These guidelines were designed specifically for the Rice University Historical Images collection

Workflow steps: Batch process metadata, manually add files

Download file >Filter data >QC>Prepare data>Save to CSV>Import & review>Post>Add files

1. Download Shared Shelf spreadsheet and open in Excel

2. Filter : Determine items ready for ingest by applying filter to all records to select only those of "Item Status" = "2. Basic Metadata Entered"

3. Transfer filtered metadata to new sheet

• select filtered rows of data

• Find > Goto special > select option : visible only cells

• copy selection

• create new spreadsheet

• paste

• save new file (e.g. batch01Nov2011.xls)
  

   

4. Perform quality control checks:

• Check for duplicate items (WRC#)

1. Run pivot table test : count should be one per WRC# (confirms no duplicate records in spreadsheet) – see below for steps on creating pivot tables

2. Search DSpace by wrc# (to confirm items have not already been ingested in any collection)

3. double check the internal WRC spreadsheet of digi-object searching by words to ensure the item does not also exist under a different wrc#
   

    

• Run spell check

• Review capitalization and punctuation (i.e. DCMI type, format.medium, no periods after titles, etc)

• No abbreviations

• Multiple values separated by double rod ||

• Check metadata is populated in required fields (i.e. rights, source)

   

note:

Since data may be changed outside of Shared Shelf, be aware that once item is ingested, DSpace becomes the primary source (i.e. Shared Shelf may not reflect all changes)

     5. Use this clean data for photometadata emedding, proceed with creating derivatives

5. Prepare data for ingest:

• delete unnecessary columns

• update metadata headers –copy from spreadsheet

• insert columns for ID and collection

• Assign collection # per sub collection and visually confirm – see table below, copy to spreadsheet

• tip: sort by dc.subject and visually compare collection id to subject term

• Add boilerplate metadata to all records
  

• dc.publisher : Rice University

• dc.subject.lcsh : Rice University -- History (broad term. Narrower or more specific terms may be added later)

• check total items is no more than 20 (the maximum number of records that may be edited through the web user interface (XMLUI)

6. Save Excel file > Open Excel in Open Office Calc > Save as CSV encoded with UTF-8

7. Goto Dspace and run batch import, visually review changes (ie collection name matches subject, diacritics, etc) if ok then post changes.

8. Attach image files per new record. Tip: Search by WRC# (Edit this item > edit bitstreams > add file)

9. Update status in Shared Shelf

Tables

Note:  for metadata work copy lists directly from excel spreadsheet located on fonlibstor project folder. Below tables are provided for easy reference only.

Element Headers

Collection Ids

Boilerplate metadata

• dc.publisher : Rice University

• dc.subject.lcsh : Rice University -- History (broad term. narrower or more specific terms may be added later)