Page History

...

1.  Discussion of the upcoming releases.  Update to the Fedora ontology.  See fedora.info (removal of elements)
2.  There is a desire to have fixity around RDF sources comparable to checksum fixity on binaries.  What are the options?  Some contexts such as import and export were spoken about.   Some approaches:
   1. normalized serializations.  Pitfall - serializations may be different when using a triplestore even when there has been no significant change to the data.
   2. graph comparison algorithm approach
   • https://en.wikipedia.org/wiki/Graph_isomorphism_problem
   • http://blog.datagraph.org/2010/03/rdf-isomorphism
   • http://www.hpl.hp.com/techreports/2001/HPL-2001-293.pdf
   Need to better understand what we mean by 'same' with RDF; the desire is to have assurance that the underlying information is not changing unexpectedly. Some use cases:
   • import/export RDF resource
   • auditing accidental changes over time
   • conditional only update if strong etag
3.  Discussion of a couple of outstanding conversations related to the draft spec.  These can be revisited after draft has gone public as a platform for further public discussion.  Need to lockdown GoogleDocs after call.

4.  How do we intend to generate GroupId after refactoring.  Conversation to be carried on at a later time (future agenda item).

5. Reminder  Reminder about extended deadline for Open Repositories 2017 proposal deadline.  Query about current proposals and ideas for potential proposals (import/export, year-in-review).
6.  Item not covered in call.  Add to agenda for next call?
7.  Discussion of  FCREPO-2313.  Some more investigation is needed, but if there is a fix, the aim will be to get this into the 4.7.1 point release.  Depending on the outcome of the investigations and potential fixes, determine if delaying the release of 4.7.1 is required.