Open ljeremias opened 6 years ago
Running re-harvest of HC collection, keeping existing subject strings
After review of 2.21.18
Here is the list of columns from the History Colorado source spreadsheet:
['Object ID', 'Object Name.Term', 'Description', 'Title', 'Non-Original Title', 'Count', 'Inscription', 'Maker.Term', 'Dates.Date Range', 'Dimension', 'Subject.Term', 'Used.Term', 'Locale.Term', 'Period.Term', 'Culture.Term', 'Publisher.Term', 'Style/Tradition.Term', 'Credit Line', 'Collection Name', 'Collection Type.Code Description', 'Collection.Code Description', 'Copyright.Copy Right Category Type.Code Description', 'DPLA Rights', 'Copyright.Rights Granted']
Originally we were mapping the ' 'Dimension' column to bf:diminsion property for bf:Instances. Should this property be mapped to the dcterms:extent for the Map profile? Is there another field in these columns that should be mapped to extent?
Yes, dimensions should map to extent.
Subject headings don't always seem to match HC's collections. See row 285 and 1805. Its almost like some Wyom State Pub and/or Marmot subject headings are mixed in with HC's data. Look for WSC, I believe that stands for Western State College. I have not look at other data providers yet. Round2.HC.xls.xlsx
Dates did not map - but this is an overall issues with all the partners
HC supplied extent but this field did not map
Alternative title did not map
Identifier did not map - I don't think this is necessary for the first load to DPLA but would like it to be part of their profile
Record totals do not match pre and post aggregation. HC supplied 7292 records. After aggregation they are credited with 8200