Governance
I've spent a lot of time building systems that keep running without constant intervention. That requires governance, not just goodwill.
Governance work can look invisible from the outside. In practice, it's how programs scale without collapsing into a few overcommitted people.
UC Carpentries
UC Carpentries is a distributed, cross-campus teaching network. I built it into a formal 10-campus network with shared norms and decision-making structures. During the review period, it delivered 28 workshops reaching 2,896 learners.
The point isn't volume. It's resilience. The network runs without depending on any single campus.
Library Carpentry Curriculum Governance
In Library Carpentry, governance shows up as process:
- How lessons are reviewed
- How adoption decisions get made
- How quality standards stay consistent
- How lesson teams get supported instead of burned out
As Chair of the Library Carpentry Curriculum Advisory / Governance group, I formalized adoption pathways and helped bring multiple lessons into the official curriculum, including lessons developed under the IMLS open science work.
Redivis Governance for Restricted Data
Restricted data environments fail when governance is ad hoc.
When we implemented Redivis, I designed a governance model intended to scale. The goal was to support routine access requests and Data Use Agreement controls without turning the platform into a bottleneck that routes everything to the Director.
That governance work is part of why the platform could grow from early workflows into high-throughput research use — from 5 workflows and 444 GB processed to 2.9 PB processed in 2024.
Governance as Sustainability
My bias is toward systems that outlast individuals. Programs like UC Carpentries, platforms like Dataverse and Redivis, and curricula like the IMLS open science lessons only matter if they keep working over time.
Governance is what makes that possible.