- From: Henrik Frystyk <frystyk@bay.lcs.mit.edu>
- Date: Wed, 15 Feb 1995 11:54:23 +0500
- To: http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com, mau@beatles.cselt.stet.it
> I am sorry to peruse this list (and Roy time!), but is seems that > www-talk is down - and yes, I am subscribed to it. www-talk and all the other www-* mailing lists which is run from CERN are down right now. We are about to set up a new mail service at W3C, but it will still take a couple of days before we are there. > Am I correct in inferring that the only difference between a (generic) client > and a user agent, as far as is concerned in HTTP specifics, is that the > latter is *your* client which is making requests (or maybe just the part of > the application devoted to this)? The `client' and `server' definitions are separated from what we normally think of as beeing the client and the server which of course is the Line Mode Browser and the CERN server respectively ;-) The reason for this is that we would like to decrease the difference between a client and a server in HTTP. Any application should be able to at any time to be a client and/or a server depending on which action it wants to perfom. This is to keep the possibility open for extending HTTP to have server communications for exchange of cache information etc. We use `user agent' and `origin server' to actually describe the Line Mode Browser and the CERN server or whatever other application which has chosen to be the origin location for either a request or a response. -- cheers -- Henrik Frystyk
Received on Wednesday, 15 February 1995 09:01:24 UTC