RE: new-flex-completion-style

[Drew Adams]

> BTW, the issue of scoring quality is not nearly as obvious as one might
> think because the score one should give to a particular match depends
> on the user's expectation and the user's expectation often depend on
> the past behavior of the tool, i.e. on the scoring.

+1 - good point. There's room for the tool to learn too.

And it can depend on the context - intended use of
the chosen completion(s).  The completing function
(completion table etc.) typically does not have a
good view of the user's context, including intent.

Put differently, if scoring only has the matches
and the input pattern to go on, the most it can do
is offer some measure of comparison based on just
those two inputs.  A score that's really useful to
a user might depend on more.