I think that it can help to fight the myth that the way to "extend"
HTML is through new tags. My personal feeling is that the draft
standard[1] contributes to this problem with these lines:
To facilitate experimentation and interoperability between
implementations of various versions of HTML, the installed base
of HTML user agents supports a superset of the HTML 2.0 language
by reducing it to HTML 2.0: markup in the form of a start-tag or
end-tag, whose generic identifier is not declared is mapped to
nothing during tokenization. Undeclared attributes are treated
similarly. The entire attribute specification of an unknown
attribute (i.e., the unknown attribute and its value, if any)
should be ignored. On the other hand, references to undeclared
entities should be treated as data characters.
The standard goes on to say things which I very much agree with about
not depending on this behaviour. I'm not sure what the point of this
paragraph is, though. Is it meant to encourage a behaviour? or
specify it? Of course any good Internet client program tries to
gracefully recover from bad input, but why single out this particular
mistake? Why is this paragraph in the HTML specification?
The SGML mechanism for "experimentation and interoperability" is the
DTD, not client-side down-translation of incorrect data. If clients
do not know how to read a DTD, then servers must do the
down-translation.
Paul Prescod
[1]ftp://ds.internic.net/internet-drafts/draft-ietf-hmtl-spec-04.txt