Last week in Denver, I had the pleasure of moderating one panel on a topic that has been a fascination for me since entering this industry: Open data. From my perspective, one of the biggest things that’s always stood out to me about the geospatial industry is the sheer amount of data involved, and how siloed much of it seems to be. Of course, there has always been some element of sharing and openness around this data, but it has seemed like that idea has been embraced to a much greater degree in recent years. It felt important to talk about that at this year’s Geo Week, and we were fortunate enough to have an esteemed panel from some of the biggest players in this realm to discuss the issue.
The session, entitled What Open Data Means for Geospatial Professionals, featured the following panelists:
Chris Andrews, Cesium (part of Bentley Systems)
Alex Chu, Google Maps
Amy Rose, Overture Maps Foundation (OMF)
Scott Simmons, Open Geospatial Consortium (OGC)
The 90-minute conversation touched on a number of different topics around open data in the industry, with each panelist coming to things from a slightly different perspective. Andrews and Chu, for example, work for public companies that are embracing open data, while Rose works for an organization made up of some of the world’s largest companies – Meta, Microsoft, AWS, and TomTom were the founding members of OMF – all coming together to create an open map. Simmons, meanwhile, works for a large consortium that includes companies and organizations from around the world, with the OGC creating standards used throughout the geospatial industry.
These different perspectives made for an interesting discussion around not only the positives of open data, but also some of the negatives and barriers that come from really embracing an open data ecosystem in the industry. Andrews, toward the beginning, talked a little bit about the history of this idea as he has seen it throughout his time in the industry, noting that it hasn’t exactly been a linear line. From his perspective, he saw plenty of open data ideals in the late 20th century, with things closing up a bit post-9/11 before recently starting to trend back toward open.
Andrews talked about his experiences in the 1990s with CDs that were shipped out by Esri as an example of how some of this open data, along with other examples like GIS parcels being shared and GIS data being put out by utility companies. However, as he noted, after 9/11 happened, much of that data went away. Since then, that sharing has started to trend up, and he points to the creation of Google Maps as being an impetus for that trend, along with other technologies driving more of that sharing.
After establishing some of this history, the conversation turned toward some of the challenges and barriers that come with embracing open data. As noted during the session, it’s great to want to embrace this idea, but it’s not as simple as just flipping a switch. There is some friction involved, and a 100 percent open data world is simply not possible, even in the eyes of the most optimistic.
Chu called out a few of the challenges that they have encountered at Google around opening up some of their data. For example, she acknowledged that at this point Google Maps is 20-plus years old. So, for any feature they have to dig through years of information and terms of service to figure out exactly where they got information, whether or not they can open up that data, and if so how they can go about it.
She and Rose also talked a lot about the issues around data governance and how challenging just interpreting data licensing can be.
“We would prefer that everything is permissive, and it makes it a lot easier,” Rose said. “But there’s going to be licensing issues where you just have incompatibility. Putting them together makes it very difficult for users downstream to be able to use those, so it really matters when it comes to building data or putting data out. I know we won’t even look at data if it doesn’t have a license attached to it because there’s no information on how it can be used and if there needs to be attribution, or what the adaptive impacts might be.”
There’s also the issue of regulations, particularly in an industry like geospatial where all of the data is necessarily global. Different parts of the world have different regulations about sharing and data governance, and working around that is a challenge for massive datasets like what we see in the geospatial realm. Simmons, who works with organizations around the globe, touched on this challenge, noting that different areas have different interpretations of what a regulation could mean.
“Sharing data across the US/Canada, US/Mexico border can be a fairly significant issue because we have three different regimes for every aspect of our privacy and rules,” he said. “Who pays for what depends on the implements, how it’s re-used, etc. And then that varies even across the [European Union].”
It wasn’t all about the negatives for open data, of course. The panel also discussed some of the ways to help encourage more adoption around open data ideas. Rose was able to provide some key insights into this idea given the big companies involved in OMF. Asked how it’s possible to not only get these companies to work together, but to do so relatively efficiently around open data, she said it comes down to a shared mission.
“I think it’s something you don’t really have to actively motivate in order to push it. All of these companies, particularly the founding companies, they all very much had a shared vision. In fact, they came to separately, but at the same time, and that’s how they came together on it. They have actually invested in this idea.”
She noted that they are actively investing their own resources into the project, with dedicated engineers whose job is to work on Overture as their full-time job.
They also talked about some of the benefits around this idea for things like disaster response. Chu talked about some of the ways that Google is leveraging open data and sharing its data in response to major disasters. For example, she pointed toward the crisis in the Middle East right now, with Google sharing the location of bomb shelters, NGOs, and nonprofits working in Israel and Gaza. Additionally, she talked about how the data that they have available can be used for environmental purposes, like flood forecasting.
These types of conversations are really important for the geospatial community, and really anyone who works with data, diving into both the merits and the drawbacks of open data. It’s easy to say you want to embrace open data, but without accounting for the barriers, challenges, and drawbacks, it’s just a setup for failure. These trends will continue to change over time, but this session represented a good snapshot of where we are now, and where we could be going. The biggest takeaway, though? It’s not as simple as just being open, or closed, as Rose said.
“I would encourage everybody to think about open – whether it’s open data, whether it’s open standards, open code – as not the opposite of closed. I think a lot of people tend to look at that as very black and white, even though we all know in our lived experiences that it’s a very gray world out there.”
