Re: Wrong MIME type for XML DTDs

I disagree.

The content provider's choice of "text/plain" is wise in this case.

If someone browsing goes for it, that individual wants to read it.
That is text/plain consumption of the resource.

If an application, say, something using the SP library or Perl
LPW, grabs it, the combined effect of information provided by the
user and the application's author already provides the context
the application needs if it is to be used as an XML DTD.  This is,
in fact, something that has been in place for 5 or 6 years now and
works well as it is.  Conceivably, a change could even cause breakage
somewhere.

Beyond that content providers need to be very wary of possible side
effects for security that may arise from auto-generated processing on
remote platforms triggered by naive users.

Now if we want to talk about user-based content negotiation, there
might be something to discuss.  But that discussion should be in
www-talk rather than www-html.

                                    -- Bill

Bjoern Hoehrmann <derhoermi@gmx.net> wrote:

>   % http-head http://www.w3.org/TR/html/DTD/xhtml1-strict.dtd
>   HTTP/1.1 200 OK
>   Date: Sun, 29 Jul 2001 04:42:57 GMT
>   Server: Apache/1.3.6 (Unix) PHP/3.0.11
. . .
>   Content-Type: text/plain; qs=0.4
> 
> This should be application/xml-dtd according to RFC 3023. I've just
> reported, that .svg files were delivered as image/svg instead of
> image/svg+xml (this is fixed now), maybe someone should take some time
> to check for more wrong MIME types in the server configuration files.

Received on Sunday, 29 July 2001 08:24:06 UTC