2.1.  Published Documents

2.2.  New Internet Drafts Submitted to IETF

2.3.  Updated Internet Drafts and Published RFCs

3.1.  TC AUTODISCOVERY

We discussed the status of the current draft which has now moved over for consideration in the IETF. Recent changes include a switch to using JSON instead of XML. The IETF should be holding a BOF at the next IETF meeting to develop interest there with a view to setting up a working group to develop the specification further. The TC will in the future work on supporting the IETF effort and promote interoperability testing at CalConnect.

3.2.  TC CALDAV

3.3.  TC EVENTPUB

We briefly covered the state of the new properties defined for event publishing, STRUCTURED-LOCATION, PARTICIPANT, STRUCTURE-RESOURCE and STYLED-DESCRIPTION. These have been described and discussed at previous Roundtables so this was largely a status report. The internet draft has been refreshed and there are a number of comments to be circulated to the group and discussed.

Most of the session was taken up discussing the need for new properties to enable enhancements in itineraries. Examples are the need to specify starting and ending locations for events, chaining of events and tasks, and properties to hold information such as flights. It was agreed that detailed information may be better kept elsewhere with well-defined references to that information. Next steps for itineraries involve a discussion of needs and identifying which TC or TCs will handle the work. It is hoped that writing a “7 Things You Should Know About Itineraries” or a white paper may help attract new members to the TC. The other extensions draft which covers images has also been refreshed, and in the next few weeks we will discuss how to make progress with both drafts.

3.4.  TC FREEBUSY

The TC hosted the Consensus Scheduling Workshop on Wednesday afternoon. The workshop covered the key aspects of consensus scheduling and described CalConnect’s proposed solution: VPOLL. The workshop started with a detailed overview of what consensus scheduling is, why people use it, and why standards — currently lacking — would enhance the process. This was followed by overviews of various products that do some form of consensus scheduling or booking. We then had a detailed overview of the VPOLL specification, covering the key elements and recent changes made by the TC. This included a description of how the POLL-MODE property will be used to define various modes and levels of complexity for polls. Initially there will be a simple “basic” mode that will be used as the basis for initial interoperability testing and will be used to encourage implementors to develop products. Going forward the TC will work on refinements to the VPOLL specification based on feedback from implementors. We plan on VPOLL testing at the next Interoperability Test Event event.

3.5.  TC IOPTEST

The interoperability testing sessions were fairly busy and featured both new and established testers. There was a significant amount of work testing many of the basic CalDAV functions and beyond that to testing implicit scheduling. Some of us spent most of our time working on iSchedule, testing the new DKIM canonicalization. This was a successful test with events and freebusy being transferred. Most problems related to the non-standard disposition of the services. There was some work in getting the CalDAV test and performance suites running. In addition there was some timezone service testing with the new JSON format and a certain amount of iMIP.

A useful feature of the sessions was the occasional informal discussions which involved most of the participants. We covered some of the detailed issues related to sharing and notifications.

3.6.  TC ISCHEDULE

After an general introduction to iSchedule, we reviewed presented the main changes in Draft 03.

Interoperability testing feedback:

IETF status:

3.7.  TC RESOURCE

We had a short session where we reviewed the work of the TC to date, current discussions, and work remaining. The new vCard objectclass property and its value of a schedulable was introduced and we went over an example of how it would be used. The TC is working on 3 drafts to get the work standardized — those just mentioned plus vCard Representation of Resources.

3.8.  TC TIMEZONE

The current state of the Timezone Service Protocol was presented. The only change is from an XML based approach to JSON. This approach is more in line with the expected use by web developers. The main outstanding service issue is how to deal with inactive timezones. While there appear to be about 80 timezones currently in use there are over 400 in the Olson database. We discussed timezones by reference: that is, servers don’t send timezone data and clients use their own cached copy. This is largely in line with current practice though not in line with the specifications. We talked about a registry for timezone related information and how we should proceed. Finally the issue of allowing a TZID for DATE values was mentioned — RFC5545 disallows it, but users want it.

3.9.  TC USECASE

Lively first discussion about the merits of calendar user community involvement in CalConnect and possible ways to enhance CalConnect discovery (that is, finding out CalConnect exists and what it is), to provide useful value to the user community, plus some concrete first steps CalConnect can take to support the commerce of ideas and needs between user and vendor communities.

3.10.  TC XML

We reviewed the history of xCal: the XML format for iCalendar, how we developed the SOAP and RESTful CalWS protocols and how we had influenced the development of WS-Calendar within OASIS for the SmartGrid. The current state of the jCal data model was described and we finished up by agreeing to try to get server side libraries working for the next Roundtable.

3.11.  VTODO AD HOC

The session covered the proposed charter and work items that the Ad Hoc has been working on since the last Roundtable. We have carefully analyzed the requirements for improving task handling in iCalendar, prioritizing the work we think needs to be done. We looked at other task-based systems, including WSHumanTask, to inform our work. In addition, we took care to rule out certain issues to keep the scope of any TC constrained to a reasonable set of deliverables. The Ad Hoc anticipates that a number of items will re-use work done in other TCs. At this point we proposed that a TC be setup in CalConnect to move forward with this work, and expect a decision on setting up a TC to be made during the Roundtable Plenary. TC TASKS was approved, and we will move ahead with the work items.