Al Gilman writes:
> To follow up on what Harald.T.Alvestrand@uninett.no said ...
>
> So, I would stick with Location, to minimize the divergence.
> If the rest of the community (hi, folks!) doesn't agree with me, let's
> do Content-Location.
>
> Harald A
>
> I don't think that we can simply adopt the HTTP usage, because (I
> believe you said) they have imposed the additional rule that only
> one such reference shall appear. At least in the faq-maintainers
> community, there is a desire to mark up a copy of a document with
> headers which identify more than one "Where to get this FAQ"
> items.
HTTP uses the Location header field to indicate the exact location
corresponding to the entity being sent in the message (for 2xx responses)
or the location for automatic redirection (for 3xx responses).
It does not necessarily indicate the "best", "master", or "nearest"
location.
HTTP will use the URI header (with a much more complicated syntax) for
indicating multiple variants or locations of a single resource.
It will perform essentially the same function as message/external-body,
but without the goofy multiparts (the multipart forms will also be
available, assuming someone implements them).
I strongly recommend against using the Content-* prefix for header
information which does not describe the actual content of the message.
Location and URI describe control attributes of the message source
and not the message content.
...Roy T. Fielding
Department of Information & Computer Science (fielding@ics.uci.edu)
University of California, Irvine, CA 92717-3425 fax:+1(714)824-4056
http://www.ics.uci.edu/~fielding/
[
Usenet Hypertext FAQ Archive |
Search Mail Archive |
Authors |
Usenet
]
[
1993 |
1994 |
1995 |
1996 |
1997
]
© Copyright The Landfield Group, 1997
All rights reserved