Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!samsung!crackers!m2c!umvlsi!dime!cs.umass.edu!pop
From: pop@cs.umass.edu
Newsgroups: comp.lang.functional
Subject: Typography, Off-side and Concrete Syntax
Message-ID: <26844@dime.cs.umass.edu>
Date: 18 Feb 91 22:25:08 GMT
Sender: news@dime.cs.umass.edu
Reply-To: pop@cs.umass.edu ()
Organization: University of Massachusetts, Amherst
Lines: 22


       Concrete Syntax, off-side, Typography.

The benefits of uncluttered indentation can be obtained far more cleanly
by relegating parsers to where they belong - the past. What is needed is a way
of allowing users to create an abstract syntax as a result of interacting with
a workstation. This permits one to exploit the resources of mathematical
typography as a tool for communicating with one's computer.

OK - I overstated the case against parsing. It is desirable to be able to use
parsers to interpret sequences of key-strokes during interactions. One of the
capabilities that is most useful for this purpose is that of creating a
smallest legal completion of a sequence of key-strokes, so that a meaningful
display can always be maintained. The only system that does this to my
knowledge is Theorist on the Macintosh (in another domain). 

Going back to indentation -- initial experiments with a Scheme presenter 
implemented in Pantechnicon indicate that it is reasonable to create 
indented structures with a template summoned up from the "soft keyboard"
rather than by keying in the conditionals, function definitions etc..

Robin Popplestone.