A Love Letter to Zotero (with a caveat)

I am a recent user of Zotero, working in a Microbial Biochemistry lab in Colorado. We are trying to implement the use of Zotero first in our lab and, potentially, across the department of the University and beyond.

There are multiple reasons for the enthusiasm we have with the software. Of note, the open source distribution and the integration with browsers are fantastic and have helped making the transition from commercially available reference managers a breeze.

But I have a caveat to this, as my title indicates. I am, personally, an above average user of computers: I do a bit of coding and I am not really afraid to dig into code. That is not true for rest of the users in the lab at all, much less across the department. So, when we had to add revisions on the first publication we submitted using Zotero, we were faced with the need to change the bibliography to include, among other small modifications, the DOI.

Since it was the first time we were faced with such a task, we were surprised to find out that Zotero has no intergrated way to modify styles. Moreover, the Visual CSL Editor is not, I am afraid to say, user friendly. I eventually solved the problem by digging into multiple CSL files and copy/paste pieces of code to solve our issues, so this is not a post asking how to do it.

What I want to tell the community is, for the people involved in developing future version of this wonderful piece of software, an intuitive, integrated way to edit styles should be included in the base package. I do not know if the same happens in all the fields of science, but journal where we normally publish change their bibliography formats constantly, at a pace that having an ever-evolving library of style may not be the best option. I also think that, despite being a great community service, reliance on community to generate custom made styles is a poor alternative to what seems to me would be a great feature to add to the software.

I hope this reaches the ears of the developers, to whom I still want to imprint the idea that their work is amazing and I would love to be able to recommend their work to more people. Keep it up!
  • What journal were you submitting to? If they are requesting DOIs, the official CSL style can be updated.
  • We were submitting it to ACS Infectious Diseases, though I would imagine any journal based on the ACS style (American Chemical Society) also requests it.

    My point was not requesting a change on the official CSL, though I appreciate it. It's more about having an intuitive tool for any user to modify their style. The Visual CSL Editor, unfortunately, is not such a tool, despite being quite powerful.

    Thank you for the help, though!
  • Point taken and we'd certainly like to make the visual editor easier to use, but that's not an easy task. The editor was developed under a six-figure grant by Mendeley with Columbia libraries that included work with Mendeley's UX team. Turning metadata into precise citation styles involves a somewhat dizzying number of possible changes that are hard to turn into an easy-to-use GUI.

    The one part I'd disagree with is this:
    I do not know if the same happens in all the fields of science, but journal where we normally publish change their bibliography formats constantly, at a pace that having an ever-evolving library of style may not be the best option.
    I have a very good sense of journals in any discipline and their citation styles and it's simply not true that journals anywhere change citation styles all that often -- unless we have a different idea of what constant change means. But I'd be surprised if you could point to several journals with more than one change in citation format in the last 3 years, say.

    We also have a good sense of often styles are outdated via error reports (understanding that not everyone reports issues), and while that certainly happens, it is a comparatively rare occurrence.

    Also, honestly, scientists are really bad at following citation guides (ask any copy editor...), so the curated existing automated styles almost certainly produces better results then everyone fending for themselves. That's why you'll find most journals recommending reference management software. \

    As for ACS -- I can't really help them. They made a big announcement that they wouldn't be requiring authors to follow a specific citation style for any of their journals anymore (to which, yay!) and specifically requested that we remove any individual styles with journal-specific features. So the problem there is really that one hand at the publisher doesn't know what the other one does/wants. Not something Zotero can solve...
  • edited November 7, 2018
    Thanks for the answer. I had no clue it was as complex to create an easy to use GUI as you kindly described.
  • @adamsmith Would it potentially be a good idea to add DOIs to ACS styles generally though if they aren’t already?
  • edited November 7, 2018
    I'd have to go back through the correspondence. I thought they told us not to. @Rintze do you happen to remember?
    Otherwise always happy to add DOIs where I can.

    Edit: Rintze poked around in his correspondence with the ACS team and that suggests that DOIs are, at a minimum, acceptable. We'll add them.
  • @adamsmith Adding DOIs to ACS style would be a welcome change. I hope that will include ES&T, right?

    Do you think DOIs should be added for all styles that are silent or ambiguous about DOIs? I've had the feeling that journal editors are willing to accept DOIs when their style says nothing about it (a style categorically denying DOIs is another thing). In any case, deleting DOIs from the final proof is easier than adding them.

    Or, maybe create a radio button kind of feature that will change a style to include DOIs? But I am sure this is complex enough that you cannot do it in the limited time you have for this task. As always, heartfelt appreciation for the work done by Zotero patrons/users for CSL >> https://citationstyles.org/about/
  • dding DOIs to ACS style would be a welcome change. I hope that will include ES&T, right?
    yes.
    Do you think DOIs should be added for all styles that are silent or ambiguous about DOIs? I've had the feeling that journal editors are willing to accept DOIs when their style says nothing about it (a style categorically denying DOIs is another thing). In any case, deleting DOIs from the final proof is easier than adding them.
    While I'm philosophically in favor, we get so many complaints already about DOIs that are included in styles that I'd rather not unless we can point to explicit guidance or instructions.
  • edited November 8, 2018
    Just to add to the confusion "academic" versions of the standard styles sometimes demand DOIs always when available and I'm aware of at least one academic case where no DOIs are allowed--ever.

    I really cannot see a non-confusing UI change that could accommodate a DOI radio button but a control of that kind would be really nice to have.
  • Fair points, both. I find it frustrating when researchers/editors/anyone resist DOIs - they are the easiest way to get to the resource! While volume, serial, and page numbers should still be included, all of these are from a time when scholars accessed publications in print form from a library stack. Anyway ...

    @adamsmith is there an easy way to find styles that include DOI and/or URL in the citation? https://www.zotero.org/styles/
    Sometimes I really like how a style is formatted but can't use it because the DOI is not included. Would be nice if I can find a similar style with DOI, or browse all styles that allow DOI.
  • Not really, no. You can try http://editor.citationstyles.org/-- search your preferred style by name, then copy over the sample citation, add the DOI, and search by example.

    But if this comes up often enough to be relevant -- I understand we can't expect every end user to edit styles, but for power users such as yourself, adding
    <text variable="DOI" prefix=". https://doi.org/"/>
    as the line right before
    </layout>
    </bibliography>

    and then changing the style id in the built in style edit pane of Zotero really shouldn't be insurmountably difficult or take longer than 3 minutes
  • That's great and answers my need. Thanks.
Sign In or Register to comment.