(Slightly Off-Topic) Emacs-like Office App

[bobstopper@australispro.com.au]

Hi

This isn't really a request for help with Emacs but a request for knowledge
on Applications inspired by Emacs, specifically office type Applications.

I love how Emacs is so extensible and provides so many features due to its
use of modes and elisp. So what I'm (very idly at this stage) considering
is the idea of an office kinda suite (yeah, like M$) all within the one
program through the use of emacs style modes and retaining a lot of
extensibility through a lisp dialect (probably guile).

Emacs can probably do something similar to what I'm thinking already if 
support was written for it, but I'm thinking Emacs would probably remain,
well, ugly in the eyes of typical office workers and it would thus
be unappealing.

Instead I'm envisioning something that looks a lot like current gnome
office products... only it's capable of doing all of them simply by loading
the appropriate modes (perhaps after first writing them ;).

Gnome office is a great idea and I think it's a great alternative to M$
office but I think the idea of having separate programs attempting to 
integrate through some additional system like bonobo or OLE will still
remain somewhat unintegrated and perhaps not as consistent in its interface
as what I'm suggesting would be. I imagine typical office users would be
much more satisfied if they only needed the one app that could do everything
and I imagine advanced office users and programmers would be much more
satisfied if they can extend that app to do almost anything they please.

So if something *better* (rather than just equivalent) than M$ office is
written as part of the GNU/Linux project then we have much more clout for
winning the average user over. I imagine for this sort of goal the app
would almost certainly have to have a fully featured M$ Windows port so
it can first appeal to all these M$ windows users, and once they're won
over it's a small step to realising that if everything they need for
office work is in that app, and that app is available on GNU why not just
use GNU?

So, with that description of my infant idea in mind, I have a couple of
questions:

Does anyone know if something like what I'm describing already exists
or is in the process of being written etc?

and of course:

Is my idea stupid, infeasible, lacking in some major consideration, too
damned hard etc or is it actually a good idea?

I'm looking forward to any kinds of comments/suggestions. Thanks!