...
Even though delivery services are out of scope for the DRS Futures project, the changes necessary to improve delivery files begins at the DRS level, by creating ad-hoc deliverable copies of archival and production masters, which are currently serving the dual purpose of archival and delivery. The solution to this is to establish By establishing post-deposit processes that automatically generate an ad-hoc delivery file for each archival file deposited, thus separating we would be able to separate the two purposes.
At the same time, a mass conversion process of existing media files that uses the same tools can be planned and executed in parallel on a different timeline. Beginning with groups of media that are deemed to have the most negative effect on current delivery performance, the full set of deliverable media would be converted to delivery-optimized copies. This endeavor, which can span multiple years, should be subdivided in sub-projects focused on a specific media batch, which will be described by individual project charters under this section. The various media collections for each project would be defined based on assessment actions, carried out by automated analysis of our media collections and existing error reports.
Metadata completeness and accuracy may be assessed by establishing a set of criteria that can be computationally evaluated, to rank and classify existing digital records using automated tools. Remediation actions may be decided once an exhaustive report is available from this assessment step.
The current DRS content model and ontology should be reviewed in the light of today's DRS stakeholder needs and advancements of digital preservation concept and scope within the DP community. Once RS Futures services are live, which should allow for an easier modification of the content model configuration, batches of digital resources can be migrated one at a time with the help of ad-hoc conversion tools. These tools would be Harvard-specific but flexible enough so that they can be reused for any content model migration.
Our vision aligns with the following Harvard Library multi-year goals and objectives (MYGOs):
...
III. In Scope/Out of Scope
In Scope
- [ To Do ]
Out of Scope
- [ To Do ]Al projects within this section deal exclusively with data migrations for improving the quality of data and metadata handled by DRS.
- If one-off scripts and software need to be developed for a specific migration, that would fall within that migration's scope.
Out of Scope
- General-purpose tools that can be reused for multiple migrations (e.g. DRS Pipelines) are out of scope for this section, but many projects in here will depend on those tools.