2015-11-19 Meeting notes
- Jessica Wesley Meyerson
- Jennifer R Hecker
Date
Attendees
Agenda
- Housekeeping
- EAC-CPF Virtual Chat with Ellen Doon and Mark Custer from the Beinecke
- Questions for Ellen and Mark:
- Lessons learned from the 2012 Connecting the Dots collaboration with Harvard· Time-action study for EAC-CPF, impact on workflow for staff· Measurable benefit to users for EAC-CPF creation
- Current implementation strategy (JH)
- Settling in on a master list of relationTypes (and other controlled vocabs that would have some institutional variation based on collection strength but hopefully can work towards a community standard CV for things like relationType) with associated URIs
- pain points
- solutions(?!),
- what-not-to-dos
- Integration between DAMS and AS
- Drawing from data in your EAC-CPF records in order to populate your local names controlled vocabulary in the Beinecke DAMS (Preservica?)
- SNAC SNACK time
- Questions for Ellen and Mark:
Discussion items
| Item | Who | Notes |
|---|---|---|
| Housekeeping/check-in | ||
| Virtual chat: EAC-CPF with Mark Custer and Ellen Doon of the Beinecke Rare Books and Manuscripts Library | Introductions:
Decided to do the project in a non-SNAC way
What is to be gained? Open question because both insitutions lack the IT support to put up a stand-alone interface for the EAC-CPF records. There was prototype developed for the pilot but once the funding dissappeared REview of the pilot: 2 interns working full time - they were documenting decisions, what decsions were and why Summary of the time study - 4 hours per record with the bulk of the time on the research and writing Least amount of time spent on control reocrd Next less time intensive was description Where to put certain parts of information and because they didn't know how they would use the records it was hard to make some of those decision those they just entered the same data into multiple fileds. How to put, how to encode it. It wasn;t just typing Adding relations was suprisingly time consuming - they were using Oxygen for these records (had no dropdown lists no controlled vocabs stored) You had to keep track of what you entered in each record. Mark came on the project the second year - he was looking at the relationship vocabularies - he looked at all the records and he indexed those in a xml database to see all the lements that were actually being used in most of the records. (BaseX) - gives an overview of the data. No way to put them online but still wanted to analyze the data. Next step for Mark was to recreate controlled vocabs of the relationship type - at the end of the first year > they felt like the least satisfying part were the generic relations so they started creating more nuanced descriptions of relationships > creating as a few They had a hierarchy of sources: bibliographies, the collection material itself - in the beginning they set down and determined names and then they looked those folks up through secondary resources.
What is the most important relationship between two people - Susan and Melanie at Harvard we could check in - they oversaw the intern. They didn't start with controlled access terms (people) created by the archivist - they started just knowing the names. Some of the n' They also created EAC-CPF for collectors of the records because they had a huge influence on.. Mark - talking about archivesspace and dams - AS will have a connection to the SNAC database so that they can pull in data from that so they can enhance existing records and use info that already exists:
How anyone updates those authority files - who knows they are looking for that two-way communication as well. Updating records is likely better done in the national database so they can download those. They wouldn't ever hand-encode again when there is at least a skeleton of one out there in the world or a mechanism to do it. They could have really strong local content - create a dynamix subject guide that would use EAC-CPF to increase understNDING > AN ENTRY POINT into the collections based on social networks. Also just providing it as a dataset. The other thing about AS - it is still early in its interpreation and adoption of EAC-CPF > Mark already tested the import into AS but it didn't work cause the data model for EAC-CPF is being tweaked. He tested again about a month ago, and one new development is that when you create a relationship reciprocal realtionships automatically (which is not always appropriate) it depends on the relationshipTYpes Other thing that doesnt work - Yale haas different descriptive notes based ont he relationship type (currently in AS, you have to a note that works for both sides of the reciprical relationship). They did have some unidirectional relationships - they were for the collectors in most cases (relationship the papers and non to each other, but they Advisors on the project were Kathy Wisser - lots of conversation about relaitonships (like celebrities they don't know - you could have a relaitonship where it's important on one end but not on the other) If you are knighted by someone you might feel that you have a personal relationship with the soveregn that knighted you but they knught. When you implement archivespace as a public discovery system - new questions:
style and formatting for the records that are currently in SNAC EAC-CPF exists as a standard decoupled from SNAC but it is defined for most people as whatever happens with the SNAC project: neither at Harvard or Yale have the tech support to build a new system for creating these records. Mark feels that the biggest thing is - 2 big parts: you have narrative description of someone which matches with - the relationships though don't fit as well in XML - the relationships should be stored in RDF because everything has to post processed out fo XML frst. No consistency about standards for EAC-CPF but because we have systems driving us in one directions and endless possibilities with hand-encoding - there is a need for best any community effort for best practices guide - kathy wisser's research into what elements are in biographical and historical statements > needs to build from that for best practices. They didn't do any structured geneological stuff for people cause people don't have family trees, families do. Mark - he has a colleague that refers to these records as "concept records", unless they are being used by researchers or being used to manage collections internally (like a finding aid) it is hard t justify doing. There has to be a closer connection to researchers. Best practices conversations should include researchers themselves - because otherwise we are limiting ourselves. There is geo data, subject data, etc. - a rich data set. Hopefully researchers can tell us how to take full advantage.
|
Welcome to the University Wiki Service! Please use your IID (yourEID@eid.utexas.edu) when prompted for your email address during login or click here to enter your EID. If you are experiencing any issues loading content on pages, please try these steps to clear your browser cache.