RE: Incorrect terminology in Customize doc

[Drew Adams]

    My understanding is that the original definition of "user
    option" was: an
    option that an end-user (not familiar with Elisp) can change.

I have no pb with that.

    This happened to be limited to user-variable-p at the time for lack of
    anything else but now also includes faces.

OK, but I don't think that is stated anywhere, and I think that "user
option" is used in various places explicitly with the more restricted
meaning.

    I don't think it's particularly confusing.

The only confusion comes from doc or doc strings that might refer to the
more restricted meaning (`user-variable-p') of "user option". If those are
eliminated/corrected, then there is no pb. I don't know if this has already
been corrected.

    > That will also require, BTW, a term for `user-variable-p'

    Why?  It's not like we often need to talk about those
    particular variables
    which are user-variable-p but are not defcustom'd, is it?

No, I agree. We just need to say clearly that not all `user-variable-p'
variables are "handled by Customize", or something to that effect, in places
where the distinction might be important. Otherwise, people will be
surprised when they try to `customize-variable' and they are told that they
cannot, even though the variable is `user-variable-p'.

    And IIRC the last discussion (where we decided that M-x
    set-variable should
    be extended with the functionality of M-x
    customize-set-variable), we'd like
    ultimately to get rid of the distinction between user-variable-p and
    custom-izable variables.

I missed that discussion & decision, but I support the idea. Is that already
planned for post 21.4?

 - Drew