Skip to content

Latest commit

 

History

History
112 lines (90 loc) · 6.47 KB

File metadata and controls

112 lines (90 loc) · 6.47 KB

Data Dictionary — Drawing from the Crowd (Smapshot import datasets)

This dataset contains curated metadata for Japanese landscape prints and illustrated books (meisho-zue), prepared for georeferencing on the Smapshot platform. No image files are included. Image columns are URLs that point back to the holding institutions (NDL, ADEAC via Japan Search, the British Museum, the Met, the National Archives of Japan), which retain all rights to the images themselves. The standardization, enrichment, and geographic tagging are the project's own contribution.

Files

File Rows Role
bm_ndl_meisho_landmarks.csv 1,345 Print/book records (British Museum + NDL meisho-zue + National Archives of Japan)
jmet_ndl_taito.csv 1,757 Print/book records (NDL via Japan Search + ADEAC) , Taito Lifelong Learning center
artists.csv 196 Artist lookup table

Key relationship: photographer_id in the two record files is a foreign key into artists.id. (The Smapshot template reuses the field name photographer_id for the artist of the print.)

File format note

These follow the Smapshot import template, which has a banded header:

  • Row 1 — a category band (Metadata displayed, Used for georeferencing, etc.)
  • Row 2 — the real field names listed below
  • Row 3 — a human description of each field
  • Row 4 onward — data When loading programmatically, use row 2 as the header and skip row 3.

Record files: bm_ndl_meisho_landmarks.csv and jmet_ndl_taito.csv

Both share the same columns.

Field Meaning Notes / known issues
(first column) Template label column Empty in data rows; a template artifact, safe to ignore or drop
download_original_image_link URL to the original image at the source institution Pointer only — not an embedded image
link_id Identifier linking metadata to the image file
title Title of the image BM/NDL: 100% filled. JMET/Taito: 2 rows blank
caption Description of the image BM/NDL: 100%. JMET/Taito: ~93% filled
original_id Identifier linking the record back to the source catalogue BM/NDL: 52 rows blank; 796 rows share duplicated IDs, reflecting multi-page meisho-zue book objects (incl. the ~181 Miyako meisho zue 都名所図会 rows) where many views derive from one catalogued object. JMET/Taito: all present and unique.
date_shot Exact date (dd/mm/yyyy) if known Empty by design — exact dates are not knowable for these works; date ranges are used instead
date_min Start of date range 100% filled
date_max End of date range 100% filled
license Image rights + source attribution + project's curation statement 100% filled; verbatim per source institution
photographer_id Artist of the print → artists.id BM/NDL: 31 blank; non-blank values resolve. JMET/Taito: 11 blank; a few Utagawa-lineage values are pending addition to the lookup
download_link Optional user-facing download URL
link Optional deeplink to the image info page
shop_link Optional purchase link Empty in both — not applicable to these sources
date_orig Original free-text date (e.g. "19th century") BM/NDL: ~95%. JMET/Taito: ~47% filled
longitude A priori viewpoint longitude 100% filled
latitude A priori viewpoint latitude 100% filled; 25 JMET/Taito records fall outside Japan and are flagged in_japan_bbox: false in dftc_points.geojson
view_type One of: terrestrial / highOblique / lowOblique / nadir BM/NDL: lowOblique (1,126) + terrestrial (219). JMET/Taito: all lowOblique

Lookup file: artists.csv

Field Meaning Notes / known issues
how it appears in ARC file Source-form name string (kanji + romanized variants) 192/196 filled
id Stable artist identifier — FK target for photographer_id 196/196; primary key
first_name Family name (kanji + romaji) 193/196
last_name Given/art name (kanji + romaji) 191/196
link Authority link (mostly Wikidata) 175/196
company Workshop/school (sparsely used) 3/196 filled
start Birth / active-from date (ISO 8601) 169/196
end Death / active-to date (ISO 8601) 168/196

Stub entries: 32 artist rows are incomplete (missing dates and/or name fields) and are marked provisional.


Schema crosswalk (Dublin Core)

A mapping of dataset fields to Dublin Core Metadata Initiative (DCMI) terms, to support interoperability and harvesting. Terms use the dcterms: namespace.

Record files

Dataset field DCMI term Notes
title dcterms:title
caption dcterms:description
photographer_id dcterms:creator resolved via artists.csv; artist link (Wikidata) serves as a creator authority identifier
date_min / date_max dcterms:created date range of the work
date_orig dcterms:created original free-text date
license dcterms:license, dcterms:rights per-record image rights and attribution
download_original_image_link dcterms:source link to the source image at the holding institution
link / download_link dcterms:source source record / download page
original_id dcterms:isPartOf / dcterms:identifier source-catalogue identifier; shared values group multi-page book objects (the meisho-zue case)
link_id dcterms:identifier record identifier within the dataset
longitude / latitude dcterms:spatial a-priori viewpoint coordinates (DCMI Point)
view_type project-specific (Smapshot); no Dublin Core equivalent
holding institution (in license) dcterms:publisher digitizing/holding institution

Dataset-level (Japanese-language works; resource types are still images and illustrated books): dcterms:language = ja; dcterms:type = Image / Text.

Dataset (collection) level

DCMI term Value
dcterms:title see CITATION.cff
dcterms:creator Santschi; Panday; Tsuji; Richardson
dcterms:publisher ThePrintLab / University of Zurich
dcterms:rights CC BY 4.0 (metadata); images remain with the holding institutions
dcterms:subject ukiyo-e; meisho-zue; digital art history; citizen science; georeferencing
dcterms:type Dataset
dcterms:date release date

Coordinates (dcterms:spatial) are free-text lat/long and are not linked to a gazetteer (e.g. GeoNames, Getty TGN).