LEMONADE Notifications chat
January 25, 2007
=======================================================
Full log:
http://www.ietf.org/meetings/ietf-logs/lemonade/2007-01-25.html
Online personalities:
bernard.desruisseaux
eburger – Eric Burger
alexeymelnikov
Glenn Parsons
Dave Cridland
Ams - Abhijit Menon-Sen
Lisa – Lisa Dusseault
Arnt – Arnt Gulbrandsen
Rohan – Rohan Mahy
Discussion
-------------
Current documents:
draft-newman-lemonade-msgevent-00
draft-ietf-lemonade-notifications-03
OMA MEM REQ & AD
draft-mahy-sieve-notify-sip-00
Types of notifications
Types of out of band notifications
Things to decide about notifications:
Concern about clients avoiding sync really being a subvertive IMAPv5
Concern on frequency of SIEVE messages for notifications has been alleviated, however there is still an issue on how to deal with multiple devices.
SIEVE scripts (for mobile devices) will tend to be semi-permanent, there was a concern noted that designing for this might prohibit dynamic scripts. SIEVE is obvious for ‘user oriented’ notifications, but it is overkill for MWI. MWI is already defined using SIP (RFC 3842) but is that necessary for email?
Notifications are best-effort. And account related notifications are also out of scope.
Consensus: We are only talking about email-related notifications, not calendar notifications, and these can be sent with a number of protocols.
Structured notifications or free text?
Use cases?
Security needed for notifications
Focus on MWI first (e.g., using IMAP IDLE), leave the user details to SIEVE draft. That is no subject or other data. However, the SIP MWI already does this – counts are reliable, the other data is not.
Consensus: Reuse RFC 3842 – or an XMPP clone of it if necessary, unless a good argument is made to the contrary.
Summary:
_______________________________________________ lemonade mailing list lemonade at ietf.org https://www1.ietf.org/mailman/listinfo/lemonade Supplemental Web Site: http://www.standardstrack.com/ietf/lemonade