Huda Khan to follow up on Vitro/VitroLib security issue that Randy asked about
2018-08-17 Test was done, Huda has yet to complete analysis. Will post results as information for the VIVO community but not do any work to address.
2018-08-24 Not yet. Have to read report to see if issues are detailed. Will report back to Harvard and VIVO community. Will attempt to do by 8.31
2018-08-31 Putting Vitro behind SSL did address some issues (e.g. login not secure) but not all. There remain cross-site scripting and SQL injection issues. Will forward document to Randy, and will report on the VIVO developer call.
Simeon Warner to work with Mary Beth to set up an LD4/D&A discussion. Should include Melissa & Frances. Frances has done something in an innovation week. Goal is to understand areas of work that might happen in LD4 that would align with existing hopes
2018-08-02 Have sent email to MB, Simeon will follow up again
2018-08-24 no status. Steven Folsom : this is his largest work area; can help coordinate if needed
2018-08-27 MB has suggested Tue Sept 18 at 11am for 30mins. Proposed 5-10mins intro and 20-25mins discussion
Steven Folsom will create a repo for the Sinatra 45s work item... and add the JSON file there; we'll also post app profiles, etc there. IF we prioritize the QA extension work, we'll create a repo/folder for that work and this can go into there...
samvera/questioning authority (QA) is preparing for release pending merge of CORS header support and Rails 5.1 support. LD4P/qa_server uses the min_context branch of QA and was updated to include everything that will be in the new release of QA.
LD4P/qa_server was updated to be an engine instead of an app. Remaining work is to bring the history work into the new engine. It's in a branch now, but needs testing to be sure it is working as expected.
cul-it/qa_server was a fork of ld4l-labs/qa_server. It has been detached and is now stand alone. Remaining work is to convert it to be a simple app that gets all its functionality from the LD4P/qa_server engine.
vocabulary support: There is a list of vocabularies that includes 8 vocabularies we already support. 3 substantial vocabularies that we should work on supporting. And 32 smaller vocabularies from id.loc.gov/vocabulary. I believe that all/most/some of these are currently filled using id.loc.gov/suggest API for autocomplete. Should these 32 smaller vocabularies be supported through QA or remain using suggest API? Are there UI considerations?
Enhanced Discovery
Planning Tue 2018-09-19 meeting with D&A, 11am. Who should/can attend?
Need discussion with Stanford and to have some brainstorming – if we add extra connections in out data (e.g. a wikipedia link on a catalog record, how will that show up?)
When is this being set up? Who is responsible for planning?
Tom mentioned the same issue that we are having; Stanford Blacklight devs are not the same as LD4P devs... similar to Cornell. Both institutions need local discussions
During monthly all-hands call, Tom mentioned a breakdown of work, which implied distribution between Cornell and Stanford; can we confirm? We heard him say they are working on schema and knowledge panel... and that Cornell should work on semantic browsing.
How much of this will be put into production in local discovery environment?
this will be determined based on discussion with D&A;
QA can already facilitate some of these enhanced discovery components (e.g.: pulling into data from DBPedia, etc.)
We should begin from a position that we WILL implement in production.
Question of source data for production discovery environment... will we be adding URIs to MARC data for this specifically?
Question of what is contextual data here... and how to facilitate that more broadly than individual entities in DBPedia linked from MARC records / RDF, etc...
Question of what is added to the source data? what is indexed in SOLR? what is added on-the-fly?
What resource implications need to be identified on the 'general' D&A side to ensure that work in LD4P2 is leveraged?
What is the tangible discovery output? Is there a centralized discovery environment of _____ data from the grant? Is it a GH repo with some code? What is promised?
Prep for Cataloging Sinatra and other 45's (Discogs data)
Application Profiles
In mid-to-late September, Steven, Tracey and Jason will start planning for Sinatra BFE profile. Hip Hop SHACL was for 78s... Sinatra will be 45s... HipHop used bibliotek-o.
Question: whether we want to help out with Art & Rare Materials BF Ontology Extension profile to ensure that others are facilitated...?
Discogs data / API
Tim pulled data from Discogs
Steven assessed data. Some can easily convert some into local resources while others need reconciliation. Some data will be conversion of literals to resources
Issue: importance of distinguishing between Release and Master in Discogs
Want API to feed form that can be supervised into conversion... cataloger has API that searches Discogs for a 45. Once selected, data are then put into a form but no RDF yet... since we might then duplicate resources.
Can we identify resources for the intermediate hand/auto reconciliation steps post-JSON & pre-RDF.
OCLC' LD prototype has a tool that searches VIAF, ISNI and one other. Select one and it brings into editing environment and you then claim bits of the conversion and hit save. Diff = only using RDF sources... so creating redundant RDF in different environment rather than non-RDF to RDF
Building this out would be really beneficial for community...
Could be integrated into BFE OR a separate form/app
How do you go from pre-populated field to a QA look-up? What do we show them? Do we do a pre-look-up with a suggested list?
Piece that may not be in BFE – you have a way to create or edit a record... but nothing where you go out and search for things you can pull in. This includes a lack of connection to OCLC or other datasets. Astrid and Christina are discussing / looking into this. No search box to look up things that already exist... go look them up and pull them in. This could all be the same functionality (OCLC OR Discogs)... where you look up & pull in... but nothing saves or writes RDF until you hit save.
How does QA fit into this? There are two components ... how to retrieve the data and how to process the data.
would need to add a mapping module into QA as a post-process.
from the processing side of QA, inside that does a term fetch; get back JSON ... with mapping module that then translates data from JSON into another format
QA the server... not QA the Gem supported by Samvera community
Action item: discuss during 8/31 mtg whether this could/should be prioritized for work
Action item: watch the space of importing external Works, etc., data into BFE – to see if this links in
Travel and meetings – what meetings should be be present at between now and end June 2019?
DCMI (Jason, not on LD4P2 funds), SWIB (Steven, Huda, Lynette, Simeon), CNI (Dean, not on LD4P2 funds), US2TS (March 11-12, Duke), LODLAM (Steven), ALA (Harvard doing PCC liaison)
Jason, Steven, Lynette, Huda, Tim, Dean, Simeon is 7 people, budget is $1500 each – all go ahead provided within budget, pre-travel to dept head and Simeon