Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site cornell.UUCP
Path: utzoo!watmath!clyde!cbosgd!ihnp4!houxm!vax135!cornell!jqj
From: jqj@cornell.UUCP (J Q Johnson)
Newsgroups: net.text
Subject: Re: WYSIWYG
Message-ID: <1865@cornell.UUCP>
Date: Sun, 22-Dec-85 07:14:47 EST
Article-I.D.: cornell.1865
Posted: Sun Dec 22 07:14:47 1985
Date-Received: Wed, 25-Dec-85 00:38:17 EST
Reply-To: jqj@cornell.UUCP (J Q Johnson)
Organization: Cornell Univ. CS Dept.
Lines: 21

This discussion has generally considered a binary choice between WYSIWYG
and batch-type formatters.  I would be interested in hearing about various
temporally intermediate implementation strategies -- editors/formatters
that produce updated page images less frequently than on every keystroke but
more frequently than once per editing session.

I think most actual WYSIWYG editors in fact don't try to keep the rendering
constantly correct.  For example, Xerox STAR fixes page breaks, placement
of floating text, page numbers, and other counters only when you give an
explicit "paginate" command.  WordPerfect (my favorite WYSIWYG for
character-oriented output devices) maintains the document as correct only
up to the current cursor, so inserting characters in a paragraph does not
cause re-linebreaking until you move to the end of the paragraph.
The rest of the text is formatted, but not necessarily up to date.
One can easily imagine a WYSIWYG that reformats only on explicit command.

Clearly, delayed reformatting has major efficiency advantages (even Scribe
and LaTeX take advantage of it, getting references right only on a second
pass).  What are the best strategies for such delayed reformatting?  What
other schemes besides those I've mentioned have actually been tried?