Gaps

Minimal Interoperability

No Concept of Shared Search

No True Concept of Work Type or File Type

The concept of work types, or content models, is mostly missing from the digital library program. As a result, things aren’t always optimal, options are often limited, and software dictates where content is deposited. In my opinion, this goes back to the historic adoption of DSPACE and the fact it was historically devoid of content model. This continued on in the adoption of Fedora and how PCDM was initially implemented. For instance, all works has the concept of pages even if a work is made of one file. Similarly, files are only differentiable via mime type. Therefore, a PDF would always be treated the same regardless of its actual purpose.