search

uchicago-library / library_website

The University of Chicago Library Website
5 stars 5 forks source link

Create IIIF Collections and Manifests for Soc Sci Map collection #251

Open dbietila opened 4 years ago

dbietila commented 4 years ago

Create necessary Resource Types in IIIF for the Social Scientists Map Chicago collection, include IIIF Collections and Manifests. The default browse order of the will be chronological by publication date, so there should be a IIIF Collection that reflects that.

There should also be a cluster browse organized by Subjects, which may require nested IIIF Collections.

johnjung commented 4 years ago

Currently, our metadata converter is looking for 264 tags with an indicator_1 set to 4, subfield c. However, I think this is incorrect. According to https://www.loc.gov/marc/bibliographic/bd264.html, I believe we should be looking for 264 tags with an indicator_2 set to 1, or a 260 subfield c. Please confirm if it's ok for me to update the metadata converter to reflect this.

Then, because publication dates are encoded in a variety of ways (e.g. "[1940?]", "[192-?]", "1943.", etc.) and this collection is so small, I think I should manually sort them for the default browse. Please confirm that this makes sense, or if updates to the MARC data are more appropriate, or something else.

dbietila commented 4 years ago

In this doc, https://docs.google.com/document/d/11QaNUMEtjp9DMkwkttvm1zMFF6rV1KrfVe2SYr1brVA/edit

it says dc:date [260/264 $c subfield from MARCXML (whichever is populated)]; edm:year [260/264 $c subfield from MARCXML (whichever is populated)];

I think the above is probably correct.

I suspect that you are seeing the indicator specs in this doc: https://docs.google.com/spreadsheets/d/1Kz1nfTSBjc2PTJ8hrZ--JCBpKV061sdXQxRxVo8VY_Y/edit?usp=sharing I will ask Thomas to take a look and see if we need to be concerned with the indicators for the maps collections.

johnjung commented 4 years ago

I'm looking at the code that does the transform itself, but thank you for pointing me to the spec you're working from. (I'm also working from the DC transformation here, which is a bit easier to sort than the EDM.) Actually, I see that I was looking at dcterms:dateCopyrighted, when dcterms:issued is much closer to the date from the EDM. I was looking for 264, indicator 2=1, subfield c...I updated that to remove the indicator. Please let me know what Thomas says about this, I can add the indicator back if necessary.