I wasn’t sure if I should toss this in the open project statistics thread, so feel free to tell me to move it there.
I am working on a document that I move via USB between two computers. They are both running the same registered version of Scrivener. When I select the entire draft from the sidebar and select project statistics, two odd things happen
There is a difference of about 50 words between the Draft and Selections sections, even though they are referring to the exact same texts. I assume this may have to do with the words in the section headings or something.
When I do the same thing on the other machine, again with the same unchanged ms, I get a variance of about 200 words from the draft on the other machine.
So, when I check on my laptop, the ms draft has 55532 words, and on my desktop machine it has 55755 words. And it’s the same file. With the same settings on each machine, as near as I can tell. Both have the “count all documents” selected in the statistics options.
This is not a really big deal, 200 words either way doesn’t phase me, but I’m wondering exactly what is going on. Ideas? Thanks for your time.
Won’t be long before someone more helpful will be along, but in the meantime I’ll contribute my tuppence
You are right on this one - by default, in the Project Statistics window, the ‘Draft’ word count includes the titles of any documents/folders in the Binder, whereas the ‘Selection’ count does not. So if you had six words in a document that had a two-word title, the ‘Draft’ count would indicate 8, and the ‘Selection’ count would indicate 6 (which incidentally, will agree with the word count in the footer bar of the Editor area).
To change this so that both the ‘Draft’ and ‘Selection’ word counts are the same, you will need to adjust your Compile Draft settings so that titles of folders / documents / stacks are not included in the compilation.
If I had to guess, I would hazzard that either
a) the compile draft settings are slightly different between the two projects,
or,
b) a 200-word text has had the ‘include in draft’ tickbox unticked in one of the projects.
But I reiterate - I’m no expert - and I join Caradee in awaiting the arrival of someone who knows more!
As others have already noted, differences between compile and selection can often be explained by compile settings. It is possible to include a great deal more text in compile than what is strictly in the text itself, and selection only goes through and counts the latter. As a simple test, you can try copying several thousand words into the Document notes for one of your files and then turn on Notes export in Compile. When you view statistics you should see a several thousand word discrepancy between the two. This is working the way it should, in my opinion. Selection counting is a tool for getting a count of the actual text in your draft, but the compile stats are meant to reflect just how big the resulting manuscript will be.
I’m not sure what is going on with the two-computer situation. It is unlikely that it is a compile setting change, because compile settings should be getting saved with the project. If you change the settings on computer A and then copy the project to computer B, those settings should transfer as well as the rest of the project. But you could always verify that just to make sure that is happening the way it should be. What is different between computers are the Scrivener preferences—but I can’t think of anything offhand that would change the way words are counted. I think the first suggestion to check system level language preferences is probably more on target. There could be some setting that is causing hyphenated words to be counted as two words on one system but not the other. The setting that is most likely suspect here is in Languages & Text, under the Text tab. There is a drop-down menu for specifying Word Break style. Make sure that is the same on both systems. I’m not sure if Scrivener uses this information in its algorithms, but this would be the most obvious setting to check where two computers could be behaving differently. Note if you are not using Snow Leopard, that setting won’t exist. As far as I know, that setting only impacts double-clicking, not counting. It determines what a “word” is when you double-click on one, for selection purposes. Maybe, maybe it adjusts some internal counting routine though.
Belated thanks for this. This was exactly the issue. I had one computer using “Standard” hyphenating and the other using “English (United States, Computer)” and once I made sure they were both the same, the word counts normalized between documents. I really appreciate everyone’s help and tenacity with this.
Great! That’s good to know for future support reference. One of the ones you had selected is optimised for programmers, who tend to use several different forms of punctuation to join words together.
