Tim Bray wrote:
At 08:25 AM 7/17/99 PDT, Larry Masinter wrote:
I think there is no justification for "text/xml-dtd" in addition
to "application/xml-dtd", because there is no utility at all
for an unaware recipient to attempt to display a dtd to a
user as if it were text
Sounds plausible, but empirical evidence is against this position.
Lots of people read DTDs all the time, and in fact since they are
usually composed in monotype with all sorts of indentation and
other pretty-printing apparatus, there is clearly an expectation
that they be displayed. -Tim
Yes, each to their own; for example Tim (and I) would be happy reading a
DTD in plain text; Ned Freed would be happy reading RFC822 headers in
plain text, and so on and so forth.
And again, in practice, software does not use this "fallback" of
displaying text/unknown as text/plain so software would either know what
to do with a DTD or else it would offer to save it to disk or somesuch.
I would argue against text/xml-dtd only on the grounds of the general
charset tangles with text/*, not on the basis of readability or
otherwise. If I send a DTD to someone in email, its probably because I
expect them to read it. If XML software fetches a DTD, its likely for
machine processing and probably no-one reads it.
--
Chris