W3C home > Mailing lists > Public > www-tag@w3.org > April 2002

RE: Fwd: What does a document mean?

From: Larry Masinter <LMM@acm.org>
Date: Mon, 1 Apr 2002 09:25:51 -0800
To: <www-tag@w3.org>
Cc: <Misha.Wolf@reuters.com>
Message-ID: <000801c1d9a2$45d12bb0$6ace8642@larrypad>
> > At 11:25 29/3/02, Misha.Wolf@reuters.com wrote:
> > >Norm and Paul tackle the question "What Does a Document Mean?".
> > >They write "So, a first level approximation of a document's
> > >meaning is determined solely by its MIME type."  They then list
> > >eight kinds of additional information which may help us arrive
> > >at a "More Specific Meaning".  Nowhere do they mention the
> > >character encoding (of textual documents).
I suggest a more careful usage in the framework; 'MIME type'
is the wrong term.

In the MIME framework the "Content-Type" label includes
the media type (or MIME media type or MIME type) and parameters.
For types in the 'text' tree, there is often a 'charset'
parameter that denotes the character set and character encoding
combined.

It is common to see 'MIME type' (denoting only the media type itself)
without reference to the parameters, even when parameters
are significant. This is probably because many media types
don't use or have parameters (image/gif) and that many
implementations do a poor job of transmitting them
(e.g., map the media type to a file extension and drop
all parameters.)

For example, early drafts of HTTP "Accept" headers only
allowed for the media type; the ability to add parameters
was changed in HTTP/1.1.

There are also other "Content- " labels for MIME messages that
are also necessary for decoding, e.g., content-transfer-encoding
(or content-encoding in HTTP), or useful in subsequent
processing to determine 'meaning' (Content-Language).
Received on Monday, 1 April 2002 12:27:09 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:06 GMT