Preparing metadata for a migration can be one of the most time consuming and labor intensive parts of the project. Generally speaking, there are two main categories of metadata preparation work: Remediation and Mapping.
This activity focuses on Metadata Remediation, which refers to any clean up work that you would like to do before mapping and migrating your content. If there are metadata fields that have been used incorrectly or inconsistently, this is a good opportunity to revise them. Remediation may include field name changes and/or consolidation, standardization for dates, and other similar activities.
Objectives
- Based on current state of metadata, determine the necessity and extensiveness of metadata remediation
- If remediation is needed, document the goal, the scope of work, decision making process, and timeline in the Metadata Remediation Project Plan
- Reach the stated goal for metadata before migration
Suggested Collaborators
- Project Lead
- Technical Lead
- Developers
- Functional Specialists, i.e. metadata librarians, digital preservation teams etc.
- SME on data and data migration
Available Template
Example
Instructions
There are three main steps to metadata remediation:
Step 1: Review Current State
If you haven't done so already, export Fedora 3.x content and review the current state of metadata fields and elements.
Step 2: Plan for remediation
- Review the Metadata Remediation Guidelines
- Complete the Metadata Remediation Project Plan
Having a project plan can help clarify goals, efforts required, and decision-making process for the project team. This plan can also serve as a reference throughout the remediation process.
Some questions to consider as you plan:
- Based on the timeline of the overall migration project, how much time will you dedicate to metadata mitigation? And do you need to reconsider the goal and scope of remediation?
- Are there any experts outside of this workgroup who could help you make decisions or speed up the remediation process?
- Are there tools that could aid the remediation process?
- How easy/difficult would it be to remediate metadata after your migration instead of before?
- Which aspects of remediation (e.g. consolidating or standardizing fields) should happen before migration because they could affect the configuring of the new system, and which could wait because they are unlikely to affect how the new system will be implemented (e.g. modifying descriptions)?
- Are you following a designated standard?
- Does your metadata cohere to the same standards across all collections?
- Are you creating standards during the remediation process?
- Document everything
- Add dates to indicate when a change was decided on
- Add reasoning for the change. This can help prevent circular discussions/questions regarding why the change was made.
- If/when remediation seems overwhelming, remain focused on things that can be fixed via batch changes and document that decision. By focusing on functionality and discoverability over uniformity, you can help determine when the metadata is good enough to move forward.
Example: formatting dates in batch to help with sorting and keeping the raw, un-clean data in another field for later remediation
|
Step 3: Conduct Remediation
Conduct remediation and document your decisions, changes, and results as outlined in the Project plan.
Example: Whitman College’s Data Remediation Process (Demo), which includes thorough evaluating metadata fields between collections, deciding on future state of those fields, tracking changes being made, and editing metadata content.
Next Steps:
- Consider whether you need to restructure your metadata.
- Before a test migration or a full migration, consider exporting and reviewing your Fedora 3.x content again for a final review.
Related articles
Related articles appear here based on the labels you select. Click to edit the macro and add or change labels.
