John you highlight some points
>
> I'm not sure what you mean by "assume". Is this not exactly what you
> see happening?
>
Loose expression - I hadn't checked it as it had nothing to do with the point I was making - sorry about the cavalier response!
>
> The place where you "end up" is simply the last file that matches a
> correct spelling. As Kim points out, the cursor does not "move". It
> stays exactly where it is and reports that a file with that spelling does
> not exist. It's not clear what you find wrong with that logic.
In my example, the context was that I was searching for Wilson - but I ended up 19 entries away from my target - instead of adjacent to where it would be slotted if it stuck at Wilton. Continuing your point, I would then leave the cursor at Wilton - which would then remain presumably with a note saying "no match"
>
> But if the author does not exist it makes sense to just stay on the
> current file and tell you that it does not exist. It makes no sense to
> move to a file that does not match the spelling. Of what "tangible" use
> would that be? How would it be any better than staying where you are and
> reporting "No match"?
>
I would be an indeterminate distance away from where I wanted to be (19 in my example).
I did ask "what tangible use is the place where you end up? What would you do with it? What activity would typically follow?"
> Chester, I'm confused. I must be missing something. Do you not already
> have the files sorted in name-ascending sequence with Alt-Sort?
>
None of the example would make sense if it wasn't.
I do not question the logic or consistency of the approach implemented, I couldn't see what practical use it could be used for. My example had a real context.
Bottom line - on a failed search I want the cursor to be adjacent to the place where it would have gone - not 4, 19 or 55 entries away.
So I ask - does anyone else have a need for this approach?
It could easily (?) be implemented by a CO6L - the L is currently unused.
regards, Chester