While working on one of my languishing manuscripts, I noticed something that started to bug me.
You see, when the muse isn’t on duty I fall into the trap of read and invariably revising what I’ve written in order to refresh my memory of the vision(that my excuse and I’m sticking to it). So I tend to keep the underline misspellings feature on.
Since this particular manuscript is a SciFi one, many of the proper nouns are not found in the dictionary. In the past, I’ve tried adding these proper nouns into the dictionary, but it suddenly struck me, why couldn’t Scrivener access the Characters and Locations lists (Perhaps other custom lists too like articfacts, spells, races, planets, or anything else setup by the author) like a user added dictionary? That way I wouldn’t have to worry about the name of the alien premiere being checked against the words I’m using for something a little closer to Earth.
Personally, I fear that you’re letting minor details distract you from drafting.
I would turn off spell-checking and fix the errors later on.
Maybe after you export to a word-processor.
I know what you mean about keeping track of the vision; it can be tough.
I use the Synopsis cards and Document/Project Notes as reminders.
I don’t think a feature like this (assuming for a moment that it were hypothetically possible to do this) would be broadly desirable until some way of addressing the global spelling dictionary issue were present. The way the Mac works currently is: every single application, and every single document within every application, uses the same spelling dictionary overrides. You would consequently end up with your SciFi names in Mail, while typing in URLs, and anywhere else you see the Mac highlight misspelled words. Add to that how difficult it presently is to view or edit this global list. Again, that assumes it can even be done—I don’t believe there is a way for a third-party program to do this (without resorting to editing system files directly). All of the spelling engine paraphernalia is being generated externally, by the OS framework. Scrivener itself does very little beyond saying, “Use it”, or “Don’t use it”.
There is also the problem that Scrivener is wholly unaware of what a character is, in this sense. It can be a bit more active about it when scriptwriting (and in fact character names are handled automatically when writing scripts; they will be added to the project auto-complete list for you as you write), but otherwise it has no good way of knowing that the word you typed is significant, or that this folder you just so happened to call “Characters” is significant in any way. We’d have to start building dedicated database features into the program, and that is a direction that Keith has never been keen on taking.