Re: progress on HTML 2.0 reconstruction

Albert Lunde (Albert-Lunde@nwu.edu)
Wed, 29 Mar 95 12:35:21 EST

>>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.
>
>Right. Beside isn't it an absolute no-no to dictate the implementation
>of a browser? Fixing a certain set of behaviour is roughly
>equivalent.

This seems a rather legalistic view.

We may wish to make it a recommendation for tolerant browsers rather than a
requirement, but it seems to me the most sane recommendation we could make.

I don't want to see browsers that dump core when they see unrecognized
tags, refuse to display the content of unknown tags, or refuse to display
the document at all!

There is also the matter of precident ... ignoring the tags but displaying
the content is the suggested behavior for dealing with unimplemented tags
in the oldest versions of the HTML spec I've got, even if it's not spelled
out in such detail.

Making recommendations on how to treat unknown elements or extensions is
NOT outside the scope of a specification.

I'm more concerned about preventing the HTML world from fragmenting into
mutually incomprehensible subsets than I am about preserving total purity
in any particular version of HTML. Thus I favor things that ease evolution
of the standard and practice rather than sudden breaks.

(This doesn't mean I'm opposed to authoring and checking tools with a more
strict view of correct HTML.)

---
    Albert Lunde                      Albert-Lunde@nwu.edu