Applied Linked Data Call 2015-10-01
Attendees:
- Arwen Hutt (UC San Diego)
Linked Data Fragments Update:
- Trey had plugged it into the talk he had at Hydra Connect and it worked.
- Viability?
- Discussion over whether Linked Data Fragments is actually a spec.
- Viability of triple fragment pattern.
- Advantage over LDP: Thing you are requesting can come from multiple datasets.
- SPARQL endpoint would be weird that it would change the query so that it does crosswalks for you. Could build this into the fragment implementation spec.
- Expose a subset of what you can do over SPARQL and limits what the user can request. Doesn't just give them an open text box that allows for such complex requests that it crashes your service.
- Advantage over LDP: Thing you are requesting can come from multiple datasets.
- ActiveTriples integration?
- Done for Trey's talk but needs to make it pluggable. But ActiveTriples 0.8 changes the interface and 0.8 is "broken" so can't be done at the moment.
Is Long term preservation in scope (Hydra Connect interest) or still out of scope:
- Preservation of predicates? They could disappear as well.
- UCSD create local definitions of predicates?
- Most things have not been tied into the external uris.
- Most things have not been tied into the external uris.
- UCSD create local definitions of predicates?
- What do we do about predicates? What do I do about objects? How deep does one go in the preservations?
- Problem: lack of available bandwith
- We don't have the bandwith to take on this additional thing. We highly recommend that someone does this!
- As such, this topic is likely best suited for another group.
- We don't have the bandwith to take on this additional thing. We highly recommend that someone does this!
Next Official Meeting:
- Next meeting will be October 15th at 9:00 AM PST / Noon EST.