*MACE Conference Call*
November 4, 2002
*Attendees*
Bob Morgan (chair) - Washington
Ton Verschuren - SURFnet
Renee Frost - Michigan/Internet2
Neal McBurnett - Internet2
Steven Carmody - Brown
Ken Klingenstein - Colorado/Internet2
Scott Cantor - OSU
Ben Chinowsky (scribe) - Internet2
*Discussion*
Bob opened the discussion by noting the high level of interest in Shibboleth at the recent Internet2 Member Meeting. There was general agreement that Shibboleth has successfully made a speedy transition from research mode to marketing mode. Bob said that he'd had interesting conversations with WebCT representatives; they're evangelizing Shibboleth and raising "be careful what you wish for" concerns. Steven noted that the campuses are starting to get calls from people looking for support for Shibboleth; Ken noted that some Internet2 support resources are forthcoming, but that marketing may need to be scaled back to adjust to current support capacity.
The group discussed what it would like to see in future middleware meetings. Ken summed up the consensus as:
- more participants beyond the usual suspects
- better mechanisms for encouraging discussion, e.g. more discussion time in the sessions, better room layouts, setting up a booth specifically for the purpose of discussion
- more CAMPs
There was also interest in organizing a WebISO workshop that would include vendors. [AI] Ken and Renee will look for appropriate January or February meetings on which to piggyback a WebISO workshop.
Ken noted that NMI is getting pressure to start including components from a wider range of sources, and suggested that NMI also needs to address mechanical issues like report and announce lists and a CVS, as well as licensing structures and the issue of who pays for integration. Ken noted a difference between how the GRIDS and EDIT sides of NMI are presenting their material: on the GRIDS side, it's being presented as the software you should install if you're a Grid site; on the EDIT side, it's "a bucket of stuff" for sites to use pieces of as they find appropriate. GRIDS makes a commitment that all its pieces will work together; EDIT doesn't. [AI] Ken will circulate Randy Butler's views on criteria for inclusion in NMI releases, and will draft some requirements for NMI software components, with a view to PERMIS in particular.
Ken noted that the need for LDAP-to-KX.509 credential conversion has come up in both the authorization context and the Grid KX.509 deployment context, and asked the group for thoughts on how to pursue this. Bob described this as an aspect of PKI Lite, and suggested that there is no need to create a credential-conversion working group. [AI] Ken will ask Jim Jokl to get HEPKI-TAG to take up credential conversion.
Finally there was a short discussion of plans for the Internet2 document library. Bob noted that he's gotten feedback that the current presentation of documents on the Internet2 web site is confusing and doesn't meet the needs of working group chairs. Bob stressed the importance of having a consistent naming convention; he also stressed that the goals of a) helping improve working group process and inter-WG coordination, and b) making clear what working group products have been officially released to the wider world, may not have the same solutions, and that confusing these two goals doesn't help us reach either one. There was general agreement that the principal goal of the Internet2 document library should be to serve as a permanent repository for material that has been vetted for official release by the working groups, and perhaps also vetted at one more level (e.g., the Internet2 Area Directors).
*Action Items*
[AI] Ken and Renee will look for appropriate January or February meetings on which to piggyback a WebISO workshop.
[AI] Ken will circulate Randy Butler's views on criteria for inclusion in NMI releases, and will draft some requirements for NMI software components, with a view to PERMIS in particular.
[AI] Ken will ask Jim Jokl to get HEPKI-TAG to take up credential conversion.