- From: Daniel DuBois <ddubois@rafiki.spyglass.com>
- Date: Wed, 24 Jan 1996 16:04:58 -0600
- To: Peter J Churchyard <pjc@trusted.com>
- Cc: http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
I think the idea that content negotiation should be scrapped is the most
ridiculous thing ever proposed to this working group.
Being of the opinion that content negotiation is such a good thing that it
should be generalized for multiple protocols, and hence, given its own
document outside of HTTP is one thing, but to say that it Accept headers as
they exist have no value is beyond preposterous.
There's not a person in this group who is currently content with the status
quo User-Agent-based negotiation, nor should they be. And removing content
negotiation from HTTP is not going to change or deter User-Agent
negotiation. Give up NOW on the idea that URL=entity=cache and that no one
will ever send different things to different browsers/users.
>> I would be happy to review real figures on how content negotiation is used.
>0% for charset.
I don't know how much it's being used, but there are software products in
existence that support content negotiation based, including charset, as per
the latest HTTP spec. The Spyglass Server reports charsets properly, and I
believe Alis detects the charset parameters and renders responses
appropriately. (I don't know what Navigator does with/means by its
'document encodings', but certianly if Navigator doesn't do the auto-charset
adjustment now, it will soon. It's too trivial not to make that next step.)
-----
Dan DuBois, Software Animal http://www.spyglass.com/~ddubois/
Download a totally free copy of the Spyglass Web Server today!
http://www.spyglass.com/products/server_download.html
Received on Wednesday, 24 January 1996 14:10:05 UTC