I am writing a biography with events sorted by date. I need to be able to tag certain sections or even just sentences with custom labels so that I can compile different versions of the same document for different pursposes. For example, controversial or unsubstantiated information can be easily included or removed in the compile process. That way as I verify facts, I can change their labels to “verified” and include those sentences or paragraphs in the official version while other facts remain labeled as “unverified” and don’t get printed until they’re changed. Is this feasible without cluttering up my projects with thousands of separate documents inside the project or is that what I should be doing? Won’t that make further organization and editing more complicated?

I think you can use custom Labels, and then create Collections that only include certain Labels or groups of Labels. You could then choose to compile a specific Collection.

The basic answer is that it is relatively simple to do what you want (using the method Lunk suggests of labels and collections) but only if you do this on the individual Scrivener document level. A document can be as short as you like (down to a single paragraph) so if that’s as granular as you need, then it’s doable. You’ll have hundreds (thousands) of documents, but it will work (albeit it will be fiddly to administer).

But you can’t label anything within a document, so that approach won’t work for sentences within paragraphs.

Off the top of my head I can think of one approach which you may want to test and try — it works, but I’m not sure whether you’ll find it practical or not!

It makes use of the fact you can compile documents to print or to ignore inline annotations. So, if you highlight a sentence or paragraph with an inline annotation (cmd-shift-A)…

then when you compile, you can opt to ignore the annotation or not as you choose in the compilation options:

This is what it looks like on compilation (2 PDFs from the same text with inline annotations printed / ignored respectively, of course):

You’d have to be careful to make sure that the annotation includes sensible spaces around it, so the text runs together, but perhaps it’s worth experimenting with to see if it can work for you?

Hope this gives you something to explore, anyway.

NB: you can of course search for annotations using Find by Formatting, and if you combine it with a tagging convention (e.g. start every relevant annotation with, say [Dubious: This is a big fact lie ]), you can log what’s going on, and also make it easy to spot any marked sentences which get through by mistake by compiling to PDF and doing a search and replace on Dubious:

Thanks for the tips. Neither of these options are very practical so I may have to use another app.

For anyone interested, I decided the best solution would be to outline my project in Aeon Timeline which has all the features I need for tagging events in my project, then link those events selectively into Scrivener for writing. I can tag verified biographical information in Timeline as green and unverified as red. Hopefully I will be able to figure out how to easily create versions of the book with both verified and unverified information by filtering in Timeline for those events and exporting the linked text coming from Scrivener. I’ll try to update on my progress down the line.