Jason Kovari will start drafting cohort report... very very soon!
Agenda
All-hands meeting on Tuesday which will cover discovery and QA
See documents linked below
Huda will summarize report-out from focus groups in order to better see where there was most interest.
Want to ensure that everyone with input on discovery provides it, align plan with the shared goals
Lynette feels that there is good agreement around lookup work between Cornell and Stanford teams. Longer term open questions around approach to sustainability
Discovery
Huda successfully explore Solr Cloud index import
Tim focused on D&A call number browse this week
QA - see below
Next week
Huda will walk through proposal for rest-of-year work for LD4P3
PI group discussions: last 90 minute meeting is today
LD4 Conference highlights (in addition to those noted on 7/10, 7/17, 7/24 and 7/31)
None. Great times!
On Monday, the conference survey closes... we'll have much information from attendees about what worked, didn't, etc.
LD4P2 wrap up
Discovery: Much documentation completed. Still to do: Indexing Processes (DONE unless people suggest further info). Committing to doing demo intro/closing today to be done with it (Using wham! autosuggest video Tim sent earlier).
DONE. Huda did video and sent it to Lynette. May want to edit down a bit in the indexing section. Documentation was made into PDFs and pieces for WHAM was added to wiki.
Greg has been working on getting the 4 green buttons working. Has 3 working but can get the syllabus browse going. Huda is going to follow up John Skinner to understand that
DONE.
Greg has also looked at combining Solr additions with Blacklight Solr. Found out that is isn't trivial to combine data from aux standalone-Solr and the blacklight Solr-cloud, Suggestion from the net is to re-index -→ Lynette notes possible implication for container work, we will need to build an index for the right type of Lucene setup
Most can stay where-is but will try to migration one to see what it takes to do (export as JSON, import, reindex); process for querying the sources is time consuming but would do this if making this more production ready. If happy with these staying as stand-alone SOLR cores, can move around to other stand-alone servers but cannot implement into SOLR cloud deployment. Indexes are comparatively small (as compared to big Blacklight indexes) so maybe fine on stand-alones. If decide we want this all on enterprise level deployments, only options is to reindex source material... and need to maintain access. For next week: will do test export, reindex
Open Syllabus API: call now has version included... so need to make sure URL is correct. Will be moved into main LD4P2 branch
Simeon put in request to retag all AWS resources for LD4P3. Greg working on this now, which should be done today
WP2 - A key element of this work package is a sustainable solution that others can deploy. Questions of budget for deployment. Need to get all code into LD4P repository. What would a good end-product look like both for our maintenance and for others to use
Cache Containerization Plan - Had 2020-07-24 meeting including Dave, agreed on some revisions to this plan with a prioritization of containers
Lynette has started containerization process for QA service. Some tricky stuff with docker compose. Have web app coming up, still need to get mysql going. Would like people to choose which authorities go in a container but that is currently is code rather than external config – still need to work that out but that can be a next step
Open sustainability questions about who runs authorities where.
First PCC Sinopia profiles meeting this week, focus was logistics and scope. Looking at previous work on BSR (bibliographic standard record - monograph but also special collections records) and CSR (CONSER standard record - the one for serials) and what makes them PCC things. Agreement that the group will focus on monograph and serials
Also discussions of how to coordinate with Sinopia development work (coordinated via Nancy Lorimer who is a stakeholder in Sinopia dev meetings)
2020-08-07: Lynette worked on a different project this week... but: set up issues for two parts of containerization for QA that are not resolved: 1. how we will deal with passwords and other private data wrt: GitHub; Greg has some best practices for this. 2. Making connection to MYSQL... Lynette left off on this last Friday as was not working. Will make progress on this next week.
2020-08-07: asked Stanford whether they would test out containers as we complete. Received an affirmative. Dave and Lynette will also test each others
2020-08-14 Creating the basics was pretty easy to put together. But getting everything to work well together is not easy. Docker commands are long to type. So in an effort to speed the process of working with Docker, I put together a bunch of shortcuts and shared with others in repo elrayle/docker_shortcuts. Working with tutorials to narrow down why certain parts aren't working.
WP3 - Which parts of LD4P2 should we take forward? What is a good candidate for production? Are there new areas of user research?
ACTION: look at Blacklight and Cornell meeting notes to better understand what people liked... and what is near-to production.
Different strands of discussion (links to brainstorming, getting closer to production, and previous discussions) included in this document . Please comment/update
ACTION - Huda to reach out to CUL staff who expressed interest at March session for informal discussion – 2020-07-31 Huda has reached out to CUL staff
Three responded so far. No specific additional suggestions, except comment regarding shelf browse
ACTION - Discuss possible organization of brainstorming session with CUL staff on directions
what data sources are useful? what interactions do we want to facilitate?
suggestion to start with some big/open question - may be fruitful, may be quiet
have some of our ideas too
be ready to switch to examples from work so far and where we might take it
2020-08-07: consider doing this after we have more of a plan. Big ask to have people design something new from scratch... in their heads
ACTION - Work out plan for auto-suggest
auto-suggest work has progressed since March and would be worth showing to folks again - want to get this on D&A agenda
Need to be very clear of value proposition for our approach
Need to understand what would be require in order to get this into production
2020-08-07: bullet in closer to production document that goes thru what would be required to move this to a more production ready state; incl: index was populated by only some source data... would want all data incl. an update process; some design work possible; maybe knowledge panel at end to show what selected. Huda will add a lessons learned to the closer to production doc.
Can proceed with any of the four items in the above linked "closer to production" doc.
Should there be a partner deep-dive on WP3? Yes
ACTION - Huda to work with Michelle on setting this up (DONE as in followed up and "scan the waters" discussion to be included in next all hands August 18th). Half of the next all-hands meeting is on Q&A and other half on Discovery. All look at feedback document mentioned above.
Hurrah for D&A selection of Tim's visual browse from LD4P2 which will be implemented alongside the textual call number browse
LD4P3 demo blacklight site
Greg wonders whether worthwhile working on full CI/deploy
2020-08-07: decided worthwhile, just not triggered automatically whenever there is a push. will push when we want to deploy. System will otherwise be full deployment system. Working on this in coming week
Need to work out whether we want to resync our fork of BL with the current Cornell code – better to start with a new fork and then port in anything we want to keep
ACTION - Greg/Tim/Huda to design github branching strategy for LD4P3 work - 2020-07-31 Not done yet
2020-08-07: Discussions link. Capistrano setup like D&A. Jenkins job for deployment which will be manually triggered. Development off dev branch . LD4P2 code saved on LD4P2 branch, dev/master to reflect latest on CUL-IT upstream.
Note that various browse buttons on LD4P3 demo site do not point to the correct Solr indices (but perhaps we don't worry about this if we're starting from a fresh copy of the production code)
What does success look like? And then how do we get there?
2020-08-07: Held off as LTS sorts out some succession planning issues... coupled with the PI discussions that may yield some output that can inform CUL's documentation. More soon-ish