-
Notifications
You must be signed in to change notification settings - Fork 14
2018.11.07 Editors Meeting
Andrew Hankinson edited this page Dec 31, 2018
·
10 revisions
https://duraspace.zoom.us/my/fedora
- Andrew Woods (DuraSpace) - Regrets, Fedora/Samvera Camp
- Julian Morley (Stanford)
- Rosy Metz (Emory) - Regrets, vacation
- Simeon Warner (Cornell)*
- Andrew Hankinson (Bodleian Libraries)
- Neil Jefferies (Bodleian Libraries) - Regrets
(* indicates the notetaker)
- Getting to Beta
- Incorporating community feedback
- Which issues need resolution?
- Community feedback
- Would a process change be helpful in moving updates forward?
- Progress with OCFL article
- Moving Simeon's fixtures to OCFL GH Org
- How do we organize them?
- Getting to Beta
- Incorporating community feedback
- Agree to continue to channel all feedback through GitHub issues https://github.com/OCFL/spec/issues
- Happy with feedback so far. Will send out to personal contacts for more
- Which issues need resolution?
- Logical organisation of OCFL storage root
- Suggestion to go with a JSON file that has two keys defined:
uri
&description
, not sure about name, it should beocfl_
something, not sure abouthierarchy
, perhapsmap
--> @zimeon to write this on the ticket
- Community feedback
-
Choice of digest algorithms. We note that tools reading OCFL objects do not need to understand the digest algorithm used for content addressing in order to get any version. Understanding of the digest algorithm is needed for validation or write operations. Agreement of those on call to add
blake2b
(full length, no salt, hex encoded) to the table, change to "SHOULD use sha512" without any "MUST use sha256 and sha512" for content addressing. Will create new issue to discuss how we support extension beyond digests listed in the table --> @ahankinson will add issues for use of other digests for content addressing (and our proposal), and another for extension beyond digests listed in the table
-
Choice of digest algorithms. We note that tools reading OCFL objects do not need to understand the digest algorithm used for content addressing in order to get any version. Understanding of the digest algorithm is needed for validation or write operations. Agreement of those on call to add
- Incorporating community feedback
- @zimeon to update https://github.com/OCFL/spec/issues/249 based on discussion
- @ahankinson will add issue for use of other digests for content addressing (and our proposal)
- @ahankinson will add issue for extension beyond digests listed in the table