dataset_notes.md

Format

### [Name](URL)

- Type: (CSV, JSON, etc)
- Object URLS: (True/ False)
- Image URLS: (True/ False)
- Notes: 

Tate Gallery

• Type: CSV
• Object URLS: True
• Image URLS: True (with processing)
• Notes:
  • 10 years out of date
  • High quality image urls are not included by default, but thumbnail URLs are (but they're slightly wrong). Here's what to do:
    • Thumbnail URLs included are formatted like: http://www.tate.org.uk/art/images/work/D/D13/D13283_8.jpg
    • However they need to be: http://media.tate.org.uk/art/images/work/D/D13/D13283_8.jpg
    • And then to get the high quality image: http://media.tate.org.uk/art/images/work/D/D13/D13283_10.jpg
  • Also note that their repo contains a nice list of things people have built with the data

Cooper Hewitt, Smithsonian Design Museum

• Type: JSON, but spread across lots of files
• Object URLS:
• Image URLS:
• Notes:
  • Its another strange spread across lots of JSON files. Need to come back to this one later.

The National Gallery of Art (DC)

• Type: CSV
• Object URLS: False
• Image URLS: False
• Notes:
  • No URLS at all in the dataset, but I'm sure I can find a way to get them.
  • They have an API: https://api.nga.gov/

Minneapolis Institute of Art

• Type: JSON, but spread across lots of files

The Metropolitan Museum of Art

• Type: CSV
• Object URLS: True
• Image URLS:
• Notes:
  • Doesn't include image URLs in the dataset... I'm sure I can find a way to get them (legally lol), just move on for now and come back later! Get the 90% solution first!
  • Note: there's an API: https://metmuseum.github.io/