First WebDAV book now available

Jim Whitehead ejw at cse.ucsc.edu
Tue Nov 4 10:41:44 PST 2003


Lucas Gonze asks:
> A thing I've been wondering about -- how come DAV is IETF rather than
> W3C, Jim?

When we were first starting up WebDAV in the spring/summer of 1996, it was
initially conceived as a W3C activity. I initially responded to a call for
volunteers sent out by Dan Connolly (W3C), who was looking for people to
work on this project. At the time, the W3C was coming under significant
pressure from its members to focus on an achievable set of activities. As
the Web was taking off, there were lots of things the W3C could do that
would have significant impact, but trying to do them all at once would just
dissipate the energy of the W3C staff. Additionally, the W3C liason for
Netscape at the time, Carl Cargill, was somewhat supportive, but also openly
skeptical that WebDAV would have significant rapid impact. He viewed it as a
technology that could have significant effects on organizational information
sharing, but that these effects could take decades to achieve. He may yet be
right.

While there was significant interest from W3C members to work on
collaborative authoring, in the end it wasn't as significant as for, say,
XML. There also weren't any W3C staff members who were willing to champion
collaborative authoring. Henrik Nielsen (W3C) and Dan Connolly both thought
it was important, but it wasn't the top priority for either of them.

Another current was concern that the W3C didn't have much experience
developing network protocols.

We had a BOF session at the IETF meeting in November. Despite the
traditional IETF BOF hazing process, a post-BOF meeting with Keith Moore
(the Applications AD for the WebDAV BOF) made it clear that he was amenable
to an IETF working group, and so instead of waiting for the W3C to maybe
create a WebDAV activity (and probably not), we formed an IETF working
group. Another factor in this decision was that it freed WebDAV from W3C
oversight. As a nascent W3C activity, I was clearing most significant
decisions via the W3C (specifically Dan Connolly), and since he was
fantastically busy at the time, he often was unable to be especially
reactive to email. This annoyance went away once we were an IETF working
group.

In the end, I think we ended up with a better protocol than if we had worked
through the W3C. The IETF process allowed us to leverage the talents of many
people with deep protocol development expertise.

To this day, the W3C hasn't really internalized WebDAV. There are many
opporunities for leveraging DAV for RDF and XML Query, and these haven't yet
been realized. Obvious synergies between XSLT and DAV properties are only
just starting to come to light (use of XSLT to automatically generate HTML
from DAV property values, as in
http://www.soe.ucsc.edu/~ejw/papers/cdrom-production.pdf). Similarly, WebDAV
has been slower to incorporate relevant W3C specifications than if it had
been a W3C activity (though luckily Julian Reschke, with his deep knowledge
of the XML standards space, has pushed WebDAV in the right direction with
many of these).

- Jim



More information about the FoRK mailing list