Present: John U., Stan, Matt, Martin, Stefan, Amit

Agenda:

Which institutions will actually have spent down all funds billed to this grant by the end of December?

Let John U. by December 15th.

Interface report:

On hiatus for documentation for the first two weeks of December. M5 release has brought a lot of new bugs.

News flash: Search by example is actually clustering classification; we're not actually using a search by example. If you have only one category, you can get rankings within the category. Middleware has calls that would support true search by example, using TFIDF, but this is different from getting the Naive Bayes stuff working right.

Dunnings is now inside the workbench, as is the concordance view, in M5; that should be available through the wiki by this afternoon. The database is 1.07 (not including the newest data); new database will be up in early January. Martin et al. will tinker with Dunnings and concordancing.

Documentation:

Amit's been documenting code and calls. Andrew is setting up a sample javascript component for documenting interface components; he's also working on a component creation guide. Carl has created some help documentation. Catherine's been working on help text: https://apps.lis.uiuc.edu/wiki/display/MONK/Wiki+version+of+help+text

JMU will start pulling things together for the web site in the last two weeks of December.

Production server for the library @ UIUC:

Quote on Tuesday of last week, Thursday the quote was up by $2K. Library wants Dell: we're looking at 2 2.7 GhZ quad core, 64GB of RAM, 2 or 3 terabytes. The idea is for this to be the public MONK site at the end of the project.

Future MONK-related proposals?

Martin's still interested in the Book of English proposal idea: a large datastore or an aggregation of datastores. This is related to working with Google Books, OCA (UIUC Triple-Deckers), Hathi trust materials, is another project, probably tied to an NCSA computational center; this could also include a shared development environment.

Philologic: We have better data, they have better interface for searching. Should we propose something that brings these things together? Or should we have interoperability only at the level of the datasets? Should the project here be to allow side-by-side comparison of different tools with common datasets?

Editing environment tied in to a more robust, user-friendly ingest workflow, is another project. (A good fit for Bamboo?)

Workshops: January workshop for working with SEASR/MONK. Future applications for NEH workshops?