Re: progress on HTML 2.0 reconstruction

Bert Bos (bert@let.rug.nl)
Wed, 29 Mar 95 04:37:42 EST

Joe English writes:
|Ignoring unrecognized elements and attributes is clearly a
|good implementation strategy, and it is appropriate to
|assume that browsers are implemented this way when
|designing HTML extensions. However, a specification
|defining a *document format* has no business describing
|what implementations should do with documents that *do not
|conform* to that format.

I agree entirely. If an author needs other tags, he should use a
different DTD, or flag the document as text/sgml rather than
text/html. Of course, it's a sensible stategy for a browser that
doesn't know SGML to try to display everything as HTML, in which case
ignoring unknown tags is the most reasonable approach. Style sheets
are just around the corner (aren't they?), so HTML-only browsers will
soon die out.

("ignoring" may not be exactly the right word, E.g., in
"<BLUE><P>blue</BLUE>" the end-tag should also end the <P>. But then,
it's a compromise anyway, so let browser writers decide for
themselves.)

Example: the following document should get type "text/sgml", despite
the public identifier:

-----------------------------------------------
<!doctype html public "-//IETF//DTD HTML//EN" [
<!element html o o (head, body) +(blue)>
<!element blue - - any>
]>
...
<blue>
<p>blue, blue, blue,...
</blue>
...
-----------------------------------------------

Bert

-- 
                          Bert Bos                      Alfa-informatica
                 <bert@let.rug.nl>           Rijksuniversiteit Groningen
    <http://www.let.rug.nl/~bert/>     Postbus 716, NL-9700 AS GRONINGEN