I believe this is a bug, albeit a minor one:
When I start writing a keyword in the Inspector, Scrivener will autocomplete (or suggest) a keyword, if there is another keyword starting with the same sequence of letters. The problem is that it is not case sensitive. So, if there is a “Descartes” keyword already assigned to some document and I want to write a new keyword that starts with “des”, such as “design”, Scrivener will suggest “Descartes” up until writing “des”, but will keep the first letter capitalised even after that point. So, I end up with “Design” which I have to then manually correct to “design”.
It’s not a major thing but it can get annoying when the difference in capitalisation is important (2 identical tags, one capitalised for a title of a book, and one lower-case for a descriptive term, for instance).
This isn’t a bug, just a limitation of how auto-complete works in general. If it weren’t case-insensitive, then you would end up with the wrong case for completions you selected, so I’m scratching my head wondering how I could work around this in code. It’s never come up as an issue before, I assume because most uses capitalise keywords anyway.
Hmm, I don’t understand the “you would end up with the wrong case for completions you selected” part. Maybe I wasn’t clear in my description, but I’m 100% sure that Zotero, for example, doesn’t auto-capitalise or auto-decapitalise my new tag entries based on previous entries. In fact, I’ve never seen it work like in Scrivener before.
But now, after checking, I see where the difference may be. Scrivener actually autocompletes, while Zotero and other programs only suggest in a drop-down list. Sorry, I got it all mixed up…