search

ycba-cia / blacklight-collections2

5 stars 2 forks source link

Online records for artists and sitters #316

Open edgartdata opened 3 years ago

edgartdata commented 3 years ago

In late July Eric and I talked with Martina about publishing YCBA records for artists and sitters to our online collections catalog on the model of what the Tate and MoMA do, including links to Wikipedia:

Can we think of what metadata elements we would want/be able to vend to our online catalog?

yulgit1 commented 3 years ago

gender nationality ethnicity life dates links to authority records/wikimedia/official websites links to works, exhibits within YCBA (beyond?) YCBA biography
related artists genre tags

edgartdata commented 3 years ago

This is related to https://github.com/ycba-cia/blacklight-collections2/issues/317

Today 8/6/21 the group talked about what would be included in YCBA online collections catalog webpages for artist, thinking about what the Tate and MoMA do (suggested by Martina): https://www.tate.org.uk/art/artists/zanele-muholi-18872 and https://www.moma.org/artists/38?locale=en . Both Tate and MoMA use external authorities (Getty and Wikipedia) but the group felt that we might just want to point to these sources rather than print their content in our webpages their content (that way the content stays fresh without YCBA having to do anything). We would want to consider other external authorities such as the British Museum and we would want to see if we could reuse biographies published by the British Museum. We would want to let the Getty know about us reusing their bios and artists records too. We also want to continue talking about what would make worthwhile a YCBA campaign to write artists biographies so that we can publish our own YCBA written bios.

yulgit1 commented 3 years ago

Also perhaps part of YCBA campaign identifying a representative image for the artist if available, whether it's a head shot, portrait, self-portrait, or image of a work, etc.

edgartdata commented 3 years ago

Update from Martina: CJM likes the concept of artists webpages in the online collections catalog.

yulgit1 commented 3 years ago

Preliminary thoughts to consider moving forward.

(1) what do we ideally want the collections page to look like when including artists/sitters (2) understanding constraints in blacklight that may affect what we can do with (1) (3) sourcing and making available the data driving the artist pages (direct TMS queries, scraping LIDO, solr indexing, using LUX as source, and/or external sources, etc.)

edgartdata commented 3 years ago

Ed and Martina would like us to consider these additional data points to artists pages:

Artists to start with as examples" hewe locke yinka shonibare: British Nigerian brunias tilly kettle joseph wright of derby joscelyn gardner – barbadian vs. Barbados: impact search In general: living artists – look at how they describe themselves on their own website – manual process

Consider this authority: RKD

edgartdata commented 3 years ago

Eric demonstrated an early version of the artist page but this is not a priority. Universal Viewer is the priority for now. Question to think about for the group: what might the YCBA online catalog search look like to accommodate searching on artists?

edgartdata commented 3 years ago

@yulgit1 Eric, something to keep in mind as we develop online records for artists and sitters: let's not publish their nationalities but instead their countries of origin/place of birth, places of activity, visit/tour, and death.

yulgit1 commented 1 year ago

Preliminary thoughts.

Project management: Maybe have a design team (what do we want BL to look like, search, facets, item page), and a content team (metadata issues).

I think this can be TMS driven, using constituent and related tables, in combination with altnums and text entries. One first step might be to create altnums to LUX person pages. In that way we could initially index a link from blacklight to our LUX artists: (although moving from BL to LUX might be a little jarring as far as UI, but at least the info is available, something subject to discussion). Nexts steps would be to scrape the LUX json, use TMS info and keep the artist page within the BL UI environment.

Also to consider: ruby 2 support drops 12/25/2023, and upgrading to ruby 3 will be a big endeavor. It would probably make sense to do the upgrade first, so we're not coding all of this into a hole that will be hard to upgrade.

edgartdata commented 10 months ago

@yulgit1 @flapka Revisiting this project after many months. Here are the data points coming from TMS and Voyager (?) I would like to see included in YCBA online records for artists represented in the YCBA collections below. The data for now should only come from TMS and Voyager.

Mention where the information is coming from

Reach out to Jeff at LUX on how to use LUX's API.

NyxeeNight commented 9 months ago

From: Sanderson, Robert robert.sanderson@yale.edu Sent: Wednesday, February 14, 2024 10:14 AM To: Delmas-Glass, Emmanuelle emmanuelle.delmas-glass@yale.edu; Campbell, Jeffrey jeffrey.campbell@yale.edu; Prown, Sarah sarah.prown@yale.edu Cc: James, Eric eric.james@yale.edu; Wickell, Carly carly.wickell@yale.edu Subject: Re: LUX's API for YCBA

Hi Emmanuelle, Eric,

I think you have everything that you need in order to do this today?

Taking Turner as an example…

From your record you need to know your URI for Turner: https://ycba-lux.s3.amazonaws.com/v3/person/d6/d6e4ba54-eeb7-4b54-b6dd-cd60b3027f89.json

Then you can do a search against the LUX API for it people or organizations with that URI as an identifier: https://lux.collections.yale.edu/api/search?q=%7B%22identifier%22%3A%22https%3A%2F%2Fycba-lux.s3.amazonaws.com%2Fv3%2Fperson%2Fd6%2Fd6e4ba54-eeb7-4b54-b6dd-cd60b3027f89.json%22%7D&scope=agent&page=1

(Note that we’re about to make this URI simpler by putting the scope of agent into the URL, rather than the query, and page is optional)

Then in orderedItems, there’s a single record, which has an id field, which is the URI for Turner in LUX.

You would then retrieve that record: https://lux.collections.yale.edu/data/person/f778f2f8-6b04-44af-8bef-cbfb8eccdc6f

And extract from the JSON whatever fields you would like to render.

Rob

yulgit1 commented 9 months ago

Spec's for this project using fields outlined above (TMS for now)

  1. Use LUX transform LMDB database and filter for actors
  2. Script this pipeline in python on the LUX transform server
  3. Use LUX YCBA contributing records for the following
    • reconciled artists names
    • gender
    • life dates
  4. Use LUX API to get LUX json from YCBA contributing records
  5. Use LUX json for the following
    • place of birth
    • place(s) of activity
    • place(s) lived in
    • places of visit
    • place of death
  6. Make python SQL connection to TMS to get:
    • objects made by the constituent
  7. Make python connection to SOLR and create SOLR record using (3),(5), and (6)
  8. BL reconfigure search categories display LIDO, MARC, Artist recordtype_ss
  9. Determine which fields to display in search results
  10. Configure BL for search results and item page from solr
  11. modify BL to display objects by artist with links