If we are becoming so much more open and transparent, why are we still seeing this?

I work in the world of Open Data. I am passionate about community engagement in a movement that so often focuses on subjects like interoperability, data standards, data file types, and other terms that I still have a hard time wrapping my head around. However, I am highly committed to the core values for which this community advocates: transparency, openness, and better participatory government.

I, like so many, am so deeply bothered and saddened by what I am seeing coming out of the United States, my home country, over the past few weeks. I am an American, but I work in France.

Over the past year that I have been in Open Data, I have seen the Police Data Initiative take off. I think that it’s a great idea, and can clearly see that its underlying ideas are well founded in the principles of the Open Data movement. When the White House launched the project last year, it proclaimed it as a commitment to “improving the relationship between citizens and police through uses of data that increase transparency, build community trust, and strengthen accountability.”

After a year of this movement officially being in place, and after a few days of images of riots and militarised police showing up at protests taking over my social media feeds once again, I’ve been wondering quite a few things. I’ll sum it up by saying this: if we are supposed to be so much more open, transparent, and accountable to communities and our citizens, and collaborating for transformative change, why are we still seeing this?

It all looks great on paper. . .

The Police Data Initiative’s Open Data portal lists five main desired outcomes of the project:

1. Community analysis of Open Data could yield important insights into policing.
2. Open Data helps the community understand what police do and provides opportunity for two-way engagement.
3. Open Data demonstrates transparency and can promote legitimacy.
4. Public safety data is important in addressing broader community concerns.
5. Opening law enforcement data can help identify new tools and better processes to improve public safety.

It’s worth drawing attention to the first two points here. The first refers to cross-referencing data that doesn’t necessarily directly concern policing, but also education and public health data for example, to show insights into our community that may be associated with more heavy policing in a community. The second point is also a call for productive engagement between communities and their governments, in this case police agencies, in order to produce more effective exchanges to change communities.

It all sounds great on paper, right? I think we can safely say that it’s not working on a national level in the way that its creators had likely envisioned. There is still an enormous amount of mistrust and unnecessary violence.

Shortcomings on the Open Data side

The Police Data Initiative portal itself has 56 participating agencies, including some where more high-profile incidents have occurred, such as Baltimore. The site is not inclusive of all open police data, as many other agencies have published their data as part of a city’s Open Data initiative, but not this larger project. Thus, there is much more police data out there, but mostly it follows the main themes of the White House’s initiative.

There are several categories of interest for visitors:

• Assaults on officers
• Calls for Service
• Incidents
• Officer Involved Shootings
• Stops, Citations, and Arrests
• Use of Force

Visitors could be excited by the potential insights they might get here, right? Unfortunately, there are only 4 datasets containing Use of Force information, 1 of which is up to date for this year. Only 12 agencies published data on Officer Involved Shootings, of which 6 are up to date. Concerning Use of Force data, there are only 4 datasets, one of which is up to date.

There is a shortage of data. The data that are there are, as a whole, not up to date. The data also lack the context which could make them remarkably more valuable as well. Sometimes, the data also come in the form of a CSV file, which is highly inaccessible to the average user. It’s no wonder why these data are not playing as much of a role as they could be!

The Open Data community is still, as a whole, fairly techie-centered. Many Open Data initiatives are built for transparency, however developers remain a primary target audience instead of citizen groups and activists that do not have the same technical skills. If Open Data is supposed to be about transforming our communities for the better together, then this needs to change. Open Data needs to come with tools that anyone can use to make data be a resource that anyone can connect with and understand.

Here’s an example of how we can take the data to create visualizations that serve a real purpose:

Go ahead and play around with the data directly on the site. I’ve listed just a few of the highlights. This doesn’t tell the full story, but it does certainly raise questions that beg further interrogation. The power of Open Data as a resource lies in how easy it can be shared and the ease with which anyone can interact with data. Put some more context with them either in an article or in a presentation, and you have a tool to illustrate injustices taking place around you.

Where do we go from here?

There is already a wealth of policing data out there that, as mentioned before, needs to be better connected with people outside of the technical community, and Open Data community. The data are being reused, but how? Too much attention is given to applications that could only exacerbate complex problems that already afflict our communities rather than solve them. Crime data is often cross-referenced with data on school performance to give families an idea of where they should move to live in a safe neighborhood with good schools. Nothing is wrong with wanting the best for a family. However, there is a risk of further gentrifying and segregating communities based on who can afford to live in these areas and who even has the technical know-how to search for and to use such tools. Where are the reuses to show correlations between transportation and job access, low performing schools, public health issues, poverty, and crime in order to advocate for resource allocation strategies that would strengthen communities in need?

There has already been at least one hackathon that has used Open Data to improve access to low income housing. A Code for America Brigade has also recently succeeded in building Comport to measure police officer interactions with residents to build safer communities rather than to just reduce crime. These are great initiatives to use data in a productive way for all, but they are simply too few and far between.

We need a larger diversity of actors to be able to access the data in order to think of new reuses that the original publishers may not have envisioned. We need a diversity of thoughts to be able to look at, analyze and question data, drawing from their own unique life experiences, to stimulate more critical debate around what is happening in communities. We know that the tech community is not a hotbed of diversity, so this is why Open Data must be more accessible to people of all backgrounds. For the sake of the movement, it must evolve and grow past the current small tech bubble.

There is so much potential for Open Data to empower actors in cities and communities everywhere. It is a resource to bring forth evidence to those in power in order to demand changes. It concerns each and every one of us, and can tell stories about how our individual actions collectively have an impact on those around us. When we can more easily understand the data, reuse it, and share it, we are getting closer to a way that can help us see what impact our actions have on others, while still giving us the insights we need to build change collaboratively. Let’s start moving in this direction. We can start with improving Police Open Data, as there is a certain immediate need, but let’s not lose sight of this overall going forward.