Present: Steve R., Bill P., Martin M., Catherine P., John U., Stan R., Matt K.

Stumpers and proposed answers, gathered from email in response to my request for input on fork-in-the-road issues:

1. Who is responsible for data standards and content development? Who oversees? Tim Cole, Martin Mueller, Brian Pytlik-Zillig, Steve (and users of collections across the project). This group should be closely tracking uses and users for requirements, and responding to datastor for requirements as well. Martin oversees. Next steps: nail down TEI-simple and process the Wright Collection.

2. Who is building the backend? Who oversees? Bill Parod, John Norstad, Phil Burns; Bill oversees. Next steps: what collections do users need processed, and what affordances does the data stor need to offer to the proxy layer? What's the object model for the data stor, and how is that exposed, and to what? What counts of what objects will be provided? What complex queries will be supported? How will worksets be handled? Answers to these questions are to be sought from the proxy group and output requirements goes to data standards and content development.

3. Who is implementing the proxy (data layer interface definition and development)? Who oversees? Amit, Steve, James. Proxy calls should be enumerated by the analytics developers and communicated to datastor developers. Amit oversees. Next step: proxy needs to talk to mysql and/or wordhoard data model? Maybe--Amit needs to confirm; Steve will ask him.

Update: Sept 26th 2007 James and Amit have started work on the Proxy calls for metadata search. We are also exploring how Fedora could be leveraged in the middle ware architecture. Regarding Wordhoard and mysql, Amit has been in contact with Bill and will wait for him to suggest ways to integrate the Wordhoard model and proxy.

4. Who is implementing the D2K/analytics portion of the project? Who doles out that work? Loretta, Vered, Duane. Loretta oversees. Input to this group comes from interface, output feeds requirements to the proxy. Next steps: we will want very shortly some search functions. I think we need several, which will respectively search collection metadata, collection (or workset, or document) fulltext, and features (word lists, n-gram lists, lemma lists, parts of speech lists).

5. Who is building the interface? Stan and Alberta folks, Stefan and McMaster folks. Stan oversees. (input to this group comes from uses and users, output goes to analytics) Next step: build the lightweight workbench to meet the current designs. Once we have the workbench itself up and running, we'll begin adding tools. In the meantime, the designers will continue looking at tools for collection browsing and feature displays.

6. Who is driving all of the above with, and checking it against, use cases? Catherine, Martin, Tanya, Sara, Ted, others as they flesh out their cases. Catherine oversees. Catherine needs to communicate requirements to the interface developers.

Note that what we're concerned with here is actual development work: overseeing this may not always be done by the chair of the relevant cell. The chair's responsibility is to make sure meetings are happening, work is getting done, and decisions are being made, but he or she may not always be the one doing the implementation. Also please note that I fully expect others beyond those named will make substantive contributions in these areas, and I recognize that there are other areas (like collaboration) that aren't mentioned above, but here I am trying to address those areas of the project where lack of forward motion is keeping other areas from advancing, and I am trying to name those who seem likely to have major responsibility for development, coding, specifications, implementation.