We are often unaware of how we exercise power in our metadata choices. Librarians can unwittingly perpetuate social inequality, especially based on identifiers of race, gender, religion, sexual orientation, gender expression, class, and ability. Metadata schemas rely on controlled vocabularies paired with searchable descriptions, but those vocabularies can be rigid, dated, and hard to amend, thus alienating users by making it difficult for them to find affirming, humanizing representations of themselves in the library.

We created a group to review our digital collections metadata practices. The group investigated how we contextualize collection creators and contents; assumptions about any form of identity; and addressed the perpetuation of “whiteness as default.” This group met bi-monthly for a year to discuss readings, make decisions, and create documentation. The group also held a session with LGBTQIA+ staff to reflect on representation in metadata; and created a public resource on harmful content in our digital collections.

Through this project, we updated our internal guidelines and recommended changes in donor relations. Throughout the process, we publicly posted documentation about the benefits and limitations of LCSH, how and why we use alternate controlled vocabularies, and provided context about collection creators and contents. Identity is unique and personal, so we recommend donors are consulted during the donation process about how they wish to be described.