Time/Place
This meeting is a hybrid teleconference and IRC chat. Anyone is welcome to join...here's the info:
- Time: 11:00am Eastern Daylight Time US (UTC-4)
- Dial-in Number: (712) 775-7035
- Participant Code: 479307#
- International numbers: Conference Call Information
- Web Access: https://www.freeconferencecallhd.com/wp-content/themes/responsive/flashphone/flash-phone.php
- IRC:
- Join the #fcrepo chat room via Freenode Web IRC (enter a unique nick)
- Or point your IRC client to #fcrepo on irc.freenode.net
- ? add slack channel info here?
Attendees
- Danny Bernstein
- Bethany Seeger
- Jared Whiklo
- Peter Eichman
- Andrew Woods
- Yinlin Chen
- Esmé Cowles
- Doron Shalvi
- Kevin Ford
- -Randall Floyd-
- Carrick Rogers
- David Wilcox
- Ben Pennell
- Aaron Birkland
- Joe Harrington
Daniel Lamb
Agenda
- Welcome to new Fedora Committer: Yinlin Chen
- Alignment Sprint 1 Progress Report https://jira.duraspace.org/issues/?filter=14401
- Memento-Datetime header granularity issue RFC-1123
- The memento spec does not support version sub-second granularity
- There is no problem with us supporting it internally (when creating mementos without specifying the header).
- However is we support sub-second granularity it is not clear how we will be able to import versions without losing information.
- Challenges with acl:Append
- https://fcrepo.github.io/fcrepo-specification/#append-ldprs
- The spec defines rules at the HTTP method layer while the ACLs are implemented in the jcr layer.
- It makes it difficult to determine whether a POST, PUT or PATCH has occurred and whether a node existed when the transaction began.
- The javax.jcr.Node interface uses the same permsission (org.modeshape.jcr.SET_PROPERTY) for adding, updating and removing properties and generates the same:
- reference tickets:
- 5.7.3 LDP-NR - Patching Binaries (LDP-NR) - do we plan to support?
- Memento-Datetime header granularity issue RFC-1123
- Specification of External Content interaction
:Request: Link: <some-URI>; rel="http://fedora.info/definitions/fcrepo#ExternalContent"; type="image/jpeg"; handling=<copy|redirect|proxy> :Response: Content-Location: <some-URI> Content-Type: image/jpeg :Fedora advertising support: OPTIONS Accept-External-Content-Handling: <copy|redirect|proxy> :Comma separated list like Accept, if supporting multiple handling approaches:
- Feedback on Peter Eichman's writeup on the use of userAgent and groupAgent base URI's and their relationship to WebAC in order to clarify whether or not what if anything needs improvement/clarification/alignment
Sprint tickets
Ticket Summaries
Please squash a bug!
Tickets resolved this week:
Tickets created this week:
Minutes
Separate tickets
- Creating Mementos when not providing Memento-Datetime
- no body
- Converting dynamic properties to static properties
Other
- ACTION: BenP to submit an initial, minimal PR for creating no-body LDPRv's
Updates from spec editors meeting
- Discussion of handling external-content
2.a.
Two options
Reduce the granularity of our label
Modify Memento spec
Change the current implementation to not have label in microsecond but just second.
escowles:
409 Conflict seems reasonable to me
agree that server-managed or distributed backends could have trouble with that
Bethany Seeger & awoods:
Memento Tools - Validator: http://mementoweb.org/tools/validator/
whikloj:
https://jira.duraspace.org/browse/FCREPO-2727 & https://jira.duraspace.org/browse/FCREPO-2728
2.c.
Change the Fedora spec. Change to May instead of a Must. [Spec 5.7.1]
whikloj: Section 5: "To configure access control restrictions, implementations MUST follow the recommendations of Web Access Control [SOLIDWEBAC] with the following additional requirements:"
Bethany Seeger: Can you read this the above this way “If you choose to configure access control restrictions, you MUST follow Solid Webac”? (edited)
awoods:https://fcrepo.github.io/fcrepo-specification/#append-ldprs
Modeshape doesn’t give delete and add property permission, only set property. This problem is because authz is being enforced at the Modeshape layer, and one logical API operation translates into multiple modeshape operations.
Option 1: using injection. like injecting it in the RDF diffing code, so you could e.g., look for deleting triples and throw an exception if the user only had append permission. But could break the design also need to think about performance impact @whikloj.
@Peter Eichman suggests moving the ACL check out of Modeshape and into the HTTP layer or at the very least outside of modeshape.
Find a feasible way to get out from underneath Modeshape the webac code has lots of jcr in it.
3. external content pr: https://github.com/fcrepo/fcrepo-specification/pull/343
Use POST/PUT requests to create external content nodes.
It looks great on the current proposal in 3.a
Action Items
- Danny Bernstein determine whether there is any case where a PUT on a TimeMap would make sense. If there is it should be raised with the specification editors.
- Peter Eichman to do a short writeup on the use of userAgent and groupAgent base URI's and their relationship to WebAC in order to clarify whether or not what if anything needs improvement/clarification/alignment. ACL Agents - Strings vs. URIs