SAN DIEGO, CA — Here are some quick takeaways after two days racing around a shoreline conference center in this sunny city at my first FOSS4G-NA. So far, we’ve taken Resiliency Maps to varying venues – including a United Nations event and a couple of open-source conferences (SCALE in Pasadena and FOSDEM in Brussels) – but aside from State of the Map, this was the first one dedicated to all things geospatial. And open source!
Definitely a well-run event (stellar volunteers, on-site childcare, many-branched donut trees, rivers of coffee) with way too many compelling talks to catch. The size kept it manageable (
300 people? Actually 500, much obliged for the fact check Jody Garnett) and friendly – but it deserves to be absolutely mobbed with at least 1,000 attendees. Let’s stick to the friendly vibe though, eh?
Two amazing women rocking the stage in #FOSS4GNA this afternoon @MicheleTobias (Digitalization of 240 Wine regions with collaboration across the #USA) and @VickyVvergara (Keynote #OSGeo abstract) 👊👊👊 pic.twitter.com/McKpaiZT0f
— Mapanauta (@mapanauta) April 18, 2019
How FOSS can be used to monitor sustainable development goals
How’s this for perspective: monitoring the SDGs from space! It turns out that satellite imagery can help report on about a quarter of SDG targets. Drilling it down, Azavea’s Ross Bernet and Jacob Bouffard talked about how their team calculated one of these indicators, SDG 9.1.1. For anyone (or all of us?) not familiar with that specific point, it’s the proportion of rural population living within two kilometers of an all-season road. (The related goal is to build resilient infrastructure, promote inclusive and sustainable industrialization and foster innovation.)
The Azavea team used OpenStreetMap plus population data from WorldPop, NASA’s socioeconomic data and applications center (SEDAC) and GeoTrellis. After that they created global set of vector tiles from OSM road data and calculated the percentage of the population in that distance range by country, for the entire world. What’s next? The work is ongoing and but the hope — beyond the SDGs — is that it will also be a useful tool for transportation planning.
— Ana Leticia (@AnaLeticiaGIS) April 17, 2019
The culture shift: Open source for 911
Randal Hale set up a 911 service for Henry County, Tennessee (population 22,000 and home to a killer BBQ fest) using open source. In about a week and using a Lenovo machine the government already had, it’s up and running thanks to QGIS, PostGIS and Geoserver. He reckons that the government saved over $30,000. (Read: mostly not paying for ESRI licenses.) Interestingly, the $360 dollars he tallies as the only cost was for the one piece of proprietary software they used, a data-collection app called Fulcrum mobile.
My calculations on the true cost of the set-up would be slightly different: include the people hours (sometimes higher for FOSS when you’re essentially writing the manual as you go along) and factor in the eventual cost of replacing that recycled machine. Still, this shows that you can run something as important as 911 service on open-source software. And that some governments already do it. Hale emphasized that educating the users was an important cultural shift — and that was what led to more counties asking how they could copycat.
— Matthew Hanson (@GeoSkeptic) April 15, 2019
Lucre isn’t filthy, open source folks
Here’s a talk that needs to be on every open-source conference agenda: “Please sell something: FOSS funding for the next decade of innovation.” Alternative title: Making money with open source, why it matters and how to do it. In a standout session from a schedule packed with very smart people sharing knowledge, Piero Toffanin implored the audience to think about turning a buck. Not every open-source gathering completely overlooks the money question but what would be standard fodder at a “regular” conference (marketing, revenue models, metrics, profitability) can seem a bit, well, base at one focusing on free and open source software. It doesn’t have to be this way!
Making money is one of the ways to convince more people that the model works and that it’s worth spending time and investing in. Toffanin’s own path makes him an interesting harbinger for this message, his is the classic itch-scratching to paycheck that still happens fairly often. A software developer, he wrote installers to keep his own drone from collecting dust in the garage and then open sourced them. Now he’s a core dev at OpenDroneMap, among other pursuits. His take on the basic funding models is especially worthwhile. He sets them out as open core, services, dual licensing, sponsors, donations and crowdsourcing.
After walking through the pros and cons for each, his conclusion is: ditch the donation button and the crowdsourcing stance and just sell something. It could be first-rate documentation in the form of a book or maybe it’s training. Just.Sell.Something. Because people will pay money for the valuable work that you do. It’s a necessary message that more people in open source need to hear. Take a look at his slides for more ammunition.