- From: Koen Holtman <koen@win.tue.nl>
- Date: Sun, 11 Aug 1996 00:35:32 +0200 (MET DST)
- To: hallam@etna.ai.mit.edu
- Cc: http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
hallam@Etna.ai.mit.edu: > >Having been in the same meeting as Dave I agree that this is an >important issue. Dave, Phill, I believe that the feature negotiation mechanism in transparent content negotiation solves many of the negotiation and registration problems you have identified. Please review the draft spec (see http://gewis.win.tue.nl/~koen/conneg/) and tell me what you think. I feel that feature negotiation is fundamentally more powerful than any user-agent header or version number dependent scheme can be, because a big limitation to the power of these latter schemes is not present for feature negotiation. The limitation in question is the need to produce a small encoding, to be sent in request messages, for _all_ capabilities of the user agent and its components, and _all_ preferences of the user. Transparent content negotiation does away with this limitation, and feature negotiation exploits the absence of this limitation to the fullest extent, yielding a framework which allows an open, evolutionary approach to the problem of creating a shared language in which to express capabilities and preferences. Koen.
Received on Saturday, 10 August 1996 15:36:55 UTC