I last wrote about the cost of knowing in the particular case of McKinney, Texas, where government staff appear not to have the ability to search through email correspondence that was sent or received before March 2014. While our experience at Code for America bears out that the city of McKinney is probably typical — many local governments struggle to have sufficient knowledge or capability in technology — the news isn’t all bad.
One government department that’s moving against this trend is Seattle’s Police Department, who are taking significant steps in opening and improving their data. Seattle in general is pretty interesting. Its Public Records Act is probably one of the most liberal in the country in terms of what the city must release to the public.
Seattle’s Police Department is not without its own problems and history. In 2011, after an 8-month investigation, the Department of Justice found “a pattern or practice of excessive force that violates the Constitution and federal law” and serious concerns about biased policing. To make sure that the findings of the report would be addressed, the police department and Department of Justice entered into a consent decree in September 2012, estimated to last around five years.
It’s against this background — with our new focus on safety and justice — that Seattle’s open and improving relationship with the local technologist community is of interest compared to, say, a city like McKinney. Here’s a story about how you turn someone requesting public records who might be an adversary into an ally, and where you can end up with an outcome that’s demonstrably better for everyone.
This story starts with Timothy A. Clemans making a public records request late last year for pretty much all of Seattle PD’s video footage. There are a few telling pieces in that article from the Seattle Times: that Mike Wagers, the department’s chief operating officer sees common ground between the request and both parties wanting the police department to be transparent and accountable. But the flip side of records requests is that they sometimes legitimately require a lot of work — or at least, a lot of thinking about — to fulfill. For instance:
In the past, requests for police video have forced agencies to pay employees to work full-time to scrub audio and video, frame by frame, of sensitive information — rape victims, children who have been abused, suicides and violent-crime scenes — and are being cited by law enforcement throughout the state as a reason not to place cameras on officers, according to the Washington Association of Sheriffs and Police Chiefs (WASPC).
- Costly public-records requests may threaten SPD plan for body cameras – Seattle Times, 20 November 2014
Seattle police spokesman Sgt. Sean Whitcomb said “the technology does not exist” for the department to “scrub” police videos of sensitive, or redactable information.”
This is pretty much a hands-down “I bet you can’t do this” challenge to a particular type of technologist or hacker. So Wagers, the COO mentioned above, invited Clemans in for a meeting to help figure out what to do with Clemans’ request, because it turns out that the police department, generating around nine terabytes of video a day just from patrol cars, didn’t really have a good way of dealing with all of that video data. (And, to be fair, most organizations wouldn’t, really.)
This is an interesting technical problem where it looks like if you understand technology and you agree on the aims, you can at least try to work together to solve the problem. Or come up with a better solution that doesn’t involve having to employ people to laboriously watch and redact video.
So that’s exactly what the department and Clemans did: first, a hackathon, to invite technologists in to play in exactly the messy gray bit of the venn diagram where transparency, accountability, and privacy intersect.
The end of this story is where the police department essentially ends up hiring Clemans, and Clemans becomes part of how the department starts to see technology as an ally: in other words, as an incredibly powerful tool that needs to be understood and directed to provide a service that works in the 21st century.
This type of lightweight problem solving using technology — where there are new ways to work quickly with video, to identify faces, to do transcription on audio and to recognize names, to do redaction — is being directly applied to massively improve the transparency and accountability of the police department, as well as help the public understand better the jobs that police officers do.
Here’s one part that I like about that Stranger article covering the hiring of Clemans: Wagers, the department’s COO (who, you’ll notice, is one of the people who’s been quoted as saying smart and reasonable things throughout this post), pointed out that Clemans has a talent that the department doesn’t possess internally. (It might not be a coincidence that the department knows a good thing when it sees it, given that it also recently hired Greg Russell from Amazon to be its new chief information officer.)
Seattle Police Department and Clemans working relationship is pretty much a poster child of how we’re thinking about and applying 21st century government. On the one hand, 21st century government is about being in control of enough of your infrastructure — of the how you deliver government — that you’re able to search through your old correspondence without it costing too much. Actually, that’s not even a hand. That’s just showing up. That’s just the basics. On the other, a vastly more important hand, is how we move forward, it means recognizing that transparency and accountability are vital to delivering not only good government, but government that improves, and that government leadership needs to make sure they have the people with the capability to build, commission, or rent the tools that are needed to deliver.
In this case, both the police department and Seattle’s residents and citizens are getting something out of the deal. Residents and citizens are getting a new way to make sure that their police force is held accountable. Finding a way to automatically publish police footage turns out to be another productive way of increasing trust: helping the police be collectively accountable and helping the public understand better what policing in their city looks like in the thousands of interactions with the public that don’t make the headlines.