Re: The Reference Concrete Syntax is not Current Practice (Was Re: Standards, Work Groups, and RealiThe Reference Concrete Syntax is not Current Practice (Was Re: Standards, Work Groups, and Reali

Marc Salomon (marc@matahari.ckm.ucsf.edu)
Tue, 26 Sep 95 15:52:56 EDT
Arjun Ray <aray@pipeline.com> writes:

|I have trawled untold megabytes of the mail archives and have yet to find
|a good discussion of the *really* insidious problem. I'm nearly convinced
|that the SGML gestalt actively hinders its appreciation, by conflating
|parsing with validation as a practical prescription, so that *categories*
|of error fail to be distinguished.

I remember in '93 reading a paper (from cern?) asserting that browser user
agents should be forgiving in parsing HTML. This is probably the reason that
we see tens of millions of pages, both conforming to a DTD and not, as opposed
to thousands of conforming pages at CERN, fermilab and SLAC.

Some documents will require full SGML in order to be rendered and many will
not. Those who chose the 'penalty' of SGML compliance at render time will be
rewarded in other ways. Browsers should still be foprgiving in what they
accept, caveat author, making the best effort to render something.

Not all data are worthy of being reused. Not all documents destined solely for
rendering by the browser class of user agents require full compliance with HTML
as an SGML application. Robot indexer user agents will probably have an easier
time with those that do.

Authors now have enough information to make the choice on where to draw the
line between fancy presentation and interoperability/reusability for their
data. The most interesting task for this wg is to ease the task of those who
want to allow their data to live long after HTML 2.0 has expired.

-marc