[calsify] Re: Question about organizerCalendarAddress in JSCalendar bis

[Robert Stepanek <[e-mail]>]

On Wed, Oct 8, 2025, at 3:08 AM, Neil Jenkins wrote:
> On Tue, 7 Oct 2025, at 23:37, Robert Stepanek wrote:
>> Alternatively, we might register the OWNER role now for iCalendar if we can describe in both iCalendar and JSCalendar definitions what being an owner means without presuming some future scheduling extension or referring to JMAP server permissions (both of which are out of scope of JSCalendar as an exchange protocol).
> 
> That sounds like the best path forward to me. I think it's useful information for the data type format (rather than just being implicit from the ORGANIZER, which also restricts it to only one owner). The JMAP Calendars spec then defines permissions/restrictions based on this attribute, which is the right split in IMO. Future scheduling protocols are likely to make use of it as well, which is why I don't think think should just be defined in JMAP Calendars.

Great. I'll do some tests with a couple of calendar services and clients. Unless I run into any major blockers, I'll then draft the new definitions for the owner role in both JSCalendar and iCalendar.

The current definition in JSCalendar is close to JMAP for Calendars:
"owner": The participant is an owner of the object. This signifies they have 
permission to make changes to it that affect the other participants. 
Non-owner participants may only change properties that affect only 
themselves (for example, setting their own alerts or changing their RSVP
 status).

I think that's *too* close to JMAP for Calendars, e.g. the current definition suggests permissions and scheduling rules are somehow enforced, but for CalDAV and iTIP flows they currently are not.

I think the following phrasing preserves the intent but more clearly sets expectations:
"owner": The participant is an owner of the calendar object. This signifies they can make changes that affect all participants (for example, rescheduling the calendar object, adding and removing participants and roles). The presence of this role only is indicative, its semantics are subject to the calendaring exchange protocol being used. See [JMAP for Calendars] for an example for making use of this role.

What do you think?

Regards,
Robert