> From: Ship Mints <shipmints@gmail.com>
> Date: Wed, 30 Apr 2025 10:33:52 -0400
> Cc: eg642616@gmail.com, 77715@debbugs.gnu.org, drew.adams@oracle.com
>
> On Wed, Apr 30, 2025 at 10:30 AM Eli Zaretskii <eliz@gnu.org> wrote:
>
>  > From: Ship Mints <shipmints@gmail.com>
>  > Date: Wed, 30 Apr 2025 10:09:36 -0400
>  > Cc: eg642616@gmail.com, 77715@debbugs.gnu.org, drew.adams@oracle.com
>  >
>  > On Wed, Apr 30, 2025 at 10:06 AM Eli Zaretskii <eliz@gnu.org> wrote:
>  >
>  >  >  But this code is not general enough for that.  It was written
>  >  >  explicitly for additional optional values for ring-bell-functions, as
>  >  >  the doc strings say.
>  >  >
>  >  > The prototype was derived from private code used in this narrow case but if we're going to adopt
>  the
>  >  > functionality in core, I think we go the extra mile to make it appropriately general.
>  >
>  >  But then the entire implementation should be revisited and reviewed
>  >  with that generality in mind.  So please let's talk about that.  Could
>  >  you or someone else please describe what general features are meant to
>  >  be implemented based on this functionality?
>  >
>  > I think of face flashing as the face equivalent to 'pulse-momentary-highlight-region'.
>
>  Then why don't we use the functions defined in pulse.el in the first
>  place? why invent a whole new family of functions,l face attributes,
>  etc.?
>
> Overlays require a buffer and flashing faces like the inner border or tab-bar can't be accomplished using
> pulse.el functions unless I'm missing something.

But the code as submitted included echo-area as well, where there _is_
a buffer.

And my question about the faces is still unanswered.