Does your workflow really have you using Bookends simultaneously on more than one machine? Using is different than installing.
Big risk for you to change workflow at this point. I get that. Been there done that and … never again.
Does your workflow really have you using Bookends simultaneously on more than one machine? Using is different than installing.
Big risk for you to change workflow at this point. I get that. Been there done that and … never again.
Yes, whilst I finalise the doctorate I have two machines running due to issues with space and monitor availability.
Agreed, which is why I won’t do anything too drastic.
I wrote my earlier thesis in Scrivener, but I managed my citations manually then, and of course some ended up missing from my bibliography. The examiners didn’t pick up on it and neither did I until well after results were returned (citations were there, just not reference in bib), so I was lucky but then shifted to Zotero.
For the doctorate, I’ve been using Microsoft. I really like the integration, fact that the citations are where they should be, and WYSWYG. I just really miss the binder from Scrivener.
I’m in a world of stress right now, and breaking things out into smaller parts makes it more easier to manage. However, in Word I end up with oodles of docs open for that and my workspace becomes overwhelming.
Right now, I’m writing in Scrivener and adding the citation using a footnote as a reference to myself. Then I copy/paste the whole section into Word, and go back and add the citations with Zotero. Not at all ideal, but the only way I can clear my work space and have a binder.
Sigh. I really wish Scrivener would facilitate integration, as from what I read on the Zotero forum, that’s the issue:
"There is absolutely nothing in place that would prevent absolutely any app developer to build a world-class integration with Zotero. You can, in principle, get the exact Zotero behavior in a markdown editor, without any changes or plugin to Zotero, since the protocol that the Word-addin and others speak to Zotero is available to any and all integration, but you’d have to do the work.
WRT Scrivener, Literature & Latte show no interest in doing the work, and they do not offer any extension points, so the people who can do it won’t, and the people who would do it can’t."
Zotero works with Scrivener (I have a collaborator who does so, see the wiki entry here) with the main limitations to do with Zotero. Zotero 7 recently broke one of the two compilation routes. If you are willing to use Pandoc + BBT the Zotero workflow is basically as good as Bookends. What problem do you have with Zotero?
I have recently had to deal with a total %^$# of Zotero + Word for a paper from another collaborator. Zotero’s Word plugin injects references into fields, and the metadata was corrupted in multiple refs causing repeated failures when trying to add or reformat (could not work with the document, Zoteros web page suggesting how to fix errors failed to provide useful suggestions, Zotero plugin could not identify which refs had problems, could be any of >80). Interestingly the problems were less serious with LibreOffice so we exported the doc to that, then had to hand edit the refs. Honestly, the way Zotero / Endnote stuff references into hidden fields is a disaster waiting to happen. I have had to fix CWYW problems from others many times over the years. Scrivener uses temporary citations, and thus is much less prone to any sort of problem, temp citations are robust and reliable.
WRT Scrivener, absolutely the same integration that is available for Bookends can be used to access comparable mechanisms in any other tool. If Zotero chooses not to supply those mechanisms, that’s their choice.
That is their perspective. My perspective is Zotero does not provide a basic and functional temporary citation workflow, as Bookends + Scrivener do. Zotero is free and Zotero developers don’t owe anyone anything of course.
Anyway – why can’t you use the Zotero + BBT workflow (BBT is a Zotero plugin that enables a temp citation workflow) for Scrivener?
I’m not sure if you’re a Scrivener dev, so I’m not sure if you mean from a dev perspective or a user perspective. However, from my user experience is that Zotero functions beautifully in Word. If what the Zotero team say is true, and they aren’t preventing the integration, it feels like a choice by LL.
I don’t understand the backend of it all, I’ve just seen the discussion on the Zotero forums indicating that they would build it.
If the app developer wants someone else to do the work, all this app developer has to do is give us extension points into the app, documentation on how to use them, and explain how it would benefit enough of the Zotero users that it would fit the Zotero mission to spend their resources on it. That’s all. This explains the MS Word, LibreOffice, and Google Docs integrations – Google, MS and the LO devs made extension points & docs available, and Zotero built the world-class integration on top of this. If you have trouble convincing on the userbase part – plenty of people on these forums will be happy to get you started on a Zotero plugin that gets you the desired behavior. Zotero is crazy extendible.
I haven’t found a clear guide on Bookends and have stopped looking as the cost of Bookends is too prohibitive for me. However, if the integration with Bookends is a temporary citation key, then I don’t want that.
In Word, we can insert a citation as it will appear if printing. Whilst I understand Scrivener wouldn’t be exactly the same, I’d still want some form of WYSWYG (e.g. in the footnotes etc.) that works smoothly with prefix and suffix etc.
The temporary citation is a feature of Bookends, it has nothing to do with its Scrivener integration.
And it is a really good feature: First, the temporary citation simply allows both the user and Bookends (when scanning the finished manuscript) to identify each reference. One temporary citation format—the one I prefer—is author, date, unique ID
. The first two parts are for the user, the third is for Bookends (or users with a really good memory for numbers).
Then it allows the user to decide after the writing process which citation format they, or more likely: their editor/publisher/whatever want to use, be it Harvard or Oxford or else.
In that aspect Bookends is like Scrivener: You create content and only later you have to decide the form you put it out in. Say, you write a paper for a magazine with the required citation format A. And later your paper becomes part of a book with the required citation format B. Not an unlikely scenario, I’d say.
If you had used the full citation in Format A in the manuscript already it would become difficult to replace it by Format B. With a temporary citation you just set Bookends to use Format B instead of Format A and that’s it.
As this is a Bookends feature Bookends works the same with Word, by the way. The only difference is that the scanning process—i. e. replacing the temporary citations with the formatted ones in the preferred format—with Word takes place in Word itself but with Scrivener not in Scrivener but in another program (e. g. Word).
Again: Temporary citations are very useful, especially in the long run, because over the course of time citation requirements may change. But if you prefer the formatted ones in the manuscript Bookends allows to do that too: Just use Edit/Copy Formatted/Bibliography
instead of Edit/Copy Citation
(wether you use it with Scrivener, or Word, or any other writing app) or their respective keyboard shortcuts.
I’m just used to Zotero achieving all of this efficiently, displaying the citation as it will appear. Switching between styles is no issues, just change it and the document updates.
I’m a user, but at least a user with fairly deep knowledge across many reference managers (thus wrote the wiki here). I’ve use Endnote, JabRef, Bookends and Zotero recently across differerent collaborations. I’ve worked with CWYW for both Endnote and Zotero in both Word and LibreOffice.
My extensive experience of CWYW across many projects is that on many occasions over the year I’ve had to fix problems when CWYW failed to work properly (mostly with Endnote both before and after the Travelling library change, but I had a major issue with Zotero just a few weeks ago). As I said, CWYW injects content into hidden fields (Mendeley, Endnote and Zotero all use the same blob-of-stuff encoded as text injection mechanism), technically this is prone to all sorts of potential issues. More than that, on at least two occasions I watched students struggling and panicking with awful performance for their close-to-completion theses (Word is already really bad as documents grow with many cross-refs and figures etc.) which were heavily exacerbated due to their use of CWYW. Turning it OFF and things got better. CWYW is fine while it works (though it is a perennial pain when you collaborate[1]), but it more-than-rarely fails some way or another.
Prefix and suffix works fine with temporary citations.
The Scrivener developers if I remember correctly have stated they will not make a CWYW interface. There are valid technical reasons why CWYW is prone to bugs as I mentioned above. There are many academics who use Scrivener with the temporary citations without issue.
If you honestly think Word + CWYW is better writing environment than Scrivener + temp-citation then you of course always have the right to use what is best for you.
Zotero + BBT is not Bookends. You are missing the advice I am giving. BetterBibTeX is a native plugin for Zotero and a workflow many peope who use Zotero use, and works for Scrivener users too:
For LaTeX and plain text users it ensures each Zotero reference has a unique citation key. It is a great tool for Zotero, and works well with the latest Zotero 7.
Read this for the details:
You use that citation key when writing, and these are then transformed to a full bibliography. In Scrivener, this means using Pandoc to process Scrivener outputs to build the bibliography. It uses the same CSL styles that Zotero’s CWYW plugin use.
You can see a sample project using BibTeX output here:
This workflow works well (my personal view is Pandoc’s DOCS and ODT output is actually much better than Scrivener’s, you can use DOCX templates to inject any styling). But this does have a learning curve (though we are active to help people on the markdown & LaTeX forum here).
[1] Zotero injects the full reference, so if I have the same ref with small changes in my database there is not a clear way to update the collaborator ref with my one, it is just badly designed for collaboration.
Apologies, I misread it. I need to teach myself how to use it, and I don’t have the time right now. I have used it a little with Obsidian (integrating with Zotero) and it was all such a time suck. I have a tendency to get lost in these things, it’s time I don’t have right now. I read somewhere that the options available for Zotero 7 are not dynamic, I’m also not clear on whether I need to import a copy of my bibliography (which is not ideal if you read as you write) etc.
I just have too many questions to feel comfortable going ahead with it right now, there are too many terms/things swirling around that I need to clarify (what is pandoc, what’s the difference between LaTeX and BibTex etc.). It’s just not the week for it.
I am constantly referring to citations, I’m in the revision / re-write stage and currently working on theory chapters. I need to be able to interpret the temporary citations.
I came across a recommendation to use a workflow with Alfred a few days ago, and downloaded Alfred to test it out. However, it wanted me to purchase Pro to access the functionality I needed.
As it stands, Word plus Zotero feels a lot safer for me because I understand it. There are things I really like about Scrivener that Word doesn’t have (binder, synopsis, notes etc.), but I find the Word interface and formatting functionality far more intuitive and not so limiting. For me, Scrivener would be perfect if I found it easier to cite and format.
Having said all of that, I just want to say thank you so much for the time you’ve taken to respond to my questions, and provide links to further information. I do appreciate it. When I’m not feeling so overwhelmed by this thesis, I’ll take some time to see what I can do to make it work.
You can refer to temporary citations. for example @shipp2013
is the cite key you can use in your text, and using applescript you can click it and do a reverse search back to your reference manager
Temporary citations usually use first author+year, so they are close to a in-text as a mnemonic, for example @shipp2013
would be (Shipp et al., 2013). And as the cite key is unique, you can select it and use it to search in your reference manager. Here is me using Scrivener + Zotero + Zotpick to insert a ref and then to search that ref:
Honestly, this workflow is as fast and efficient as CWYW, you can easily get to your ref. The main issue you’ll have is compiling as you have no experience of a Markdown compile, I understand it can be daunting. Take your time and ask questions when you are ready.
Good luck with your thesis whatever tools you end up using!
To clarify a bit, the choice by L&L is to be agnostic. Rather than dedicate development resources to integrating Zotero (or Bookends, or EndNote, or 2025NewShinyLibTool), we provide a consistent interface that allows writers to use whichever tool they want. Or even to change tools when they decide NewShiny works better than whatever they were using.
This approach is motivated in part by our experience with third-party integrations generally. Companies become unreliable. They abandon products or platforms. They go out of business. And we have to answer a long stream of support queries along the lines of “please fix DeadTool integration.”
In the case of Zotero specifically, @nontroppo has explained why temporary citations are useful, not just for Scrivener users but for anyone who prefers human-readable markers in their text. Zotero’s developers have chosen to follow a different path, which is fine. We just don’t think it’s a good use of our resources to go down that road with them.
There is a fundamental misunderstanding here.
Zotero does have a Word integration (by plugin).
Bookends does have a Word integration (by plugin).
And they both (and Endnote too) do have a Word integration for obvious reasons: Microsoft not only managed to establish .docx as the de facto standard in certain fields, Microsoft has also cleverly ensured that Word is the first choice for many people for writing, even if it is not the office writing that Word was originally designed for.
On the other hand:
Zotero does not have a Scrivener integration.
Bookends does not have a Scrivener integration.
And just in case you have missed the three-letter word in the last sentence I repeat: Bookends does not have a Scrivener integration.
Bookends simply works with any writing application accessible via standard MacOS features (I’m not a programmer, but there’s no need to get technical here). And Scrivener happens to be one of them—as is Apple Mail if you ever feel the urge to write a paper in an email.
As a one-step preparation you set up a writing app as the word processor in Bookends’ settings. And when you press the keyboard shortcut for either the temporary or the formatted citation, Bookends
And that’s all there is to it.[1] It’s just the most basic stuff. It’s above all the most basic macOS stuff, which the Zotero programmers obviously don’t deem worthy of providing. Or to put it another way: Cross-platform apps very often are not fully integrated into macOS. While Bookends and Scrivener are native Mac apps.
So—a rhetorical question—what would make more sense: Have the (small) vendor of one single writing app (Lit&Lat/Scrivener)—that currently does not have any plugin port at all—integrate a plugin for one single reference manager (Zotero) or have the vendor of one reference manager (Zotero) enable access to basic macOS features and thus to every (native) writing app running on macOS?
Which leads to:
You would not. You could use it with lots of other apps on the Mac and on iOS/iPadOS devices, e. g. information managers or note taking apps.
However, Scrivener does add something to the mix: You can link a—a, not just Bookends—citation manager in its settings. Then, if you press the keyboard shortcut, the citation manager will start (if not running already) and become the frontmost app.
At best, it’s the same keyboard shortcut that Bookends uses to transfer the citation. So, for temporary citations, the default workflow would be: Press cmd-y
in Scrivener, select the reference in Bookends, press cmd-y
again and the reference is injected into your Scrivener document. If you can use two hands, one can rest on the keys for switching between the apps while the other uses the trackpad or the mouse to select the reference. It is a very fast way to cite once you have learned the muscle memory. ↩︎
As the person in control of my workflow, with the best understanding of my current needs… I think it is fair to say that this is the sole reason I would be getting it (not that I will). Just because I could use it for other purposes, does not mean I currently lack that functionality and therefore need Bookends to fill that gap.
Of course you are. I was imprecise at the end of a longer posting.
I just wanted to point out how many advantages a refererence manager—not necessarily Bookends—offers when it is integrated into the system.
By the way, I happen to know a lot about not having a lot. And I am not against Zotero either, I think the idea of a cross-platform, free programme is very good.
Oh, I wouldn’t be able to do this project without one! I’m currently looking at a final product with more than 600 references.
I have free access to Endnotes through my university, but at the beginning of my project I found Zotero more intuitive, so I went with that. I had also just witnessed someone about to submit have a major issue with Endnotes that put me off (but to be fair, the online community helped them fix it).
I’m just now accustomed to using Zotero in a specific way, that looks clean to me as an end user.
I love Scrivener for the binder, and the progress (for the most part, some of it I find a bit clunky) etc., but it doesn’t maintain a polished product and I inevitably have to do more work once exported to Word. Being able to cut down on that would have been great, but I guess at the end of the day this is more of a tool for fiction writers who won’t have the same needs for their finished product.
I don’t do any work in Word (well in my case LibreOffice as it is better than Word overall IMO). My bibliographic references, cross-references, figures with captions, tables, maths equations are all automatic. Here is an example from a preprint, this is the direct output from Scrivener:
Click on the intext citations, they cross-link to the bibliography and visa-versa. Note my DOI links are shortened etc. The author list, the funding information, which authors are corresponding, which have ORCID IDs, this is all directly generated from the output of Scrivener. No fussing, no post-production. The workflow I and other academics use (Scrivener is not just used by fiction writers, the compiler is way too powerful for most of their needs, but fits more technical work perfectly), is available to you. You could tell me you can reproduce that layout in Word (with quite a bit of fussing), but then I’ll tell you I can take the same Scrivener project and output it to a totally different journal format with no changes to the project itself immediately. Scrivener has almost completely separated the content from the presentation (via compilation, the opposite of Word which makes separation of concerns so much more difficult, even if you use Styles). I have access to the latest developments in literate publishing, I can run e.g. Python or R code to make figures, analyse data etc on each compile from Scrivener. Could you easily turn your paper into both a website AND a book directly from Word?
You want to use CWYW, and it is not supported by Scrivener, you of course have the right to argue for your preferences, but it does not mean anything else about Scrivener’s suitability for academic or technical work.
Re https://www.biorxiv.org/content/10.1101/2023.10.19.563116v3.full.pdf, wow. Finally get to read something from you not having to do with Scrivener! Refreshing. Will I understand? … probably not, but will read with interest anyway. Thanks!
Ha thanks, maybe good bedtime reading for insomniacs unless they also happen to be vision scientists. This preprint ended up being published here: Assessing perceptual chromatic equiluminance using a reflexive pupillary response | Scientific Reports — and IMO the final “official” PDF is much more poorly laid out than the automatic output from Scrivener in the preprint