Metadata Movement

From UA Libraries Digital Services Planning and Documentation
Revision as of 16:32, 12 November 2009 by Scpstu32 (Talk | contribs)

Jump to: navigation, search

Diagram of metadata movement for content being digitized in Digital Services effective November 2009: File:Metadata.pdf


Digital Services:

When content is ready for storage:

  1. put the Excel file into S:\Digital Projects\Administrative\collectionInfo\forMDlib\needsRemediation for Mary,
  2. export a Unicode tab-delimited text file (use Open Office) and put it with the content for storage
  3. use a copy of this txt file to generate MODS according to the instructions here: Making MODS
  4. and upload the MODS to Acumen by following the process outlined here: Uploading_MODS
  5. then delete the text file

Metadata Services:

  1. Retrieve deposited spreadsheet from S:\Digital Projects\Administrative\collectionInfo\forMDlib\needsRemediation, and remediate metadata appropriately
  2. Save the remediated spreadsheet in S:\Digital Projects\Administrative\collectionInfo\Storage_Excel
  3. Generate MODS records, according to the instructions here: Making MODS
  4. Upload these MODS records to Acumen by following the process outlined here: Uploading_MODS
  5. Deposit the exported Unicode tab-delimited collection-level metadata spreadsheet in S:\Digital Projects\Administrative\collectionInfo\JodyPickup
  6. Notify Jody it's ready for pickup

Jody: Digital Services

  1. If initial uploaded content includes text files in the Scans directory (i.e. item-level tab-delimited metadata), move those to S:\Digital Projects\Administrative\collectionInfo\forMDlib\itemMD\
  2. Notify Mary they're ready for pickup
  3. Pick up finalized spreadsheets from S:\Digital Projects\Administrative\collectionInfo\JodyPickup and transfer to the deposits directory on libcontent1
  4. Process for correct deposit into long term storage and linking in LOCKSS Manifests

Metadata Transfer and Remediation for Collections Requiring Multiple Uploads

  1. Metadata is originally held in a .xlsx file.
  2. After Digital Services has ensured that the metadata received from the archivists has been formatted to the standard metadata format, this set of metadata is kept on the first worksheet of the workbook. The "original" metadata (i.e. received directly from the Archivists) may exist on a separate tab of the spreadsheet, should the need to refer to it arise.
  3. When enough item scans have passed quality control checkls to warrant an upload of materials to Storage (i.e. a "batch"), the metadata corresponding to the items to be uploaded is copied out of the original set of metadata and pasted into a new Excel file (.xlsx). The naming convention for this file is collectionNumber.n.xlsx, where "n" is the numerical value of the batch. Ex: u0008_0000001.2.xlsx = the metadata for the second batch of items for collection u0008_0000001.
  4. This portion of the metadata is then exported from excel as a tab delimited Unicode .txt file, and is handed over to be uploaded to the storage and web access server.
  5. At some point during this process the metadata unit will want to perform a remediation of the metadata information for this collection.
  6. the metadata for remediation in the .xlsx file will either be in the main worksheet or in one of the batch tab worksheets.
  7. By editing the metadata with precedence for metadata found on a tab in the .xlsx file, the possibility of creating versions of the metadata by having it in too many places is avoided
  8. If this collection is in progress its .xlsx file will be found in the collection metadata folder
  9. If this collection has been finished its .xlsx file will be found in the storage excel folder with other .xlsx files from other finished collections.
  10. Either way once the metadata has been remediated it will be re-output to tab-delimited Unicode .TXT files and handed off to be uploaded to the storage and web access server to replace the original tab-delimited Unicode .TXT files that were uploaded.


The following needs remediation: we are using tab-delimited UTF-8 text, not CSV. As each chunk of content is digitized, the tab containing the metadata will be pulled out of the main spreadsheet into a separate Excel file, numbered with the batch number (collnum.1.xlsx, collnum.2.xslx, etc.) and that will be deposited for remediation, and exported for MODS creation and upload for storage. The original reference metadata will contain an extra column for notation of which batch each segment of metadata belongs to. In this fashion, we keep track of which batch we're on.

This means that in Storage_Excel, we will wind up with multiple spreadsheets for each of the large or ongoing collections. This will also be the case in the archive. If there *is* a collnum.2.xlsx or collnum.2.txt, and there is a collnum.xlsx or collnum.txt, we will assume that the latter is the first segment of metadata for the collection. That is to say, it *is* the collnum.1.xlsx or collnum.1.txt file.

Diagram of metadata movement for collections requiring multiple uploads this image shows a potential policy: File:Metadata transfer and remediation.pdf

Personal tools