Building New Jersey's First Statewide FOI Platform
My Role
I started the project in 2017 and launched New Jersey's first statewide freedom of information platform in June 2018. I designed the system, wrote the software, and ran it for roughly seven years before stepping down. The problem it solved was procedural: filing an Open Public Records Act request meant knowing which of hundreds of agencies to write to, in what format, and having no way to track what happened next.
I built it to handle the full lifecycle of a records request — submission, routing to the correct custodian, response tracking, and publication — and made every request and response public by default. That last decision was the important one: it turned a stream of one-off requests into a searchable public archive, so the same record never had to be requested twice.
Scaling it to 565 agencies across all 21 counties meant solving the unglamorous problems: agency contact data that no one maintained, wildly inconsistent response formats, and the storage and indexing of 250GB of released documents.
Outcomes
- 75,000+
- Requests Processed
- 250GB
- Data Released
- 565
- Government Agencies
Over seven years the platform processed more than 75,000 requests. The uses it was put to were broader than I anticipated when I started.
-
Investigative Journalism
Newsrooms across New Jersey used the platform to source stories on government spending, public safety, and local politics.
-
Academic Research
Researchers drew on municipal budgets, contracts, and policy documents to study local governance and public administration.
-
Legal Research
Attorneys pulled government records for compliance review, due diligence, and public interest litigation.
-
Public Engagement
Residents researched their own local governments, from school board spending to municipal contracts and public safety records.
Research Impact
- 2026
Countering Security-Induced Government Non-Responsiveness with Nonprofit Technology
View the OPRAmachine Data Paper → · DOI: 10.1016/j.dib.2020.106265 (Data in Brief, 2020)
What I Learned
-
Publishing by default changes the economics
Making every response public meant each request answered a question permanently, rather than for one requester. The archive became more valuable than the filing tool.
-
The hard part is the registry, not the software
No maintained list of records custodians for 565 agencies existed. Building and keeping that current was more work than the request pipeline itself.
-
Civic platforms need a succession plan
A tool people depend on has to outlive its founder's involvement. Handing it over cleanly matters as much as building it well.
Press & Media Coverage
- 2022
- 2020
- 2019
- 2018
Related Articles
- Jun 22, 2024
- Aug 20, 2022
- Mar 15, 2021
- Jul 13, 2019