W3C home > Mailing lists > Public > w3c-dist-auth@w3.org > October to December 2005

Re: ideas for research on wbedav required

From: Jim Whitehead <ejw@soe.ucsc.edu>
Date: Wed, 5 Oct 2005 14:15:24 -0700
Message-Id: <95C4A0DE-0347-4C59-9BA0-73F7DCB5AB78@cs.ucsc.edu>
Cc: WebDav <w3c-dist-auth@w3.org>
To: david muller <david_muller_8888@yahoo.co.uk>

Here are some thoughts on substantial research projects involving  
WebDAV.

* Photo transport protocol. The cost of 802.11x hardware is dropping  
rapidly, while the number of deployed 802.11x access points is  
increasing. Especially with VOIP gaining traction, it seems likely  
that widespread wireless networking access based on TCP/IP will soon  
be a reality. Given this background, it would be very useful for  
digital cameras to be able to automatically transfer their pictures  
from the camera to a remote server. WebDAV seems like a useful  
protocol to build upon for such a service, since it's well suited for  
synchronization, and has strong metadata facilities. The existing  
standard in this space, CPXe (http://www.i3a.org/) is focused on  
getting photos to photo finishing services for hardcopy. Many  
consumers want auto-download without being required to upload to a  
photo finishing service. Project: develop a standard for automatic  
photo upload from a camera via wireless networking.

* Identity management protocol. Many web sites require users to  
create an identify for that site, usually by creating a username/ 
password pair. Users interacting with multiple sites need to maintain  
multiple username/password pairs (or compromise security by using the  
same pair). It would be much more convenient to have a single signon  
capability. There are many existing efforts in this space (Passport,  
Liberty Alliance), with each effort having some concern over its  
degree of capture by one or more organizations or communities. Since  
identity management involves a fair amount of metadata management,  
WebDAV might be a good protocol to use as a starting point in  
crafting an identity management protocol, due to its strong support  
for metadata, and DeltaV versioning capabilities.. This issue can  
have significant impact on the cross-organization calendar scheduling  
problem (see CalDAV).

* Email transport protocol. Spam is killing SMTP. It would be nice to  
have a mail transport protocol that was point-to-point, and involved  
authentication or some form of expensive computation for people to  
write email to a remote inbox. An email message is a blog of content  
plus metadata, just like a WebDAV resource. WebDAV + ACLs might be an  
interesting starting point for creating an alternate to SMTP for mail  
transport.  WebDAV is already the third most popular email transport  
between mail server and mail client (MUA), since Hotmail and Outlook  
Express use WebDAV for mail transport. Note that this project might  
very well also involve a solution for identity management. For extra  
credit (and a shred of hope for adoption), ensure some form of  
interoperability with the existing SMTP-based email system (perhaps  
by putting all of this email into an "untrusted" or "unverified"  
folder).

RDF may very well be a useful technology in addressing the metadata  
aspects of these projects. Or it might not. Let the requirements  
drive your choice of technology, not vice-versa.

- Jim Whitehead
Assistant Professor
Computer Science
UC Santa Cruz
Received on Wednesday, 5 October 2005 21:15:41 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 18:44:10 GMT