Re: [WICD] comments

Hi Bert,

On 2/7/06, Bert Bos <bert@w3.org> wrote:
> So Ned Freed is disappointed with the quality of many of the documents
> that are labeled "text/html". Does that mean "text/html" is "generally"
> frowned upon? The thread you quoted shows the opposite.

While it is true that it is possible to craft HTML such that the
"display to user" fallback is reasonable behaviour, and therefore
text/html the appropriate type, it is not the case that most - or even
a significant amount of - existing HTML content is suitable for
display as plain text.

> You can count one vote, mine, less for "application/*". Is that enough
> to turn "generally agreed" into "generally not agreed"?

Well, the HTML WG at the time (and now the CDF WG) agreed that
application/xhtml+xml was the best way forward, after being convinced
of that by the IETF.  No disrespect to you meant of course, but no, I
don't think your position changes that.

> Why do I need to teach my software that "application/foo" is unreadable,
> *except* when foo is "xhtml+xml", in which case it can be treated as if
> it was "text/xhtml", line endings and character encoding and all?

application/xhtml+xml *is* unreadable in the general case.  It is not
an exception.

> I understand that the CDF (WICD?) WG doesn't want to create new MIME
> types

That's not the case.  We are prepared to register new media types when
required,  we just didn't feel one was required in this case.  This
was one of the tougher decisions we made, in fact (if not the
toughest).

> and so you depend on the HTML WG, but let this be recorded so
> that whoever next creates a MIME type for an XML-based formats won't
> assume that it is "generally agreed" that "text/*" is incorrect.

Though I haven't discussed this with the WG, I doubt it's within our
scope to document such a practice.  You might consider contributing
some text to RFC 3023bis though (which is currently dead AIUI, but
will eventually resurrect because it's needed).

Please let us know within two weeks if this doesn't address your issue.  Thanks.

Mark.

Received on Tuesday, 7 February 2006 16:54:50 UTC