🚀 go-pugleaf

Wrt to your BBS project for the ST:

The commands for retrieving and manipulating messages on every micro BBS I
have ever seen have been hopelessly primitive.  With a megabyte of RAM to
work with, why shouldn't one be able to scan or read collections of
messages by any combination of criteria of sender, receiver, time, text
string, etc.? (For example: >Read From Jones To Smith During 1985 Text
Qlisp) Why shouldn't an ST BBS (particularly on the forthcoming multimeg
machines) be at least as sophisticated as Mark Crispin's MM mail processing
program that is popular on the Arpanet, and perhaps, going one step
further, include a small-scale natural language interface similar to Q&A?

Better yet, I would like to see a full-scale communications program for the
ST that included all the standard BBS functions and the mail processing
abilities of MM, Babyl, and Zmail (the last is standard on Symbolics
machines, and is the most powerful mail processor I've yet encountered). 
Such a program would be quite intelligent about automatically downloading
and storing the message bases of Compuserve, Delphi, Arpanet hosts, TBBS
systems, Fido systems, etc., presenting a unified interface for offline
retrieval, reading, and reply-writing, and automatically uploading replies
to the appropriate hosts. 

I don't actually expect you'll have the time to write such a program for
your dissertation, but I wish someone at one of the major software houses
would get on the ball. 

Wayne McGuire
(wayne@oz.ai.mit.edu)