*MACE Conference Call*
May 20, 2002

*Attendees*

Bob Morgan (chair) - Washington
Michael Gettes - Georgetown
Neal McBurnett - Internet2
Jim Jokl - Virginia
Brian Gilmore - Edinburgh
Renee Frost - Michigan/Internet2
Keith Hazelton - Wisconsin
Ken Klingenstein - Colorado/Internet2
Steven Carmody - Brown
Scott Cantor - OSU
Mark Poepping - CMU
Ben Chinowsky (scribe) - Internet2

*Discussion*

Ken opened the discussion with a report on the NSF's recent NMI review. Overall the panel was pleased with progress so far. They said that more work is needed on integrating the various parts of the effort, and in particular stressed campus Grid integration, making eduPerson global, and creating international trust standards. The funding picture is good. While most of the people at the review meeting were Grid-aligned, interest in non-Grid middleware nonetheless appears to be on the rise. Ken stressed the need for work on authZ and came away with a sense that this will make a good focus for the second year of NMI.

After reviewing the recent Internet2 Member Meeting (Bob noted that attendance at middleware working-group sesssions was the highest yet) and plans for the upcoming CAMP meetings, the group discussed Ken's suggestion to set up an advisory board for federated security issues. Ken sees a need for discussion of how the Liberty Alliance conception of federation relates to other concepts of federation and to the business requirements of the large corporations involved in Liberty. In particular, overlapping federations and semi-lateral subsets of federations are major concerns. Scott noted that there seems to be an emerging consensus on keeping the federated "clubs" lightweight; Ken noted that doing so imposes greater burdens on individual sites, and makes the user-interface issues more complex. There was general agreement that such an effort would be useful, and also that it is very important to keep expectations modest, billing this as a place where these issues can be discussed, but not necessarily the place where they will be definitively resolved.

MACE discussed the digital rights management working group now forming within VidMid, under the leadership of Mairéad Martin. The scope of such a group is potentially very broad; Bob expressed concern that it be kept manageable, and suggested a focus on support for fair use for "the kinds of things we want our community to be able to do". Bob also warned against the illusion that if we design a technology that doesn't put draconian restrictions on fair use, such draconian restrictions won't end up being applied anyway -- "fighting back with technology when the issue is that they're greedy doesn't solve the problem." Mairead has secured NSF funding for a DRM workshop; the agenda for the DRM working group will take shape there. Keith pointed the group to http://xml.coverpages.org/drm.html for information on DRM technologies.

Finally, Ken noted that he'd met with some people working on K-12 projects within SRI International (www.sri.com); they are currently using PKI for internal authentication, and are very anxious to deploy Shibboleth. Ken described SRI's projected uses for Shibboleth as "an absolutely classical case of RBAC where the group has already defined the roles for us," and noted that bringing in Shibboleth to support this group is exactly the kind of thing that the NSF review panel wants to see happen.