Wednesday 28 July 2010

Eeeny meeny ... another meltdown

This is terrible. Another flaming can of worms. Don't you people understand how difficult it is for Crail to make decisions?


EndNoteWeb? Mendeley? Zotero?

I have really, really, wanted to use Mendeley, since seeing that charming young gentleman [was that Jan Reichelt?] who demonstrated it last year. Freely available, devised by PhD students, iTunes for references, not dominated by Big Boys, drag and drop for god's sake, pretty colour, what's not to like? Plus in June received notification that they are working with Caret to investigate using it as a repository.... though as Caret admit themselves, a lot of their projects fall by the wayside. However, as long as they drop Connotea that would be a start. Would like to know why Mendeley, which initially received equal billing in 23Things, has dropped down the pecking order. Is this because hardly anyone in the Cambridge libraries is actually using it apart from Isla? Interesting.....

Anyway, yes, back to the point. Some of these Web 2.0 wotsits are interesting, but can be taken or left, without our, er, clients being inconvenienced. But now we are talking reference management systems, and things get more serious. We should promote these tools ergo we have to have a degree of competency in them.

Trouble is, Mendeley doesn't quite work. Same as EndNote. Same, probably, with Zotero.

What is the point in telling PhD students to get organised and park their refs in one of these when it is all such a flippin' FAFF? If I am chucking the toys out of the pram in frustration, one can hardly expect them to stick with it. All that dropping into Notepad and converting to .ris, or dropping into EndNote and then into Mendeley. Why do I have to DO all that?

I hadn't even bothered to look at Zotero until now, more stuff on the Firefox pane, I can resiste anything but temptation etc. But oh heck, I have to confess that clicking that little booky-thing in the search box and having the ref drop in IS fabulously easy ... ergo alarmingly tempting. BUT the Computing Service warns against it for large nos of refs. So now I've got [personal] refs in EndNoteWeb, in Mendeley and in Zotero. Mostly the same ones. More Augean messes.

Ah, life was easy when you just showed the students how to drop from WoK into EndNote. Job done, sidle off before they started asking about the other databases and things got decidedly sticky.....

That's it. I'm propped up at the Hommage a Huysmans and it's only 11 am.

17 comments:

  1. somehow Cambridge librarians seemly oddly troubled by the ease of use of Zotero:
    http://libreaction.wordpress.com/2010/07/27/you-say-zoh-tair-oh-i-say-zoh-teer-oh/

    I've written several articles and am currently completing a dissertation on Zotero without any troubles - and I know of several dissertations (in the social sciences, i.e. with large bibliographies) using Zotero.

    Zotero slows down somewhat at collections with >5000 entries, but that was much worse with earlier versions and has been mostly addressed. There are some rare cases where the syncing of libraries to the web doesn't work for very large libraries, but that's about it.
    Zotero uses a straightforward sqlite back-end, which can handle much larger databases. So I'm really not sure where the idea that Zotero isn't suitable for large databases comes from.

    My guess is that a lot of people look at Zotero, see that it's free and open source and then see how their university pays several thousand dollars/pounds a year for an Endnote campus license and then find that Endnote _must_ be more stable and better for large projects. Because, why else would anyone pay for it? Which is, of course, a bit of circular logic.

    ReplyDelete
  2. Thank you so much, Latinamericanist - so glad to hear from a real live user, rather than a tinkerer as most of us in Cambridge libraries probably are. As a non-techie, and as someone who doesn't want anything fancy, merely a clean way to store refs, I am most certainly NOT troubled by the ease of use or the fact that Zotero and Mendeley are free, honest! I don't even have EndNote itself. I guess what worries me is committing to creating one database, and then discovering later that there is something I need the software to do which it cannot do easily, which as a non-techie I hadn't foreseen. There is also the worry of recommending one service to students above others, and the poor kids coming back and complaining it cannot do something they need to do as well.

    ReplyDelete
  3. I have high hopes for one of Zotero, Mendeley or Endnot being the solution to a current problem (compiling lists of recent publications for a journal). Haven't had time to look at any yet (have my own little meltdowns over other things right now), so was feeling a bit anxious reading Miss Crail's post. Latinamericanist, it's good to hear that you're making Zotero work. I don't imagine I shall be dealing with >5,000 references, so I'm still hopeful that I'll find my solution in this Thing. Just as soon as I get round to looking...

    ReplyDelete
  4. Depends where you are sourcing the refs I expect. It's getting the refs from database to ref mgt system that's the problem. At least you aren't trying Newton to RMS, as I am, though Zotero CAN mostly do that

    ReplyDelete
  5. @Miss Crail - thanks for the nice welcome -
    I do agree very much that librarians should be relatively neutral with respect to reference software - e.g. I think chosing between Zotero, JabRef and Mendeley depends largely on ideosyncratic factors.
    However, I'm not sure that should be the case for Endnote anymore, considering that there are now two free and (I would argue) at least equivalent products available - for my part I'd rather have our library spend the money (if I remember correctly our campus license is close to US$ 10,000) otherwise.

    @ Girl in the Moon:
    Zotero is unbeatable if your goal is to compile relatively large lists of articles - while people differ in how they prefer to organize research, Zotero is by design super-fast to get lists of articles either from journal pages or from google scholar or other databses.

    ReplyDelete
  6. Thanks again, Latinamericanist. You are certainly putting a persuasive case for Zotero!
    I have tended to use EndNoteWeb for work purposes because we push Web of Science, so ENW comes as a bolt-on, making downloading very easy. But not RSS feeds.
    With Mendeley I can do the feeds but the downloading is a bit of a pain.
    It seems to be the case that A database refs will drop into X ref management system easily, but B database refs won't, though they will drop into Y RMS, and so on, compounded by us promoting 3 or 4 main databases.
    I guess I should seriously compare and contrast Zotero and EndNoteWeb, but I'd hate to abandon Mendeley.

    ReplyDelete
  7. Thanks for the tips, Latinamericanist. From what I've read though, Zotero is a browser add-on? I don't have my own laptop, so my work tends to be done from any number of computers scattered across town (home, work, friends, libraries...). So it seems to me that a browser add-on wouldn't be so convenient. It seems that Mendeley has a Web version, which sounds just like what I'm looking for. If only I had time to trying it out!

    ReplyDelete
  8. Zotero has a web section too, which can (I believe) be synced with your computer. I haven't been able to try that bit out though because the word plugin is throwing up all kinds of error messages now :/

    ReplyDelete
  9. Oh? Well, then, I shall have to try both! All my current knowledge is based on a quick reading of the Cam23 post, so thanks everyone for providing extra info.

    ReplyDelete
  10. Just wanted to drop in here to announce myself. I'm the academic community liaison for Mendeley, and I can address some of the issues here.

    From reading this post, and Girl in the moon's, it seems that much of the trouble is caused by what I call "data friction", or the resistance to moving data from one location to another. The easiest thing would be to just have one database and one program that does everything, and Mendeley is certainly trying to be that all-in-one solution. It's true that our coverage in the humanities is a little lighter than Zotero, but that's getting better.

    Before addressing any specific points, I'd like to understand more clearly what the issues are. Starting from the collection of PDFs than most researchers have in some form of fashion, how would you go to a formatted bibliography in a word processing program?

    ReplyDelete
  11. Thanks Mr Gunn. It's great that a Mendeley representative is contributing here because, as stated, I'd prefer to opt for it.
    Yes, if one already had lots of pds sitting around, then Mendeley would be perfect with its drag-and-drop feature, which has mostly worked for me, and would be great for PhD students who had amassed lots of pdfs and now needed to create refs and a biblio. I'd unreservedly recommen it for that.
    However, for part of my work needs, having to compile lists of staff publications from eg WoK or PubMed [not all of which we would be able to download as originals] Mendeley required tweaks like dropping into EndNoteWeb first then copying across - bit of a nuisance. And for my personal use I need to drop catalogue records from Newton or other, into a RMS.
    So you could say, for items read that already exist as pdfs on a computer, Mendeley is perfect. And I do love the 'import to Mendeley' function in 'Favourites' which sometimes works a treat.
    But many of us have quick-and-dirty 'looks interesting, must check this out later' lists [of papers so Delicious perhaps not suitable], or book records. And if I want to drop Newton records into Mendeley it has trouble. I keep going back to Mendeley and then giving up again. I'd like an easy bibliographic life and I'm pretty certain most PhD student to too. Mine are all biosciences by the way

    ReplyDelete
  12. @Niamh I didn't need to use the Word plugin to get the synching to work. It worked for me by 1) making sure I was logged out of Zotero online. 2) Hitting the 'synch' button (like a little green circular arrow, at the top right) on the Zotero pane in Firefoz 3) Filling in account details when prompted.

    And for 'quick and dirty' citations, you can (I think - not looking at it as I type) select ref(s) in the pane, and choose edit-copy, select the style, and then it copies it to the clipboard, and you can Ctrl+V it into the doc. Not as quick as a plugin, but an OK workaround, perhaps?

    ReplyDelete
  13. @Girl in the Moon:
    quick and dirty citations can also be generated from Zotero simply by drag&drop to Word/Ooo or any text editor (though I've read - but not confirmed - that there have been google doc trouble recently due to changes in how that works). You can specify the default output style for that in the Export tab of the Zotero preferences. If you frequently want different styles, using the method you describe works better (the option is called "Create Bibliography..." in the context menu).

    That said, of course Niamh's word integration should work - have you looked here
    http://www.zotero.org/support/word_processor_plugin_troubleshooting
    or asked on the Zotero forum?

    @Miss Crail - Regardless of what Mr. Gunn says, I doubt that quick data import in Mendeley will ever be as smooth as in Zotero, simply by design (note that there is a trade-off involved - Mendeley allows you to do that from any browser, Zotero just from Firefox).
    However, one nice thing Mendeley has recently implemented is the ability to one-click import (or even sync?) Zotero data. So if you prefer the management aspects of Mendeley, you could collect mainly with Zotero and work mainly with Mendeley - I haven't tried this, but it's sugested e.g. here:
    http://chronicle.com/blogPost/Using-Mendeley-for-Research/25627/
    I know that's adding one more tool - but in that case you're using Zotero really as not much more than a better bookmarking service, so you wouldn't really need to learn much about how it works beyond clicking the URL-bar item.

    ReplyDelete
  14. Thank you again, Latinamericanist, for doing the research I should have done. My excuse - it is the weekend!
    Your suggestion sounds painless enough to be, what shall we say, a dodge, rather than a tweak. And if that proves to be the case, it might be as close to problem solved as we can hope for at the moment.
    Thanks so much for stepping in here

    ReplyDelete
  15. you're very welcome - good luck with your project.
    And the point is that neither you nor I had to do research for this - I knew this before, so I could just write you a quick note without you having to google, read instructions etc...

    ReplyDelete
  16. Oh stop it! I'm lazy enough as it is already.
    But seriously, it has been great to hear from an actual user ; ergo it is seriously shameful that none of us in Cambridge libraries appear to be au fait with software we are recommending. As Celine says in http://thingblogging.wordpress.com/2010/08/01/standing-in-the-kitchen-at-the-zotero-party/, even if we don't use the tools ourselves, we should be helping people with them. And there certainly seems to be niche market!
    And I am glad to hear from Mr Gunn too, though he might regret putting his head above the parapet if I keep pestering him with silly questions

    ReplyDelete
  17. The UL makes some effort to advise on the most popular reference management tools, and how to intergrate them with library interfaces in Cambridge.

    http://www.lib.cam.ac.uk/toolbox/rmt.html

    As a firefox user, I have a real love of Zotero, but have recently been impressed with the way the Menderley client can synch with Zotero, giving you the best of both worlds.

    There is a massive market out there. I often get queries with smaller RMT packages that I have to try and handle.

    As Librarians, I think the only way to go is to try and use these tools in anger, if we get the chance. When I used Endnote a few years ago to compile stuff for a last report, I quickly became au-fair with its shortcomings.

    ReplyDelete