About me
Our stations' are nearly a century old, so there is a wide variety of quality in both metadata and audio.
My focus has been on normalization of metadata (within and across platforms), as well as data augmentation via APIs and Linked Data. For example, we are developing tools that analyze regex patterns in the station's press releases, extract significant data, and apply Library of Congress Subject Headings (LCSH). We then embed that metadata in the files for more robust discovery.
We are also in the process of executing a grant from the Levy Foundation to digitise 45,000 of our assets. We are implementing workflows that allow us to detect quality outliers at scale. This information is valuable when communicating with vendors.