*MACE Conference Call*
July 11, 2005

*Attendees*

Bob Morgan (chair) - Washington
Paul Hill - MIT
Ken Klingenstein - Colorado/Internet2
Jim Jokl - Virginia
Renee Frost - Michigan/Internet2
Mark Poepping - CMU
Keith Hazelton - Wisconsin
Tom Barton - Chicago
Steve Olshansky - Internet2
Lynn McRae - Stanford
Scott Cantor - OSU
Brian Gilmore - Edinburgh
David Wasley - UCOP
Steve Carmody - Brown
Ben Chinowsky (scribe) - Internet2

*Discussion*

Bob noted that the recent CAMPs were successful. The integration CAMP (http://www.educause.edu/camp052) was a traditional how-to gathering. There was a good session with Chuck Severance of Sakai; they're looking for further input via the Sakai SEPP Enterprise discussion group. The virtual organizations CAMP (http://www.educause.edu/camp053) was "somewhat more speculative" -- people came expecting answers and instead got questions -- but given the state of this area, just working toward a better understanding of the problems constitutes progress. [AI] Ken will circulate a writeup of the VO camp. The afternoon vendor sessions were less successful and will probably not be repeated.

Upcoming meetings:
- Bob, Scott, and Brendan are demoing Shibboleth interoperability at Catalyst this week.
- Keith will be spending Saturday, August 20 in Tokyo meeting with middleware researchers. [AI] Ken will find a Japanese Grid contact for Keith to brief on GridShib.
- The next EuroCAMP is being scheduled for this November in Portugal.

Internet2 and National LambdaRail (NLR) are moving toward a merger; see http://www.internet2.edu/consolidation/ for current status.

Alan Robiette has retired, creating a vacancy on MICE. [AI] Ken will add Brian to the MICE list; Brian will look into possible replacements for Alan.

Bob is working to activate the InCommon Technical Advisory Committee (TAC). [AI] All who can participate in the InCommon TAC will contact Bob.

The group discussed the possibility of contracting for professional support services for Internet2 middleware -- particularly Shibboleth, Signet, and Grouper -- using a company such as JBoss (http://jboss.com/) or rSmart (http://rsmart.com/). On the one hand, this could do a lot to build confidence on the campuses, particularly smaller campuses where funds and in-house expertise are more limited. On the other hand, much of our credibility comes from the fact that we're pushing a model rather than particular software that implements it. If there was professional support for our software, it could be harder to maintain this distinction; we'd need to be clear that the point is to ensure that e.g. Shibboleth is a viable option for federating software, and not to discourage people from using other federating software. On the third hand, we wouldn't be making any money from Internet2 middleware, so our credibility might not suffer much anyway. Scott noted the related issue of vendors starting to make claims of conformance with Internet2 middleware; we need to make sure they understand exactly what this entails.

Finally, Bob pointed out that there are several current issues -- e.g. professional support, the nature of virtual organizations, and questions around the Nexus provisioning software -- that need a wider discussion forum than MACE or MACE-Dir. The mw-discuss list was intended for this purpose but has not taken off. There was general agreement that it would be good to reactivate mw-discuss; the group will discuss this further on the July 25 call.

*Action Items*

[AI] Ken will circulate a writeup of the VO camp.
[AI] Ken will find a Japanese Grid contact for Keith to brief on GridShib.
[AI] Ken will add Brian to the MICE list; Brian will look into possible replacements for Alan.
[AI] All who can participate in the InCommon TAC will contact Bob.