Re: RS/RE: basic questions

One of the reasons why people tend to find SGML's RE/RS processing rules
confusing is that people want to do different things with REs in different
contexts. Would it not be simpler to provide an architectural form that
would add a compulsory attribute, with a default value that has to be
specified in the DTD (if it is present, or defaults to a browser specified
default if it is not), that allowed users to specify clearly whether or not
they wanted RE's to be retained in a particular element, e.g.

    linebreaks   (spaces|hardline|ignore-after-markup)    ignore-after-markup

By combining this with rules such as "strip any leading and trailing spaces
in an element" it should be possible to come up with clean but flexible
rules for processing RE's that would not conflict with SGML's rules.
Martin Bryan, The SGML Centre, Churchdown, Glos. GL3 2PU, UK 
Phone/Fax: +44 1452 714029   WWW home page: http://www.u-net.com/~sgml/