Migration Planning: What Do I Need to Do to Get Ready for a New Repository?
Migrating your content from one existing library repository system to another can be a daunting task. But with careful planning and execution, the process to migrate can be relatively smooth. Here are some tips to help you get started.
The different types of migrations
Most of our custom work entails some form of content migration. Some use cases we’ve seen are: 1) Customers who have new collections that they need to get it into a new repository; 2) Institutions who want to move their content off of a proprietary platform; 3) Customers who can no longer manage their in-house platform - whether they lost technical staff who ran it or it’s in a state of falling over and they want to migrate to a 3rd party hosted solution. Here are some things to consider as you plan for your particular migration project.
Starting a new repository
When migrating new or fresh content, you need to think about how you will use the content. In our previous post, preparing your import you can learn more about structured content and file locations.
Migrating from a proprietary platform
If you're migrating content from a proprietary platform, you need to have an understanding of the differences between the content model of the old and your new application.
Self-managed to 3rd party hosted solutions
When migrating content from your in-house, self-managed system to a third-party hosted solution, you'll also have to review and analyze your content model and metadata.
How to prepare for a migration
Here are some things you need to consider when planning for your migration, no matter what type of migration you’re doing.
Content
It's important to consider the type and structure of the content you're migrating. What file formats are in use, and how are your digital objects structured? Can they be accommodated in the new system?
It's also important to inventory your file types and locations. This will help you determine what needs to be migrated and how. You may decide to upload files from your local file system, or migrate existing files or derivatives directly from your current digital library asset management.
With careful planning, a digital library migration can be a relatively smooth process. By considering all of your content and file types ahead of time, you can make sure that everything is accounted for and that your new system is able to support your digital library.
Quantify content
An essential part of preparing to migrate content to a new platform is understanding the size of your repository. Knowing what types of works, how many objects, and how large the files help determine what disk space you’ll need. It also helps avoid unexpectedly running out of disk space in the middle of the import process. You can read more about Quantifying your Repository here.
Organizing content
When building your repository, it's essential to think about how you will use the works. Are you updating a group or individual work(s)? Are you adding new collection(s)? Are you adding a child collection to a parent collection?
Metadata
When planning a migration, it is important to understand your current metadata and the metadata requirements of the new system. This includes understanding the profiles and schemas used, the data types of the fields, and any controlled vocabularies in use.
This is a great time to assess the quality of your metadata and identify any remediation needs. For instance, we’ve seen dates stored as strings and others in date format. It’s helpful to do the data cleanup work prior to a big migration project.
Map Metadata
To begin, you'll need to map your current metadata schema to the schema of your target system. This can be a challenging process, as often there are structural differences between schemas. However, there are many sample crosswalks available online to help you make the transition.
Once you have mapped your schema, you will need to decide which fields to keep and which to exclude from your new system. Sometimes, you will lose specificity or granularity when migrating to a new system. In some cases, it is important to make these decisions based on your institution's context and knowledge.
Finally, once you have completed the mapping and selection process, you will need to export your metadata into the format required by your target system. With careful planning and execution, migrating your content to a hosted third-party solution can be a smooth process.
What to do during the migration process
During the migration process, you need to make sure that all the content is transferred correctly. Set up some test imports and figure out the time and effort required to complete a successful import. Some systems may require individual imports while others allow for more bulk imports.
Summary
When migrating your digital library to a new platform, it is important to plan ahead and understand the data models and metadata requirements of the new system. By mapping your metadata and understanding your current schema, you can make informed decisions about your migration. During the migration process, it is important to set up test imports and figure out the time and effort required for a successful import.
References:
http://bit.ly/dpla-MAP4-crosswalk
https://github.com/osulp/dspace2hydra
https://github.com/UVicLibrary/vaultify
https://www.tenejo.com/blog/2022-05-10-quantifying-your-repository.html
https://www.tenejo.com/blog/2022-05-04-how-to-import-content-with-tenejo.html
https://bridge2hyku.github.io/migration/
Are you interested to learn more about Tenejo? Let's chat. Check out the links below or shoot me an email.