Attendees

Agenda

1. Introductions/Roll Call
2. Updates
   1. UC Update
      1. Data Modeling - Lucas
      2. ID Match - Benn
   2. Rutgers (OR) Update - Benn
3. Project Proposal - Eric/Jessica/Renee/Matt
   1. Requirements Chunking Exercise - Matt
   2. Draft Proposal - Eric/Jessica
   3. Timeline and Estimates Work - Renee/Matt
4. Strategy Group Updates - Bob

Action Items

• Everyone look through the chunks document and make or comment any moves/changes you'd like to suggest.

Notes

benn - weekly meetings continue with the UC subgroups looking at data modeling (for lucas), ID Match, and transport.  on the id match, working up some documents that will be shared with this group shortly.  First draft of requirements and some mockup APIs, no code writing though.  Maybe some flow diagrams, more of a straw man than an implementation proposal.  should be posted shortly in a public space for our consumption.  pretty good progress there.

renee - is there any interest in PSU's work, what's the reasoning behind this and how can it be used here.

benn - in the UC there's a joint effort and there is some convergence here.  UC doing their reqs and see what's out there.  the UC folks are interested in looking at requirements and algorithms and such.

lucas - knowing that PSU is willing to share is very much welcome

benn - tom barton has offered some of their work from chicago.  but looking at just the identity match part is what we are looking to tie together with this and other group

renee - could come out with a product or open source and people can pick and choose

benn - exactly an API layer is what we'd like to at least have to put in or link up with whichever ID Match thing you want

steve - i think that's the key thing.  in the commercial space there is a wide variety of matching algorithms and I think the OS space will catch up so having those APIs is the key to allow those to plug in what they want.  i can ask about the API question to see if they have any interest

renee - that's what we are doing, we use data flux with APIs and our algorithms based on parameters and verify

steve - that kind of structure seems to be very common with this idea that you could then plugin a variety of algorithms

benn - to wrap up, if renee can point us to the documents PSU has and if you or Steve have any hooks to your vendors to see how interested they would be in talking about the API work 

lucas - at this point we've started looking at PSU and OR data models for comparison based on how they match with their current model.  i don't know how realistic it is that we decide on one product, there might be multiple efforts where things are loosely coupled and folks can swap out what they need.  certainly interested in looking at a joint venture.  not sure why multiple things are going on, but we defiantly want to participate in things as they evolve.  we're doing our own, but closely watching what this group is doing.  but as far as the model goes we're still evaluating.

benn - at Rutgers, they have been running the code in parallel production for about 6 months now with good results with no talk about a cut over date, but possibly in the near future.  interest in bringing the project back to the open source community effort.  so it may be a more viable option that previously thought.

renee - jimmy and i, sometime, are going to look at the requirements, what we did, what we were not able to do, and see what kinds of skill sets and FTEs those would take.

matt - i know jessica has an interest in seeing that and possibly helping with that work

matt - i'd like to put out some suggestions that we get some sort of update from the strategy group or some way (like the september update meeting) where the whole OSIdM4HE group gets updated on what's going on across the entire initiative.