- From: Jeremie Miller <jmiller@mwci.net>
- Date: Mon, 29 Jan 1996 13:19:29 -0600
- To: Mike Meyer <mwm@contessa.phone.net>
- Cc: www-talk@w3.org
Mike Meyer wrote: > > > > Could most of this be handled with media type registration? E.g., if > > > Netscape were to accept: text/html and text/netscape-2.0-html, then > > > Microsoft's browser could express its willingness to accept either or > > > both. Is this a workable solution? > > I agree with others that this is probably not adequate for the > > distinctions content providers want to make. > > Just specifying text/netscape-html probably works for 90% of the > content providers who care, as most of them seem to want nothing more > than to distinguish netscape from the rest of the world. Adding a > version number to the MIME type (either as text/netscape-#.#-html or > as text/netscape-html; version=2.0) would probably keep 90% of the > rest happy, as they get the exact same information from this Accept: > header as they are now getting from the User-agent: header. > > MS could then achieve the same effect they got with the User-Agent: > hack by adding a single type to the Accept: headers. This removes > their incentive to do user-agent hacking. This should make the few > content providers who care enough to distinguish between more than a > fraction of a percent of the available browsers happier as well. > > Yes, it's not a complete solution, and it's not perfect. But it solves > the majority of users problems, and is a lot better than what we've > got now. > > <mike I am not sure if it has already been discussed, but why not have a header similiar to Accept-Charset, like "Accept-Features:" and and a list of standard HTML features like tables, frames, forms, etc... so that the browser can explicitly describe what it can do? Yes, unfortunately it is a "stopgap" solution, but it seems to cover quite a few of the content negotiation problems. If the server also replies with a "Content-Features:" header, the proxy could cache copies for all further requests with the same "Accept-Features:" header. Just an idea. Jeremie Miller jmiller@mwci.net
Received on Monday, 29 January 1996 14:24:50 UTC