Morgan Herlocker is an open source software developer and creator of the Turf distributed geospatial analysis framework. He has worked on mapping and statistics software across the geo industry, from consumer navigation to citizen counter-surveillance. Morgan has been a vocal advocate for user privacy in the design of location telemetry systems.
Morgan was interviewed for GeoHipster by Mike Dolbow.
Q: I really enjoyed your talk on location privacy at SOTMUS in Minneapolis last September. In December, the NY Times did an article on this very topic. Do you think this will continue to crop up in news media in the future?
A: We’re living in a moment when authorities are asking for more location data than ever through contact tracing to prevent the spread of COVID-19. At the same time, millions of people are out in the streets demanding that our officials abolish discriminatory policing practices that have explicitly targeted BIPOC communities across all US cities. The tension is palpable and people are asking “If I give officials access to my trips, will they use it against me or my neighbors in the future?” Privacy is a personal safety issue, a civil liberties issue, and a public health issue all in one, and it will be front and center as our society grapples with the underlying inequities that are increasingly front and center in this moment.
Q: Can you tell us how you managed to get into mapping and/or GIS?
A: It was a bit of an accident. I was always interested in computer graphics and data processing, but was mostly focused on audio, statistics, and game engines at first. In 2013, I started the Turf project to build a toolset that would make crunching geographic data easier and faster. At the time, I hardly knew how to make a map, knew little about data visualization, and did not even know what a GIS was. I came in with a healthy dose of naivete about how geo software could or should work. I wanted it to be more like the web – more like Unix, and was surprised to find a huge movement of people at Mapbox and elsewhere who were working towards the same thing. The Turf project hugely benefited from the contributors I met at Mapbox, and opened my eyes to the rest of the open mapping ecosystem.
Q: Do you think folks in our industry are more aware of location privacy issues than the general public? Or are we blissfully unaware, like some infosec experts who buy IOT devices?
A: Geo is a big industry. Some of us are actively complicit in the creation of these issues through working on ad tech and law enforcement tech, and it’s hard to claim that this is due to ignorance of the negative externalities. What we choose to work on is an expression of our values, and unsurprisingly, there is plenty of disagreement in this professional sphere. That said, privacy is almost trivially solvable on the technical side compared to the political front. We need to demand regulation that protects citizens and consumers from ubiquitous surveillance. These decisions should not be left to the whims of a police chief or a VP of marketing at BigCo. We are increasingly seeing state legislatures in the US addressing these issues, and I’m hopeful that we will see a federal response in the next several years.
Q: Do you think the current pandemic situation changes our tolerance for location privacy? Or will contact tracing have to be mandated in order to be effective?
A: Contact tracing is difficult to mandate and enforce in most constitutional democracies. Public agencies need to build a record at a small scale that shows they can be trusted to handle data at a large scale. Voluntary systems that extrapolate statistics across the entire populace seem more viable from both an adoption and effectiveness standpoint.
Q: Do you predict any adverse effects on the open data trend resulting from privacy issues? It’s only a matter of time before a whole “portal” gets taken offline as an overreaction, right?
A: It could happen. Some of the privacy disclosures I have made resulted in a portal being taken down for a few days, patched, then brought back online. This is the ideal situation, but I could imagine smaller teams deciding to drop sharing data entirely. While that would be unfortunate, being realistic about the resources required to maintain a system that includes personally identifying information is critical. If a dataset isn’t safe for sharing, it should not be on the internet. Often, this is mandated by existing laws and regulations, especially for government agencies in certain US states.
Q: Outside of location privacy, what else excited you about our industry? Any interesting projects you want to share?
A: Microsoft Flight Simulator 2020 is my “must watch” for geo tech this year. Is it a map? Is it a GIS? I’m not sure, but it clearly sets a new bar for how I think about cartography and providing spatial context to users.
Q: What do you like to do in your free time?
A: I’m really into gardening for food and flowers using the concepts of permaculture and closed loop systems. I’ve also been using the extra alone time during the pandemic to brush up on my jazz guitar chops. I try to follow my interests and don’t draw hard lines between professional and personal hobbies.
Q: What do you think has defined “the geohipster” over the last decade? Will this change in the 2020s?
A: A fundamental, at times borderline irrational, belief that empowered communities can solve challenging problems through collaboration without a central extractive authority. The groups in the geo software industry with the most lasting success have made this real by fostering community, designing for sustainability, and investing in getting the usability details right. That won’t change anytime soon.
Q: Any parting words of wisdom for our readers?
A: “Geo” is often used as a catchall industry for people making maps for a particular narrow set of purposes. This leads to less diverse teams and less useful software. Look for ideas and talented people in adjacent industries that do similar work with a different label. If we think about geo as a skill set that includes anyone pushing the limits of computational geometry, there are far more avenues for development. If we think about geography as a skill set for understanding the arrangement of human infrastructure, we can learn from people with non-traditional backgrounds like real estate or delivery logistics. There is so much untapped potential in the field, but we need to do the work to recruit and welcome these contributors.