*ITANA Call 12-April-2012*
Jim Phelps, U. Wisconsin-Madison (chair)
Keith Hazelton, U. Wisconsin-Madison
Wayde Nie, McMaster U.
Glenn Donaldson, The Ohio State U.
Paul Hobson, UBC
RL "Bob" Morgan, U. Washington
Gary , U. Washington
Rupert Burke, U. Washington
Tony Chang, U. Washington
Leo Fernig, UBC
Sharif Nijim, Notre Dame
Bob , Notre Dame
Steve Olshansky, Internet2 (scribe)
Systems of record are authoritative for what they say, but don't have control over what is done with the data they provide. How do we manage the queue, messaging, and service calls in an effective manner?
What are acceptable levels of latency for various systems, how and when do they change, and how are they determined? How are messaging systems and brokers evaluated?
See also Change Data Capture (CDC) http://en.wikipedia.org/wiki/Change_data_capture
It was suggested that it would be useful to codify requirements for an operational datastore.
Given the breadth of the space, this would be a useful undertaking to knit together the various threads.
Q: What about higher-ed warrants its own reference architecture?
A: Distinct vertical, focused on learning and research; and unique porous/fluid set of communities being served, different from gov't and private sector.
Industrialization of research and learning were cited as long term trends affecting higher ed and EA, e.g. organizing research in a systematic way to achieve defined goals, and research as a funding source. This has implications in how supporting systems can and should be structured.
Soliciting the involvement of an established EA group (e.g. OpenGroup?) was proposed, as was talking to JISC about some related work under their umbrella.
Feedback on the call was positive. This may be a good topic for the F2F.