[Standards-JIG] Re: DEFERRED: JEP-0085 (Chat State Notifications)

Ian Paterson ian.paterson at clientside.co.uk
Thu Sep 23 12:41:54 UTC 2004


> > If and when a new revision of this JEP is published, its status
> > will be changed back to Experimental.
>
> Maybe we could add in the introduction :
> 6. The protocol adds a new state 'GONE'.
>
> IMHO the definition GONE suggested trigger should be changed to :
> "The user closes the chat window."

Yes, this example would be a very useful clarification of the existing
definition of GONE.

> The current GONE suggested trigger is superfluous because we have
> already have the INACTIVE state to notify lack of user interaction
> for a 'long' period.

My client currently sends INACTIVE when the discussion window has not been
the top window for 30 seconds. IMHO the recommended 2 minutes is too long.
For example:

If User A believes that User B is reading his last message and/or thinking
about a response, then User A will probably wait for the response. If User B
is in fact working in another window, then User A needs to know this as soon
as possible.

> we should really push this JEP for Last Call as it is simpler and
> less prone to non-compliant implementations than JEP-22.

I agree. The JEP-0022 'composing' event should be depricated. JEP-0085
should be advanced.

- Ian




More information about the Standards mailing list