TechCall_17jul2017

Action Items
All - review BHL Egypt proposal and annotate before September 12Joel will write up documentation on segments in MacawJoel will add Mike to receive notifications on serverSusan to make a recommendation on whether to keep user-generated metadata and if so, for how long
Agenda
BHL Egypt proposal
https://docs.google.com/document/d/1nt133hQLk5eos700lMHCxzpDPc5NYZiQS-kagU32lAE/edit?usp=sharing

Segments
Intersection of EABL deliverables & BHL core priorities
Kristina's work in Macaw
Development in Admin Dash

Full Text
Mockup feedback
Grace's marketing plan

Notes
Segments
Coordinated BHL tech development, holistically. In other words, meshing across Macaw and Admin Dash.
BHL Tech Dev as one stream

Joel: Macaw framework in place for segments; Auto-complete suggestion tool, bits BHL API and suggests ones that might match
At this point, we could go to cataloging cttee, what we have now, what we want next, solution for ingesting for bulk article content, separate from Macaw
EABL and Australia both have segment metadata that would like to add

Lengthy document that Mike produced and Trish, Susan and Diane R commented extensively; Will share with Joel
Documentation on Macaw?

To make sure to user, terminology is the same

As part of CITEBANK, existing functionality was created, to be used from in Admin Dash
That tool is no longer really in use. But has been a starting point for requirements other work that's been going on in parallel, for bringing in content i.e., as csv as input.

Joel will write up Macaw documentation on segments

Susan - Rod Page invests a lot of time and programming in trying to avoid duplicates. He does significant amount of work in BioStor to eliminate dupes, as a result, it minimizes the number of dupes in BHL. BHL has it's own code and mechanism for handling dupes. Doesn't do a lot because BioSTOR is doing same work, and earlier, and doing well. A BHL Tool needs to weed out dupes as aggressively as BioStor does

Mike - not sure code exists. It's a good idea. Concern about programming time. Deduplication later, would help us get things out before EABL ends. Reached out to Rod and haven't heard back.

When we drop BioStor out of process, we need to be sure we're not dropping functionality.
Mike can't find that functionality. If there's code that can be adapted quicklyy and easily, let's do it. If from scratch, might be best as phased approach.

Top priority is loading segments via csv

For deduping, not looking for highly sophisticated. But if partner loads 200 articles on Mon, attempts same 3 days later, being able to catch that.

Documentation - get all in one place for all the pieces.

WikiCite - another external project looking at deduping functionality

Ultimate goal = one method to create and ingest articles; back-end tools will look the same and have the end result


BHL Egypt
How much do we want BHL Egypt focusing on user interface- how much value in that, vs metadata, OCR, gap searching etc?

5 year plan - these things we could put times on, could occur in year 5

How much would we want them to be involved?

We want involved with assistance as relevant

Joel to work with someone in Alexandria, on specification

don't need specifications to be ready until January 2018, timeline for development might not start until 2019
Would it be easier to come up with specifications and timelines for discrete products?
EOL is working as co-development; since 2007. Four active BibAlex developers on EOL v. 3

Take off Macaw multi-language support
BHL 2 would have all Macaw functionality rolled up

Multilingual built in from beginning would be best
Everyone annotate and create points or issues before September 12

BHL EC and Members to approve in principal and then have significant discussion in September.

Monitoring on server.
DBL server 2 , disk space, Joel and Mike to look into.
Joel will add Mike for notifications

Full text mock-ups
No comments at this time
Grace will be putting together a marketing plan. May have questions on us as it gets closer on how we might like to market it; ie popups vs banners etc.


Open discussion
User supplied metadata for articles. EC decided not to include user-generated metadata in article definition.
Mike had additional suggestions on form, but we'll need to change text on site. Not a rush. Martin will create a Gemini ticket

Do we want to continue saving all this information in database?
It's interesting to look at. Privacy for emails? No, privacy office reviewed.
But useful?
In short term, could be valuable in context of Gemini

If PDF only retrievable for a month, keep the metadata for the same amount of time?
Susan to make a recommendation and we'll discuss on a future call.