Re: Beta: Fatal Error: No DOCTYPE specified!

Bjoern Hoehrmann <derhoermi@gmx.net> wrote:

These issues as a whole have been assigned bug number #7.

><http://validator.w3.org:8001/check?uri=http%3A%2F%2Fwww.microsoft.com>:
>
>Fatal Error: No DOCTYPE specified! [...]
>
>1) I don't see any good reason to refuse validation completly. It's very
>   simple to choose document types to default to

This is a conscious choice; any time we try to guess at the document type
we move into "heuristics" territory and it's too sordid a mess to sort out
easily. It turned out that in practice the heuristics were wrong more often
then not and encouraged people to omit the DOCTYPE Declaration.


>2) the page should display the revalidate form

Agreed, but it may be tricky to implement. This is handled as a fatal error
which bails out with an exception that is handled generically (so it's hard
to make as special case for this).


>3) It's "document type declaration", not "DOCTYPE declaration", please
>keep the terminology straight

Actually, AFAICT, it's "Document Type Definition" and "DOCTYPE
Declaration". The "<!DOCTYPE" bit is a SGML/XML Declaration named
"DOCTYPE". The "DTD" is a "Document Type Definition"; IOW it's actually
what defines the document type (aka. "syntax"). The former refers to the
latter.


>4) The phrase talks about a "first line", while the document type
>declaration in the example takes two lines

Pedant! :-)

Yeah, thanks, good catch! I'll fix it ASAP. How does "At the very beginning
of your document" sound to you?


>5) The example should (as per XHTML 1.0) include a XML declaration

The XML Declaration is problematic in that current browsers (FSVO
"current") tend to do undesireable things when faced with one; and XHTML
1.0 also reccomends against including one in "Appendix C" documents.
Divining what was intended to conform to Appendix C and what doesn't, is
deep voodoo most of the time.

It's noted and the issue will stay open until what constitutes "current"
browsers all handle it somewhat gracefullt, or until someone convinces me
that the deployed number of such browsers is sufficiently small the they
can be disregarded.


>6) The example should (as per XHTML 1.0) use both, the lang and the
>xml:lang attribute

Hmm. Isn't the "lang" attribute a compatibility crutch? Or does WAI/WCAG
overrule that?


>7) A "List List"? :-)

Woops! Will fix. :-)


-- 
"Frailty, thy name is woman!"                    - Hamlet, Prince of Denmark.
                   See Project Gutenberg <URL:http://promo.net/pg/> for more.

Received on Thursday, 24 October 2002 22:07:49 UTC