This wiki space is deprecated

This wiki space is now deprecated. The wiki for Samvera (formerly the Hydra Project) is now to be found here: Samvera

Skip to end of metadata
Go to start of metadata

Draft Scope & Objectives

We share approaches, work-in-progress, and finished code aimed at using external linked data in our applications. We coordinate development efforts where possible.

Current and past focal points of discussion / activity:

  • Linked Data Fragments server (ActiveTriples/ldf)
    • caching to allow for high availability URI resolution
      • configuration
      • expiration
    • querying to allow for data mirrors or remote downtime stop-gap strategies
  • Questioning Authority (projecthydra/questioning_authority)
    • querying linked data endpoints
    • integration with LDF (curationexperts/qa-ldf)
  • Indexing - keeping index updated in conjunction with ldf cache
  • Navigation / Cataloger UI - providing context above and beyond autocomplete
  • Linked data predicate range issues
  • Recommendations on which Linked Data vocabularies to use that should remain around.

Out of scope:

  • Long term preservation of Linked Data.

Meeting Times & Communication Channels

Time: Third Monday each month at 12:00 PM Pacific / 3:00 PM Eastern. Next Meeting June 19, 2017.

Call-In Info: Current meeting hangout link:

Backup conference line if the hangout isn't working: 712-775-7035 (Access Code: 960009).

Updates and announcements will be sent to hydra-communityhydra-tech, and islandora lists.


  • Anna Headleyfacilitator (CHF)
  • Chrissy Rissmeyer (UC Santa Barbara)
  • Ryan E. Johnson (UC San Diego)
  • Tom Johnson (Data Curation Experts)
  • Arwen Hutt (UC San Diego)
  • Zach Schoenberger (University of Alberta)
  • Eric O'Hanlon (Columbia University)
  • Julie Allinson (University of York)
  • Lynette Rayle (Cornell)
  • Bess Sadler (Data Curation Experts)
  • Huda Khan (Cornell)
  • Ryan Wick (Oregon Digital / Oregon State University)
  • Your Name Here (institution)


Meeting Notes


  • No labels


  1. For an "LDP Best Practices" group, the topics and agenda items are suspiciously light on any mention of LDP.

    In particular:

    • Caching of what? HTTP requirements for LDP servers that want to use caching infrastructures? Or using an LDP server as a cache for triples (but why?) ? Or just the question about how to cache triples in general, which is not specific to LDP but a general architectural/infrastructure question for the community.
    • Solr ... has what to do with LDP?
    • Linked Data Fragments ... has what to do with LDP?

    Basically, I have no idea whether I should be interested in the group or not as currently described. Any clarifications would be greatly appreciated.


  2. Rob Sanderson – I think the usefulness of the group's name here loses something when divorced from it's context. It's a subgroup of the Hydra Metadata Working Group. To that end, the name should probably be something more like "Metadata Management Best Practices with LDP."

    In response to your 3 specific questions:

    • Caching of triples that are outside of Fedora 4's scope – most notably controlled vocabulary entries and other related LDP RDF Sources containing triples that do not have a repository object as their subject. 
    • Solr and Indexing is extremely important from a metadata management perspective because bits of metadata from these non-Fedora resources need to end up in the User Interfaces that we build in front of Fedora 4, Marmota, LDP-server of your choice. Basically, the LDP conversations at Stanford last month verified that many of us have to look outside of Fedora to store metadata that we care about. We're now starting a discussion about how that might work.
    • Linked Data Fragments – see above, and the considerations coming out of Oregon that Linked Data Fragments & Triple Pattern Fragments might be a path toward some of these goals.

    Much of this comes out of the "Controlled Vocabs" session from LDCX:

    That's my (limited) understanding of the group's purpose. I'll let Steven Anderson & Karen Estlund weigh in if they have anything further to add.



  3. I agree that caching and indexing are important, but I fail to see where LDP comes in to that. Those topics are important regardless of the interaction protocol used between client and server.

    It seems more like:  Linked Data Management for Applications (and btw we're using LDP for some of that, but we're not talking about that particular bit)?


  4. I suspect we will end up talking about some patterns for the LDP bits, too, they're just not in the initial round of conversation, possibly because it's still somewhat of a moving target.

    But your point is well taken. The group is probably not very well named at present. I'll bring it up on one of our calls, at which I also suspect well flesh out our scope & be better able to respond to your questions.

  5. Rob Sanderson: I believe your usage of "Linked Data Management for Applications" could be correct in terms of the primary topic of what the group is focused on. So... up to you if that would still be of interest to join in on the conversation. 

    Otherwise, as Corey Harper said, hopefully can have more details posted and available after the initial call this Wednesday. Thanks for taking the time post!