IPP Mail Archive: IPP> transaction-based HTTP

IPP> transaction-based HTTP

Larry Masinter (masinter@parc.xerox.com)
Fri, 31 Jan 1997 01:22:28 PST

HTTP/1.1 has both persistent connections and pipelining, so that you
can do several operations on the same connection without closing or
waiting for the return from one request before sending a second.

Persistent connections are optional; this is a good idea for a
small-memory server that wants to service multiple status requests.
(One job printing while multiple other clients are querying status).

Do any of the scenarios talk about the case of "50 simultaneous print
job attempts on a printer than can't buffer?". You'd probably like to
find out about capabilities and status (and even wait for availablity)
even from limited memory machines.

Larry