I am the assigned Gen-ART reviewer for this draft. The General Area Review Team (Gen-ART) reviews all IETF documents being processed by the IESG for the IETF Chair. Please treat these comments just like any other last call comments. For more information, please see the FAQ at . Document: draft-ietf-calext-jscalendarbis-?? Reviewer: Thomas Fossati Review Date: 2026-11-13 IETF LC End Date: 2026-11-26 IESG Telechat date: Not scheduled for a telechat # Summary This is a carefully written document; thanks editors and the Calext WG. Appendix A is very useful to understand the delta between 1.0 and 2.0. The issues I've found are minor, mainly around backward-compatibility and a few gaps in the IANA text. I reckon they can be easily fixed. # Major issues none # Minor issues ## §1.7.3 It says: > Any JSCalendar object including a property which is reserved in context of this object MUST be considered invalid. which confuses me because, if taken literally, it seems to imply that a 1.0 object with a "replyTo"/"sendTo" (or any other reserved property coming from the RFC8984 pool) is now invalid, which I'd be suprised if it were the desired outcome. I think the document should state a bit more clearly how properties that used to be common but are now reserved are treated in 1.0 data. Up one level, AFAICS the draft never says what a 2.0 implementation should do with 1.0 data. ## §3.3.4 It says: > A pointer in the PatchObject MUST NOT exactly match one of: > [LIST...] where [LIST...] does not include "version", which (unless I am horribly mistaken) should, otherwise an oject could end up in a weird / conflicting situation. Besides, what are the patchability rules for "organizerCalendarAddress", "mainLocationId", and "endTimeZone"? ## IANA ### §7.4.2.3 * Both "description" and "descriptionContentType" seem to be missing Group in their context. * I could not completely grasp the obsolescence story for "TimeZone". My review of this part is therefore incomplete, apologies. ### §7.6.4 * Participant progress: the enum has a couple of things that look slightly off: * the registry entries say "Since Version 2.0" but they were already defined in RFC8984, §5.2.5 * "needs-action" has been dropped and its removal is not documented in A.2.3 (whereas "cancelled" is) ## §7.1 It says: > If this parameter is set, then the values of all JSCalendar "version" (Section 3.1.2) properties in the body part MUST match the parameter value. It should probably describe more precisely how the MUST applies with "version=1.0", since that can be silent in the body (Quick related check: did you run this update through the media-types list?) ## §7.2 It says: > The designated expert (DE) decides if a major or minor version change is required and assigns the new version to the "JSCalendar Version" registry (Section 7.3). which got me a bit puzzled because a bump in major requires Standards Action, where no DE is involved. ## Sec Cons There are a few things that got my antennas tingling around the opportunity for spoofing and HTML sanitisation of the fields that allow text/html. But I'll let secdir reviewer poke at that. ## Refs * All the stale refs will adjust automatically, except ical-tasks that needs to be manually changed to RFC10051 * RFC1034 and RFC1035 could be safely demoted to informative, I reckon # Nits * §7.6.4: s/registry entry/registry/ ? * §1.9.1: are you OK with "jsversion" with leading zeroes? If not, you need to tighten the ABNF * §3.1.8: it left me wondering what are the _exact_ values allowed? "itip" only? Is the values set extensible? If so, what is the procedure? * §A.2.2.1 lists "scheduleSequence" and "scheduleUpdated" under "JMAP for Calendars", but the prose talks about iTIP. Should this rather go to A.2.2.2?