I am very pleased to announce that an effort I’ve spent nearly three years on is becoming an IETF Proposed Standard. CalDAV will have its own RFC number shortly, and the approval announcement was just last week.
This started in November of 2003, when I was attending a meeting of the CalSched working group at the IETF. I was frustrated with the circles that the participants were going in, as they worked on calendar modeling problems mixed together with protocol modeling problems, and ranted about how much simpler this would be if the protocol modeling had really been abstracted away by building on something like WebDAV (which itself, really builds on HTTP’s model). Larry Greenfield basically dared me to back this up, pointing out that the WG had no way to judge whether it would be simpler unless somebody wrote up a specific proposal. So I put my time where my mouth was, and published a sketchy -00 draft version of CalDAV.
Nothing happened immediately, but in spring of 2004 a few things happened. OSAF hired me to work on calendar standards specifically, and Oracle flew me into Montreal to talk about IETF work, CalSch, the previous CalSched proposal, and CalDAV. I had some great deep-diving-technical meetings with Bernard Desruisseaux and other very smart Oracle calendar people and we all got very enthusiastic about solving at least the basic problems. Although Bernard and Oracle had more big-enterprise requirements, in contrast to OSAF’s person-to-person sharing requirements, we figured that for basic calendar access and sharing we could solve both organizations’ requirements and the combination would be a very strong candidate for a broadly useful standard.
In September of 2004 I published an -02 draft of CalDAV with Bernard as co-author. We also added Cyrus Daboo who was already a strong contributor, particularly at reconciling different positions. We didn’t often need a tie-breaker between the three of us because we worked rather by team consensus and by polling a public list of more occasional contributors, but it certainly helped to have three in the discussions. Bernard set the conference call schedule and cracked the whip and we built momentum. Also in September 2004, the CalConnect organization started to rally implementors and calendar users together around standards.
CalConnect held the first CalDAV interoperability event in January 2005 and two more in June and September. These were really great for helping us nail down what needed to be changed, added and clarified in the draft. Oracle publicly announced its planned support for CalDAV in the next version of its Collaboration Suite. Open Source implementations also came along: not only OSAF’s Cosmo and Chandler but also the Bedework server and the Novell-sponsored Hula server, with Evolution and Mozilla Sunbird/Lightning client-side work ongoing.
Most of 2006 was spent with a generally stable proposal, overcoming the final obstacles: doing a mailing list last call, discussing contentious issues, requesting an IETF last call and more contentious issues, and finally going through IESG evaluation this summer. In August, Apple announced support for CalDAV.
My personal thanks go to everybody who contributed to this: implementors, mailing list participants, CalConnect members, people who wrote blog posts or articles. I look forward to seeing what happens now that it’s out of the protocol design process and in the hands of implementors, administrators and ultimately the users.