Tagging of Text

I would absolutely love to be able to arbitrarily tag sections of text.

Functioning much the way you add a comment to a section of text (but with the option of hiding all indication that there is a tag for a section), I’d want to be able to add a custom tag to some arbitrary length of text. You’d then be able to search and view all text snippets by tag.

I’d want this to tag all dialog by each character, and then be able to pull up all their dialog. This would be useful for developing the voice of each character.

Other uses for this are myriad. You could tag sections by setting, POV, tense, etc. You could tag text by grammatical construct (maybe with a script), finding all instances of adverb. You could even tag foreshadowing or setup for later events, finding them both easily if you want to revise.

I’m sure others might be able to think of even more uses.

I second this request. Even though some of these things can be manually done by adding keywords and painting colours, it’s cumbersome to some extent, and it sometimes requires splitting the chunks of text too much.

I think it’d be glorious if freestyle tagging was possible, and even support overlapping of tags on the same text.

This comes up from time to time, but sorry, it’s outside of the scope of Scrivener for now. I don’t rule it out for the future, but there are no plans for this in 2.x or 3.0.

Thanks for the suggestion, though.

All the best,

People have time to tag their adverbs? :open_mouth:

Much of this you can already do with document notes, inline annotations and meta-data, especially if break your documents into smaller sub-documents. The more fine grained your chunking, the closer to your tagging system you get.

The default should really be that each word typed in scrivener gets automatically tagged with itself as keyword; not only that would be a safe redundancy to help find that word with the search function, but would also provide an easy way to duplicate whole documents without using copy/paste. This list of keywords would be magical to generate indexes and glossaries, and I am sure others will find multiple new uses for it.
Maybe an optional second tag could also be attached to each word, …for a fee.


… time for my soup! :confused:

Genius! :smiley:

Self-tagging. Strange that no one has thought of it before. Now the idea is out; I expect some entrepreneur in California is putting his robotchiks to work on it.

Tags for adverbs. I want to like the idea, because pruning adverbs (and “the”) is a first-pass edit. But if I tag them while writing, why not prune them on the spot?

As The King says, “Is a puzzlement.”

For anyone working with Multimarkdown: Marked now offers “Word Repetition Visualization.” Haven’t tried it, but it too could become a first-pass edit.


I’d like to bump this, but with a slightly simpler suggestion.

If we could just automatically tag or highlight anything between quotes, that would be a huge help.

In other words, with a toggle, change this:

“That would be very cool” Bob said.


“That would be very cool” Bob said.

That’s pretty straight forward I’d think. You’d have to clear the " marking at the end of a paragraph so a mistaken or misplaced quote doesn’t stain the entire doc :slight_smile:

I too would love to see ‘is’ or other passive words be tagged too, although that’s a little more advanced and sounds like would be out of scope, although I’d sure like that to do a grammer pass. “Highlight all passive voice …” lol.

== John ==

This seems like on of those no-brainier features that would make Scrivener immensely more powerful. Anyone that is working on a document of significant length like a novel almost certainly knows how hard it is to keep track of multiple intersecting plot lines and character arcs. Arbitrary tagging of certain chunks of text combined with an easy way to recall them and (ideally) visualize where they occur would almost completely solve the problem. The only solution Scrivener offers is to decimate the text down into paragraph long or even sentence long scrivenings so you can actually use the tagging and metadata features.

I brought this up a while ago, actually, and a helpful moderator at the time offered numerous, lengthy responses. The only hope I got of ever having something like this was the thought that maybe we’d be able to use comments, and Scrivener would eventually be able to display all of the comments from every scivening you had selected, instead of only the active one.

2 posts were merged into an existing topic: Inline Tags

Maybe tagging comments or annotations would be easier for them to program. IDK, but yes, this would be a great feature. The type of writing I do almost demands it.

On those occasions when tagging might be useful I turn to corpus linguistics software especially those programs that include Part Of Speech (POS) tagging. Currently I am using LancsBox produced by the specialist corpus group at Lancaster University not least because it accepts a myriad of input formats — thanks to a Java library — and applies POS tagging — thanks to a different Java library. One of the formats that LancsBox accepts is our good friend RTF.

However, much as I sing the praises of corpus linguistic software there are times when even in that environment tagging becomes more intrusive than useful. That said I am at the moment compiling my own corpus which is currently planned to have eight tagging schemes applied to it! POS using the same toolkit as LancsBox, a simplified POS which reduces the specificity of the POS tags to broarder classes so transitive and intratransive verbs are tagged as VERB with specific instances ot comma period semicolon etc reducded to a generic PUNCTUATION, semantic tags which apportion the lexemes in the collected texts to semantic groups— this using another tool from Lancs which they have recently made opensource, and lemmatisation which collects declensions into the headword. These are all pretty standard stuff for corpus software. But I plan to go further by including Named Entity tagging with a subdivision of personal names, then speech (utterance) markup, and finally syntactical tagging. Like the semantic tagger all the tools I am using are opensource as is the web crawler (as all the texts I want are out on the web somewhere), boilerplate redactor and deduplicator.

There is no way I would want any of that stuff being added to Scrivener itself. It is all too niche. Ah that reminds me of something else I would love to tag my corpus with “foreign/scientific phrases” especially Linnian taxonomy.