O V E R V I E W

IETF WEBDAV Working Group Meeting

Chicago IETF, Chicago, IL USA

August 27, 1998

The WebDAV working group held a meeting at the Chicago IETF on Thursday, August 27, 1998, from 1PM to 3PM. There were 55 people in attendace. The meeting was chaired by Jim Whitehead, and minutes were taken by Stephen Martin, and John Stracke. Jim Davis began the meeting by leading discussion on Advanced Collections capability. This discussion ran through the Advanced Collections Requirements, followed by the Advanced Collections Protocol. The most significant comment concerned the need to clarify the text to distinguish between a URL, a resource, and a reference. Jim Davis recorded several discussion points, which he will take to the discussion list.

Lisa Lippert next led discussion on the Access Control requirements document. Participants commented that the document needs to be clear on whether the goal is to make an access control protocol which exactly maps to the access control mechanisms of underlying repositories, or whether the goal is to have an access control protocol which is WebDAV-specific, and which will likely not map exactly to the access control mechanisms of many repositories. The sentiment in the room was to support the latter.

At this point, there was brief discussion of the mailing list thread on "Hierarchical URLs and collections" which was perceived to be the issue of whether the same chunk of persistent state can be accessed via multiple URLs, and if so, whether the WebDAV Distributed Authoring Protocol specification needs to set policy for how to handle operations on these multi-URL chunks of state. Discussion of this issue will continue on the list.

Chris Kaler finished the meeting by presenting a summary of the recent Versioning and Variant Authoring Design Team meeting. There was brief discussion of whether it was OK for the Design Team to address automatic versioning (versioning for downlevel clients), and simple configuration management support (similar to CVS). No dissent was raised for addressing automatic versioning. Some participants noted that the versioning protocol might be useful even if it didn't address configuration management, but didn't object to development along this path.