# CALEXT - IETF110 (Prague virtual)

## Agenda

Wednesday 2021-03-10 15:30-16:30

Intro and Note Well: 5 min

Submitted documents: 30 min

* draft-ietf-calext-jscalendar (+icalendar) - 15 min
* draft-ietf-calext-valarm-extensions - 5 min
* draft-ietf-calext-eventpub-extensions - 5 min
* draft-ietf-calext-ical-relations - 5 min

Discussion of current work items: 20 min

* draft-ietf-calext-icalendar-series - 5 min
* draft-ietf-calext-serverside-subscriptions - 5 min
* draft-ietf-calext-subscription-upgrade - 5 min
* draft-ietf-calext-vpoll - 5 min

Milestone review: 5 min

## Minutes

* Opened meeting at 9:30am EST
* Reviewed Note Well
* Minimal Agenda Bashing
* Goodbye from Barry as outgoing AD

### JSCalendar

* Notes
    * In AUTH48
    * Differences with scheduling in iCalendar
        * ATTENDEE/ORGANIZER has single CUA
        * JSCalendar participants not a direct equivalent to CUA
        * Via ITIP there's a need to send multiple participants when talking about delegates, so you need to know which is the attendee.
        * Neil: could have some kind of UID property.
        * If you're mapping from icalendar there will only be one value, so if you have multiple then you're already doing something more powerful.
        * Robert: there's already events from Apple with a "meaningless" URI in the property.
        * Mike: has received events with iMIP sent to an address with the email parameter value as the destination for the message.
        * Mike: right now iCloud doesn't work how the spec says, so we need to make sure we are compatible with what they do.
        * Neil: being compatible with things not following existing specs is a problem!  But anything seen externally should be the mailto: value.
        * Neil: CUA value maps to sendto property; mailto: URI -> 'imip', any other URI -> 'other'
        * Mike: do we need to do anything with SENTBY?
    * Recurring events
        * Not clear if an override is an RDATE or a change to an RRULE-generated instance.  Implementations need to check
        * Standalone instance without a master needs to include a time zone id
            * Mike/Bron: this needs to be in the base spec
            * Mike: RECURRENCEID is an actual time value so knowing the TZID is important

ACTION: discussion on list to confirm cases work, then clear AUTH48 ASAP.

### VALARMs:
* in RFC Ed queue

ACTION: none needed

### EventPub:

* Notes:
    * Regarding ABNF - Barry says it's Mike's judgement call.
    * Ken: Agree that new ABNF is preferable.  Ken will check that it does what we think it does.

ACTION: Ken to check ABNF
ACTION: Mike to upload new draft

### iCal relations:

* pretty much done.
* Mike to look into adding JSCalendar mapping

ACTION: Mike to look at JSCalendar to make sure everything is covered
ACTION: WGLC

### iCalendar Series
* Neil has concerns that need to be discussed (at CalConnect)

ACTION: pause until CalConnect in April

### Server-side subscriptions:
* no progress since IETF 109

ACTION: Ken / Mike to follow up with Apple and to align with what they currently do.

### Subscription-upgrade:

* heading to WGLC

ACTION: Bron to WGLC

### VPOLL:

* needs existing implementations to switch over to PARTICIPANT
* also needs JSCalendar mapping

ACTION: Mike has a new draft ready to upload, then interop experience (again)
ACTION: discuss at CalConnect


Milestones were already updated by Daniel being efficient!  Hopefully we'll get a couple done before June.

* Closed meeting at 10:26am EST