(joint F2F meeting with TCIC at TechEx)
Klara Jelinkova, Rice University
Dennis Cromwell, Indiana University
Michael Gettes, Penn State
Ted Hanss, University of Michigan
Sean Reynolds, Northwestern University
Dave Vernon, Cornell University
Ann West, Internet2
Melissa Woo, Stony Brook University (joined remotely)
Klara Jelinkova, Rice University
Dennis Cromwell, Indiana University
Eric Denna, University of Maryland
Tracy Futhey, Duke University
Ron Kraemer, University of Notre Dame (joined remotely)
Tom Barton, University of Chicago
With: Kevin Morooney, Dave Lambert, David Gift (remotely), Chris Spadanuda, Ann West, Steve Zoppi, Ana Hunsinger (remotely)
There was discussion about the process for creating a Trust and Identity PAG (Program Advisory Group) and how to constitute the PAG membership. There is agreement that a PAG should be created by Global Summit.
There was also discussion about whether creating a PAG necessarily means doing anything with the InCommon LLC and the InCommon Steering Committee.
The consensus is that the PAG should include the governance structure for both InCommon and TIER, since both are part of the overall Trust and Identity division of Internet2. Some of the issues that need to be worked out include:
The consensus also was to create a PAG and explain to the TIER investors and InCommon participants what the PAG is and why we are doing this. Let’s do what we need to do and worry about the LLC structure when necessary.
One suggestion is that the PAG include three members from InCommon Steering, three from the TCIC, and two “at-large” members.
[AI] (Klara) will ask John Moribito about whether non-Internet2 members can be on a PAG
Kevin summarized the results from the “paths forward” meetings held in June and July; specifically the staffing models and funding requirements.
In general, the gaps exceed the funding available or, at least in the short term, the funding that can be generated by InCommon fee increases. We expect some reallocation of Internet2 revenue, which will help but will not close the gap. We also cannot have this conversation about InCommon alone - TIER and Shibboleth are the other parts of a three-legged stool. Several additional slides were recommended to be included and reconciled.
From the InCommon perspective, Kevin is looking at scaling federation operations, working on maturing the federation (and moving from “best effort” to a production service), and stabilizing the funding for Shibboleth (which is used by 90 percent of InCommon participants).
The Shibboleth Consortium needs additional support. Currently there are only three principal investors in the consortium: InCommon, JISC, and SWITCH. The consortium chair (from SWITCH) is also talking with GEANT about support, and is discussing the creation of a communications effort to gain the support (and/or increase the support) of other NRENs.
There was a discussion about a proposed increase in InCommon fees for the 2017 calendar year. The proposal is to increase fees across the board for all InCommon participants. Other potential options:
There was discussion about a fee increase and the impact on Internet2, given that Level 1 Internet2 network participation fees include InCommon fees (and there is then an internal transfer). Additionally, “system” participants of InCommon have their InCommon participation fees paid from Internet2 membership dues. Since the Internet2 2017 budget has already been set, an InCommon fee increase (and resulting increase in funds transferred internally) could require Internet2 board approval.
There was also discussion about the need to communicate the need for a fee increase and the associated urgency to do so for the 2017 year. That communication should include that this is an “urgent needs” move and not a “sustainability” move - that additional increases will be necessary.
(AI) Klara will talk with Internet2 leadership about the proposed increase, why it is necessary, and the approach regarding the Level 1 Internet2 members.
To be clear, the increase also must be approved by InCommon Steering, and that is not on the agenda for the TechEx meeting, but will be voted on during the next call on October 10, 2016.
There was also a conversation about communicating to the InCommon community not just the fee increase, but also the full picture of TIER and Trust and Identity writ large. This should include information about the short- and long-term strategy for TIER and Trust and Identity.
The overall flow of this, then, is to push the fee increase, develop the PAG, and then work with the PAG on the longer term sustainability issues.
Steve Zoppi reviewed the material that will be covered at the investor meeting. He noted that there are a number of items “in progress,” given the slower than anticipated pace. Klara noted, however, that the packaging work and the TIER demos were completed in time for TechEx.
There was a comment that the presentation conveys significant progress on TIER, but also lays out the financial challenges and makes a case for broader conversations that are needed concerning funding. It would be good to communicate that the creation of a PAG has been planned for awhile.
There was also discussion that we would like to leave the investor meeting participants to leave with the message that we need additional sites to adopt TIER.
Steve mentioned that instrumentation is now built into the dashboard, so that we will soon have fine-grained metrics for adoption. A comment was made that we need to know about adoption, but also about overall value, perhaps by gathering structured feedback.
There was discussion about whether to use release numbers for TIER. The plan has been to use a continuous release schedule, but release numbers could be added if they would be useful. The feedback was that release numbers would communicate momentum and could help in the drive to adoption. There would also be value in announcing whenever a campus goes into production with TIER.
The last meeting of the TIER investors included significant information about the existing funding gap. This time, we don’t have that message in the presentation. We may not want to make that the focus of the presentation, but it seems as if it should remain part of the conversation. If the gap is closing, we should talk about how it is being managed.
The InCommon Assurance Advisory Committee is recommending adopting baseline expectations or minimum standards for InCommon participation. This recommendation was reviewed by members of the community and none were seen as controversial or unreasonable. However, we need to consider how to deal with chronic non-compliance, as well as communicating both the risks (such as security) for such non-compliance and the penalties.
Further discussion was deferred to the next Steering meeting.
InCommon Steering and TCIC discussed some of the items from this morning with Dave Lambert. Klara said the group is committed to having a PAG structure in place by the 2017 Global Summit, and that the PAG’s first order of business will be to set up a structure that accommodates InCommon, TIER, and Trust and Identity overall.
There was also discussion about the 2017 InCommon fee increase and the potential impact on the 60 or so large universities that are part of Internet2 (and that also comprise the bulk of the TIER investors). Dave said the overall issue of an InCommon dues increase has been communicated to the Internet2 board. He will discuss these issues with Ana Huntsinger and Dave GIft in the next few days.
Klara expressed concerns about the need for communication very soon about the urgency of the InCommon funding needs. Dave mentioned that this will need to be communicated as part of a larger narrative about funding and the need for the increase. We will also need to be clear that this increase is not the end of the road, but is part of a multi-year plan for sustainability. This may require a round of validating and testing the community’s understanding of the value of the work that we do. That funding must be sustainable and fees appropriate.
One of the challenges in funding, worldwide, will be ensuring that there is a model that includes all of the NRENS that benefit from the work (say, of Shibboleth). We need a model that involves investment from those that are heavily involved and able, along with those who may be less savvy or prepared. We can’t have a model where a few are paying the way for all.