The Project’s work, spearheaded by two community volunteers, Bob Williams and Lisa Boccetti, was accomplished with no budget and at no cost to taxpayers. The Wake County Register of Deeds Office, which is led by Tammy Brunner and manages many public records including historical deed documents, supported it.
The endeavor, Williams said, is “opening a window into the past” by making historical information publicly accessible. It was important, he said, to make the data available in different formats, so it could be accessible regardless of a person's technical expertise.
“It’s one big data set sliced in different ways depending on what your audience is,” he said.
The video below, from Wake County government, shows the spread of racially restrictive covenants in the county, from 1905 to 1950.
The Project is just one of the ways that the county government is working to enhance storytelling.
Racially restrictive covenants, although illegal today, have had lasting impacts. Even now, there are gaps in homeownership for marginalized populations like Black Americans.
Wake County’s work in this area started with a legislative mandate to preserve a longitudinal database of historic records and enable public access, Boccetti said.
This historical data led to two unique initiatives. The first, the Enslaved Persons Project, digitized information about formerly enslaved people for public access. The Racially Restrictive Covenants Project is the second.
Without modern technology, assessing the racial covenants that existed in the county between 1900 and 1950 would have required reading 600,000 pages of documents and analyzing 5,000 maps, Williams said.
The project team instead used a random sample to determine approximately how many racial covenants they should be seeking, then identified deeds that may contain these covenants using Unix scripts and Tesseract Solutions software — the latter of which uses AI to identify relevant information in documents.
These tools found the documents that likely included racial covenants, based on keywords and phrases associated with racial terms — no fewer than 20,000 of them. These were shared with 200 volunteers for closer analysis — community members, people from the University of North Carolina at Chapel Hill Law School, North Carolina Central University School of Law, and an undergraduate history class at North Carolina State University.
All told, the volunteers contributed an estimated 10,000 hours of unpaid labor to this project, Williams said.
The “outpouring” of community interest in the county’s past, including from those who found it too painful to participate in the process but still wanted to see the results, was the most “heartening thing” about this project, Boccetti said.
The database was built with MySQL, and one challenge was that the historical documents did not use the same property designations used today, Boccetti said. A document might read, “this property starts at the pile of rocks and it goes to the tree stump that used to be there,” she said. In one case, a volunteer’s deep knowledge of the county’s history helped solve several such questions — adding valuable context to about 50 untraceable deeds.
The work and its resulting map revealed something about the county that Brunner had predicted would emerge: the places impacted by these racially motivated restrictions are the same parts of the county that are underserved today.
It was important for the county to put this information out there to demonstrate the historical context behind why these places are underserved even today, Brunner said: “The proof is in the pudding; these are facts."
The inclusion of racially restrictive language in these deeds was intentional, Boccetti emphasized. And the results of this systemic discrimination, which included inequitable infrastructure investments, are still seen today in issues like flooding, which disproportionately affect the historically disadvantaged — and historically Black — neighborhoods in Raleigh, she said.
The project was done using all open-source software. Williams designed its methodology so that it could be handed over to another agency that wanted to create similar tools, Brunner said, noting that she would like to see every North Carolina county use these tools to make covenant information more accessible.
