Date: Wed, 29 Jan 2003 10:01:57 -0800 (PST)
   From: Mark Crispin <[EMAIL PROTECTED]>

   On Tue, 28 Jan 2003, Timo Sirainen wrote:
   > Sure. And I don't think I've said anything that would indicate otherwise.
   > The keyword in my comment above was *LESS* resource intensive. Using
   > multiple connections may be light, but using STATUS may still be lighter.

   NEVERTHELESS...

   The specification is explicit that there is no requirement that STATUS be
   fast.  STATUS could be very resource intensive on a server, and there is
   *no* way for the client to know this.

   Therefore, a client MUST assume that STATUS will be resource intensive.

Nonsense. A client should be implemented with the author's best guess
of what is and isn't expensive. Very few servers have modes where
STATUS is more expensive than open connections. NOOPs may also be
expensive but a client would be foolish to assume that NOOPs will be
resource intensive. (The spec merely says it is the "preferred method
to do" "a periodic poll for new messages or message status updates".)

The performance of STATUS or NOOP or extra connections is a
quality-of-server implementation issue. If clients require fast
STATUSes, people will select servers that give them that. If clients
require hundreds of simultaneous connections to a single server,
people will select servers that give them that.

Larry

Reply via email to