Re: progress on HTML 2.0 reconstruction

Paul Grosso (paul@arbortext.com)
Mon, 27 Mar 95 14:27:38 EST

> From: "Roy T. Fielding" <fielding@avron.ICS.UCI.EDU>
>
> You can see my change notes at
>
> http://www.ics.uci.edu/~fielding/html/
>
> which now contains a new base draft containing the format and
> structural changes.

I know I'm jumping the gun, but I'd hate to have the wrong public
identifiers get out.

At the top of section 4, page 12, you've got it that every HTML
document should start with:

<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML//EN//2.0">

but I think we're pretty much in agreement that this is not the way.
For example, see the DTD or the SGML Open catalog later on. This
line should instead read:

<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML//EN">

to match what I read in the catalog and DTD, or perhaps:

<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 2.0//EN">

if you do want to be specific about the version.

While I'm at it, on page 11, section 3.2, the terminology about
how to "skip" undefined tags is prone to misunderstanding. In
particular, as written it sounds like you would *not* skip the
matching end tag. To replace:

...behave as though,
in the case of a tag, the whole tag had not been there
but its content had, or in the case of an attribute,
that the attribute had not been present.
I recommend:
...behave as though,
in the case of a tag, both the start tag (including its entire
attribute specification list) and matching end tag (if any) had not
been there but its content had, or in the case of an attribute,
that the entire attribute specification (for that attribute)
had not been present.

paul

Paul Grosso
VP Research, ArborText, Inc.
and
Chief Technical Officer, SGML Open

Email: paul@arbortext.com