Re: What is the most useful potential feature which Emacs lacks?

[Eli Zaretskii <eliz@gnu.org>]

> Date: Sat, 6 Jun 2020 11:58:51 +0200
> From: <tomas@tuxteam.de>
> 
> That said, as far as I understand the collaborative editing folks,
> the difference to a dVCS (which I read as "distributed version
> control system" à la git) is the "live" experience: you see other
> people's cursors (points?) running over the text making changes,
> while you change the text, too. Ideally supported by a side channel,
> e.g. audio.
> 
> Think several people doodling simultaneously over a shared blackboard.

Someone will have to explain why this is useful.  Sitting and looking
at other people's typing something, then erasing and retyping, one
character at a time, sounds like a huge waste of time to me.  I could
use that same time to modify a different section of the same document,
or suggest a solution for a problem in parallel to several others
suggesting their solutions for the same problem (which would need some
processing on top of VC conflict resolution).  I'm probably missing
something.

> There was a thread a while ago in -help or -devel explaining why
> several emacs clients connected to a common server didn't quite
> fill that bill: I could only partially understand what the limitations
> were. I think I'll have to try it in practice to get a grip on
> that.

I don't think using emacsclient in its current implementation and the
infrastructure it uses will help us make any progress in this area.
The current keyboard "multiplexing" in Emacs doesn't really support
any concurrent input in any useful sense of that word.

That's why I think we need to start from the basics, and define the
features we'd need.  In general, since we are talking about several
different individuals, the concept of having a single Emacs "server"
sounds a non-starter to me.  We should talk about several separate
Emacs sessions communicating between them in some way.