- From: Koen Holtman <koen@win.tue.nl>
- Date: Mon, 3 Jun 1996 13:47:14 +0200 (MET DST)
- To: burchard@cs.princeton.edu
- Cc: koen@win.tue.nl, masinter@parc.xerox.com, jg@w3.org, http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com, fielding@ics.uci.edu
Paul Burchard: > >koen@win.tue.nl (Koen Holtman) writes: >> Give me one month, and I can write a nice 10-page section, >> which people might might understand, that exactly >> defines how `understood variation' works. > >Yikes...then how can you be so confident that this will be >interoperably implemented? The requirements in 13.5 that tell caches when (not) to return responses with a Vary header field are clear, so I don't expect a divergence handling of _incoming_ Vary fields by cache implementations. The problems I detected are mostly in the section 12 text that tells servers what to put in an outgoing Vary field. Though this text is subtly wrong, I expect that this wrongness will almost never lead to someone putting less headers in a Vary field that there should be to make caches handle the response correctly. I expect that most people will intuitively make the right choice, especially if they have also read 13.5. >[...] >Paul Burchard <burchard@cs.princeton.edu> Koen.
Received on Monday, 3 June 1996 04:50:51 UTC