Total word count is too aggressive.


I’m probably doing this wrong.

I like to keep sub-revisions of my scenes near my active revision. It makes them easy to find and easy to compare new with old. When I do a big rev, such as adding chapters and stuff, I’ll save the whole MS to another folder. But for sub-revs I save them under the current rev. A screen shot:

The big rev is 0.8 while the letters are sub-revs. Here I’ve reved chapter 1 scene 1 once, chapter 2 scene 1 twice, and I have not reved chapter 6.

Great. They’re right there and they’re easy to find. However my total word counts are off. I unchecked the “Include in Compile” box for all obsolete scenes. When I do an export I only get current scenes, so that’s working as expected.

Why does total word count count words in scenes that are not included?

Scrivenings mode includes the unchecked scenes as well.

Help! Am I doing this wrong? Is this a bug? Is it a “feature”? How can I do sub-revs and still get a good word count total and a working Scrivenings mode? I’m using version 2.4.1 on Mountain Lion.



PS: To see total width of screen shot, which includes target word counts (which are accurate), right click pic and select “View Image” (or whatever fits your web browser).


“Total words” just shows you the total word count for everything that is contained inside the folder. “Include in Compile” only affects Compile, not word counts like this. There’s no way to show a total that ignores documents with “Include in Compile” deselected in the outliner.

As for Scrivenings, either use Documents > Open > With Compilable Subdocuments or Opt-click on the folder when selecting it in scrivenings mode.

All the best,

Awesome! Compose mode works too. Thank you so much.

As for word count, I feel that if a file is not included in a compile, then it doesn’t belong in the MS, and therefore its word count is irrelevant. Yes we can temporarily uncheck those boxes, but is it a problem if word count temporarily changes to match?

Anyway, now I’m curious. How do other users feel about this?



I use snapshots for what you’re doing in the editor. Snapshots aren’t as visible in the binder/outline, but they’re otherwise just as useful for referencing changes (even better, actually; you can have the differences highlighted), and they can’t affect the current word counts. In other words, I don’t feel one way or another about how the include in compile checkmark has no impact on the total words tally.

Personally, I think the current set up works. In my thesis, I often had segments that I was deleting, temporarily excluding, thinking of moving, or for some other reason not including in my main text at any given time. BUT unless and until I moved them into my research folder (or, more likely, my “offcuts” folder), they were still a part of the main text and I expected them to display in the word count. I could always select relevant documents in Scrivenings mode to see an up to date word count without the temporary deletions.

I think that robertdguthrie is right about your needs: Snapshots appear to do what you are looking for.
If you are looking for alternatives however, you could move old versions to an offcuts folder (like I used to do - I rely on Snapshots much more now) and then insert links to those earlier versions in the Document References pane (I didn’t figure this out until after I’d finished my thesis), or even as in-text links. This makes it trivially easy to access your revisions without cluttering your main text with multiple versions.

Thanks guys.

I create a minor rev for one of two reasons:

  • I get feedback on a chapter from my critique group–old rev is pre-critique, new rev is updated.
  • I decide a chapter isn’t working and I start over. Should I later decide the old approach has some merit, I still have it in all its glorious half-completed first-draft badness.

A major rev is for restructuring the story. For example if I need to shuffle events between days, add and delete chapters, and just generally muck about with the structure. To do this I copy my whole manuscript into a top-level folder “Previous Drafts”, delete all the obsolete sub-revs from the original, and rename the old scenes to reflect the new rev. E.g. “02.1 v0.8c” becomes “02.1 v0.9a”.

My technique is in evolution.

  1. In the bad old days I didn’t do revs. That was a mistake.
  2. Next I kept folders named “version_0.1” (for example) and put all my doc files in that. Problem was, I needed to copy everything into each rev’s folder or I’d lose track of where stuff was stored. For example my synopses, character sheets, and whatnot.
  3. I started using FreeMind to organize my documents including research, web links, etc. (FreeMind is very much like the binder). After several years of experimentation my FreeMind technique was to put old minor revs behind the newest minor rev, then do major revs as a new tree. FreeMind decouples the directory layout where the files are stored from the logical layout (like the binder). So my top level in FreeMind is the latest rev. I do bump all manuscript docs to the current major rev, but I don’t bother for character sheets, etc. The beauty and the pain of FreeMind is the total freedom to use any application–word processor, spreadsheet, web browser, picture editor, sound player, C-MAP, CAD program, anything. Of course that means zero integration between tools.
  4. I started using Scrivener about 11 months ago (about a year after after I buying my first Mac). Initially I tried copying my whole .scriv file for revisions–gawd but that’s ugly. A few months ago I experimented with putting minor revs behind the current rev, which is super duper if you’re not in the “Manuscript” folder.

I’m still learning Scrivener. Last week I was reading the manual and noticed “collections” and wondered if they might be a better way to do minor revs. Then Robert mentioned “snapshots” and I’ll look into those. When I find something that works better for me, I switch.

Integrating something like a VCS into Scrivener would be too cool for words, but it can’t the “joy” to use that is a traditional VCS. And unless working on a multi-author project, you don’t get the full benefit of a VCS. Perhaps “VCS-lite”, but I have no idea how that might look as a Scrivener component.

Well, I’ll look into snapshots. Meanwhile, Keith, how do you feel about the wishlist item of a 4th check box in the “general” section of the Inspector: “exclude from word count”?



If you can let go of having your revisions as separate documents in the binder, then snapshots are a really good fit:

  • You can name them v1.1 or v0.3b or however you like.
  • Each snapshot is time & date stamped.
  • You can revisit any revision, even restoring the file to a previous version (you will be prompted to create a new snapshot before you do the rollback).
  • Splitting the document will copy the snapshots, merging two documents will bring snapshots into the merged document from both sources.
  • You can ‘Compare’ two snapshots, or a snapshot and the current, editable version by word, “clause”, or paragraph to see what the actual differences are.
  • You can take snapshots of multiple documents at once by selecting a bunch of document in the binder and using the Documents->Snapshots menu.
  • You can even drag a snapshot onto the title bar of an editor split and view it side-by-side with the current version, or with another document entirely.

The downsides are that metadata (including title, keywords, document notes, etc.) are not recorded in your snapshots, and snapshots feel less concrete, being shunted off into the inspector.

Wow, I’ll say!

I checked out snapshots today. They are exactly what I’d been doing manually back in FreeMind, and the only reason they are not as powerful as my copy/hide thing I was doing in Scrivener is because all the extra goodies Scrivener brings to the party are not snapshotted as well. The goodies are neat. The goodies are cool. But I can probably live with only the current minor rev of the goodies (I lived with no goodies at all back in FreeMind days).

Read The Fine Manual.

Well, it’s a 528-page PDF file, so I’ve been skimming it for what I need to know at that moment. Snapshots are mentioned several times before they are covered in depth in section 15.5. I even peeked inside my .scriv file/folder to see how they are implemented.

Snapshots are awesome cool goodness.

Our local writing group is super active, meeting 3 times per week–I run the Thursday night meetings (tonight was open critique). I think I was the first to adopt Scrivener, but lately several Windows users have taken it up as well. One showed up tonight and it was a great pleasure to show him snapshots.

Thank you Keith for giving me what I didn’t know I wanted, but exactly what I needed. And thank you Robert for pointing it out to me.