I. Discussion of Mark's schematron:

Mark started a schematron with approx half dozen assertions covering issues he knows of where AS has requirements beyond valid ead2002.
We discussed how to uncover additional requirements so that the schematron can be filled out to cover all cases.
3 places to check for these requirements:

1. ruby schema file (Nathan, can you provide the specific name)

2. test json object (Nathan can provide a test object they have been using for development)

3. The user interface itself (Brad may have more pointers on this)

We discussed how the schematron could be integrated into the ArchivesSpace import stream. On Nathan's suggestion, there was a consensus that this should be developed as a separate service from the actual ead import.

- Could still be one button from the user's experience (that is, check schematron and import if successful).

- In addition, the service should be set up so that it can be run separately against multiple EAD, in prep for subsequent ingest.

- It was suggested that the schematron service might (optionally?) fix some of the issues found.

Brad suggested that the schematron service be generalized (perhaps after EAD schematron integration is implemented) for use with MARCXML and others (Digital Objects?)
As far as next steps, I (Michael) suggested that maybe we start a document on the wiki to start collaborating on collecting issues to be baked into the schematron. Mark (I think) volunteered to start the doc with his findings to date.
In addition to the schematron itself, we should start planning the architecture for the new service. Nathan (I think) can start an arch./workflow document to kickstart this effort.
Brad suggested that everyone be involved in this effort before work on any other issues. Question, how can we have everyone involved?
After schematron, we can focus on marcxml, then ead3.

II. Updates from Members

Updates from Duke, Harvard,NYU, Ohio State, Yale, all in various stages of implementing AS, and doing import/export and plugin development. Michael had question about XSLT vs plugin (with whatever native ruby xml manipulation tools are available - is it nokigiri?). As Harvard is beginning a retrospective data normalization project on EAD, I'd like to continue this conversation as Harvard, AS, and other institutions work on tools for import/export.

Browser not supported