantes academicos que Mendeley tem para ajudar pessoas a divulgar o software nas suas instituicoes? Talvez seja do vosso interesse participar. Mais info aqui: http://www.mendeley.com/advisors/join/

    Cumprimentos,
    Ricardo

  12. E não é que o Teixeira trouxe alguém da equipa do Mendeley ao blog? Grande post rapaz :)

    Caro Ricardo, o João Teixeira será mais indicado para responder ao post, mas desde já obrigado pela participação e esclarecimento! Nós estamos a planear fazer testes entre o Zotero e o Mendeley entre nós os 6 e escrevermos um artigo com mais organização e talvez mais aprofundado estabelecendo uma comparação clara entre ambos.

    Abraço!

  13. Ola’ and thanks for the great blog post. I just wanted to add a few things and possibly clear up a few others if you don’t mind.
    You mention that it is cheaper to upgrade with Zotero but forget to mention that Mendeley offers more than just online storage but also collaborative options to set up private groups within which you can work on research projects, share PDF annotations (notes, highlights, etc) and also the online aspect of being able to work with other Mendeley users.

    As for the citations styles and the plugin, I’m quite sure we support Word 2003 and up. I can understand that there are versions before 2003, but it’s now almost 2011… :-)
    Still on citations, you can navigate within Mendeley and pick your references or even search and find the ones you would like to insert. However, we do agree that some work needs to be done with the plugin and the citation style editor. In fact, we’re working hard on that and you should see some cool new features pop up soon!

    Finally, I’d like to remind you that Mendeley has a synchronize with Zotero function. Go to your preferences and you can select to keep your Zotero library synched within Mendeley. This way, you can move over or work with both applications with less hassle.

    Once again, thanks for the great overview and for the feedback!

    Cumprimentos,
    Ricardo

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s