2015 11 06 Chapel Hill NC

Notes

• Will keep to every six weeks
• Peter May is considering taking Marcus’ place on the Board
• Betsy has been administrative co-chair for four years, and Tom has been technical co-chair for that same period of time, and they are both asking for people to consider replacing them as co-chairs
• Thanks to Frederick for securing the room for the meeting
• 2010 - Reimagining METS paper
• METS Light
• Alignment with the PREMIS Community
• Workshop in London
• Decided to do it again in Chapel Hill
• Will likely be back at DLF next year

Workshop on Monday

• Tom talked about METS 1.x and the data model derived from it - also the recent changes
• Bertrand did METS RDF - METS/METS Light - METS and PREMIS
• Nancy talked about RAMLET
• Andreas talked about pros and cons of alignment with other models
• Breakout sessions - 30-40 people; lively conversations
• METS RDF not completely well received
• METS Light came up again as a need
• Don’t really need something new - what is the problem 2.0 is trying to solve
• A METS generic profile would be the base for a tool set and a viewer This is being done in Sweden - one base profile; models that are institute-specific
• We could move conversations back to profiles from the 2.0 direction
• A useful task would also be to create a data dictionary such as PREMIS - we don’t have definitions
• What is the coverage of the profiles - should we compile use cases that don’t have profiles - need to gather documents for those that don’t have (and never will have) profiles
• Need a stylesheet to render 2.0 profiles - this has been a to do for a while and we need to make it a priority
• Should we coordinate with Europeana and DPLA?
• Every profile has at least one example
• fileSec and structMap - fileSec needs to be flat - just a listing of all of the files connected to technical metadata
• If we don’t go to a METS 2.0 then do we need to talk about a METS Light - in London the METS Light conversation was in context of a more complicated METS 2.0
• PREMIS in METS need to be revised since PREMIS 3 - METS 2 could be streamlined with PREMIS
• Could reduce complications
• Eliminate redundancies
• With the any attribute - could start putting something like DC data in the divs of the structMap
• Maybe we need to be working on an RDF serialization for METS 1.x - but need a data dictionary/data model
• Primer and comments in schema act somewhat as a data dictionary

PREMIS and METS - what do we want to do together? Point of contact is the guidelines Doing a METS profile that formalizes the guidelines - is this the generic profile for utilizing in a preservation environment; Karin and Evelyn are involved in both groups, so there is overlap that could be useful to see what needs to be revised - new thinking since 2011 - they will bring to both boards Frederick - do we want to coordinate METS with schema.org - need use cases - maybe it’s just the descriptive metadata in which case it should be with the descriptive metadata schema

Subclassing - there was a preference from the meeting on Monday to not subclass, but this could make it more difficult to use different schemas within the METS

Action Items XSLT to render 2.0 profiles

1.0 Model Data Dictionary Generic profiles METS Light Education Rebecca’s stuff What else Primer Highlight Karin Technical documentation for XML schema - update it (or is this too much?) Or Tag Library like EAD and EAC; could use the TEI transform that Karin uses

Start thinking about METS Light as starting with a data model - that then develop a profile perhaps and it becomes place to serialize RDF

Possible path Don’t say METS Lite Don’t say METS 2 Model based on METS 1.x Separate working group for METS RDF - Board would work on METS 1.x Work that has been up until now could be a base to start this work

Joint meeting METS/PREMIS to also talk with PCDM representative

Karin: I talked about this overview which is an easy introduction but may need an update: Overview