‘Remix it and let it evolve’: Q&A with FieldKit developer Shah Selbe

Shah Selbe and his team were all set to unveil their new product in April 2020. The plan was to launch FieldKit, an open-source environmental sensor platform for conservation researchers, to coincide with the 50th anniversary of Earth Day. Hundreds of FieldKits were ready to go. And then, just over a month ahead of the launch, the World Health Organization declared COVID-19 a pandemic. Before Selbe knew it, flights were grounded, travel restrictions were imposed, and the world effectively shut down, putting a damper on his plans and any prospect of fieldwork.

Then, almost a year later, when the situation seemed to be improving, a new problem arose. The transportation hurdles, labor shortages, manufacturing backlogs and increased consumer demand during lockdowns had caused a global supply chain crisis that continues to snarl manufacturing and trade even now. For Selbe, the shortage of parts required to build his product further slowed down his plan to launch FieldKit.

The ongoing supply chain crisis has had an impact on almost every aspect of the global economy. Brands big and small have had to reimagine their workflow. Ikea hiked prices because of shortages caused by the crisis. Automobile giants have had to slash production because of a shortage of semiconductors. The crisis even forced McDonalds to take milkshakes off its menu in the U.K. and ration french fries in Japan.

Makers of conservation technology have been equally impacted by the shortage of parts, making it difficult for many to develop and manufacture products that they had conceived of before the pandemic. Selbe, a National Geographic explorer, says his team had to redesign their products owing to trouble getting their hands on important parts — from a chip used to study power consumption to a barometric sensor to measure pressure and altitude — all of which, he says, disappeared from the market.

As a modular system, FieldKit is meant to be a “mix and match” platform, Selbe says. Users kit it out with the sensors relevant to their specific application — air quality, camera traps, acoustics, and more — and program it to relay back that data at a frequency of their choosing. “It’s really quite varied,” Selbe says.

Mongabay’s Abhishyant Kidangoor spoke with Selbe about the impact of the supply chain crisis on his work, about FieldKit and Conservify, the conservation tech lab he founded, and about the future of conservation technology and why it has to be open source. The following interview has been lightly edited for style and clarity.

Mongabay: Have you always been interested in wildlife conservation?

Shah Selbe: I have always been very environmentally minded. It was something that I have always had a love for. I always saw my role in conservation, and did some volunteering here and there, or maybe I would go on some vacation and do some good work during that time. I went to school for engineering, and it wasn’t until graduate school where I started to understand the need for upgraded tools in conservation, and the role that technology has in that.

Mongabay: How did you go about doing it?

Shah Selbe: Back in 2015, I left my traditional engineering corporate job to start Conservify, which is probably the first solely conservation-focused maker space and tech prototyping lab that anybody had come up with. We have worked on projects all over the world, and different types of ecosystems — from rainforests to glaciers. We partner with organizations that are already doing really great conservation work on the ground, but could really kick up their capabilities if they have better access to tools that allow them to lower costs or [improve] fidelity. So that’s the work that we had been doing.

And then, the constant need for better environmental sensors is what drove me to start FieldKit, which is really focused on making sensors that can be used by conservationists in the field. But we designed it in the context of what modern electronics allows us to do, and we interfaced the sensors with smartphones.

Mongabay: Are there any specific ground-level examples you would like to share?

Shah Selbe: The projects have varied pretty greatly, because it depends on what the need is. We have used drones for coastal monitoring, and for assessments of whale health. We have built apps that work with citizen-science approaches to monitoring marine protected areas. We have worked with Parks Canada and the University of Edmonton to set up a glacier monitoring station within Banff National Park. We have a project with UCLA that’s focused around creating a radio network that can collect open data from anybody who’s doing any sort of research in that area, and then bring it back so anybody can have access to it.

We have a project with the Wildlife Conservation Society in five of the countries that make up the Amazon Rainforest. And that’s to provide FieldKit devices to communities who want to measure how water quality is changing over time and the level of the water as it changes, and to also help them study and think about the fisheries happening in the region. So it’s really quite varied.

Mongabay: How has the COVID-19 pandemic affected your work?

Shah Selbe: This is a huge issue. It has been very difficult. I will tell you in the context of FieldKit because I think what we saw is similar to what a lot of people saw. So originally, in 2018-2019, we did this big redesign of what we were building with FieldKit, and we were planning on unveiling it in April 2020 for the 50th anniversary of Earth Day. When the pandemic hit, we put things on pause for a little bit. It was too close to this massive global anxiety everybody was feeling. So I personally felt it was a little irresponsible to try and launch a product while people are worried about their lives. And so we decided to sit on it, and not open the store. But we continued to support other projects here and there. But because of the pandemic, a lot of field programs with travel ended. And so a lot of the work that we were supporting — NGOs, scientists, citizen scientists and community members out in the field — nobody was going to the field anymore. So the need for these tools just wasn’t there. And so we gave a bunch of FieldKits away. Then that ended up being a bit of a logistical nightmare because we started to see impacts on flights, and getting things to different places was very difficult. It was a huge headache.

Mongabay: Then the supply chain crisis came along. How did that hit you?

Shah Selbe: When we started to feel like we had gotten over that first wave of anxiety for COVID, and maybe people would be interested in at least getting these in their hands, playing around with it, though not necessarily going out into their study sites and researching it, we tried to get the product out. But that was around when we started to see the supply chain impacts happen. In my team in FieldKit, there are a number of folks that are pretty well connected to the hardware community. They started hearing little rumblings from folks in Shenzhen and some other places that things are starting to get hard to buy.

As things started to collapse, we decided to make some strategic purchases to allow us to keep on going, in case things get worse. We are kind of still in that weird state. Some of the stuff that we paid for almost a year and a half ago, we still haven’t received. It was so bad that our main electronics engineer would see things disappear from his cart. He will be trying to buy 1,000 of something. Then all of a sudden, things that had a stock of 50,000 are gone. It’s because the Apples of the world come in and just scoop all that stuff up.

Mongabay: How did the crisis make you rethink FieldKit?

Shah Selbe: It really put us in this position where we had to think a lot about the design of our products. I think it helped that we already were doing a good job at choosing the correct components and thinking a lot about design for manufacturing. We just had to put extra thought into all the upfront design stuff. We would have been hit a lot harder if we were just using reference designs.

But we had to change some boards. We had to redesign certain parts, like our power system for different things, because we couldn’t get those parts in the market. We have an underwater version of FieldKit and we thought we were ready to build the first five of those and put them into testing, and then one of the parts just disappeared. So we had to do a redesign.

Mongabay: What’s your biggest worry?

Shah Selbe: If we were smaller, it would be a lot more painful. Because a lot of these conservation technology grants that come out, they are not necessarily big. And, if you have invested a lot into a certain design and find out that you can’t build that design until 2024, it’s hard to keep a team going for that long without a proven product. So I worry about the smaller conservation tech projects out there that are probably in a dormant state right now, because they can’t quite last through this.

Mongabay: Are things getting better?

Shah Selbe: It’s not as bad as right in the beginning. But it feels like we are still very much in the middle of it. There’s less panic buying, which is good because that was really hurting a lot of people. But the problem is these big companies who have big supply chain budgets. If they bought a bunch of stuff, they are not going to put that stuff back out on the market and are just going to hold on to it. So right now, we just have to wait for factories to build more.

Mongabay: Given the current state, do you anticipate any of your projects being stalled?

Shah Selbe: The model I wanted to build with FieldKit is that of an open-source, nonprofit organization, where we can build and manufacture hardware, and we could sell it and that revenue will help augment the grant funding that we get for other things. And I wanted this to be a proven model that other organizations can copy — this sort of approach of selling products, and using the revenue that comes in to help fund the organization. What I wanted to prove with FieldKit is that not only can we build these sensors, but we also have this new approach to doing things that other people can replicate and do as well.

But I haven’t been able to test that side because we haven’t been able to sell a FieldKit. Right now, all the FieldKits that we build are going out to specific projects we have with different organizations, and they are being deployed through there. But I would like to get to the point where I can build, say, 500 of these FieldKits, and some of those will go to projects, but then some of them will be for people to buy and do interesting things with.

Mongabay: Has inflation affected you?

Shah Selbe: We haven’t dug too deeply into it, because we haven’t really gotten to selling things regularly to regular customers. Once we can open up a shop, we might have to go back and reassess how inflation has impacted us.

Mongabay: What was your biggest learning experience going through this process during a pandemic and global recession?

Shah Selbe: Don’t build a hardware product during a pandemic. [Laughs.] I am very proud of everything that the team has done and what we have built. But it has been really difficult to make sure that we can last through all of this to see the fruits of our labor.

Mongabay: What advice would you give to people who are setting out to do what you did?

Shah Selbe: Spend some extra time trying to understand what the supply chain looks like. Our lead engineer really understands how these companies work, and he has talked to a lot of people and read up a lot about it. That really allows us to make the right decisions about what we are putting inside of our product because there’s so much volatility on what’s available, and what’s not available. That will allow you to really build what you want to build.

It’s also important to remember that if you are building something during a pandemic, it’s not going to be your ideal final solution for that design. Maybe you couldn’t get something and you will have to go around things to find a way out.

Mongabay: What does the future of FieldKit look like?

Shah Selbe: Right now, we want to get to the point where we are not only supporting our projects with partners, but we are also making FieldKits available to anybody who wants to buy. The other thing is, because we are open source, I want people to be able to take FieldKit and mix and match and create some version of their own and build on it. That’s something I want to see more people doing — just taking little pieces of this and remixing it into what they want it to be and letting it evolve.

Mongabay: Broadly speaking, how do you think conservation technology is faring?

Shah Selbe: I think we are in a pretty good place. Every new technology gets caught up in this hype cycle, and I think that has a big negative impact. If we take drones, for example, they are very useful for a lot of conservation causes. But people were thinking that drones were going to solve everything, and so there’s a lot of places where they took drones, they didn’t think through the problem, and they didn’t eventually get anything. And there’s now a bunch of broken drones that ate up people’s budget. I think that has happened enough that the folks in the conservation world now know that we have to be a little bit smarter about how we think about applying technology.

There’s a lot of interesting projects, and new technologies rolling into it. All the stuff that you see that’s happening in the machine-learning space is amazing, but I think we still have a long way to go to understand it. I would say we are kind of on the declining side of the hype cycle around AI-related stuff. And that’s because there have been a lot of proven tools that have come out of that space. I think the hype cycle is still on its way up for things like eDNA. Blockchain, I think, is also much earlier in the hype cycle. We haven’t fully seen anything come out of that that I would say is a clear success.

Mongabay: Finally, what do you think needs to change in the conservation technology space?

Shah Selbe: One thing that I still see conservation organizations struggling with is trying to develop a technology with a team that’s not really well-suited to do that sort of thing. So there should be more partnerships between existing tech companies and conservation organizations or NGOs to make sure that the people who are building the tech are the ones that actually know how to build the tech, instead of trying to do everything in-house. We need to move toward more open sharing of information. FieldKit, for instance, is 100% open source. We are open-sourcing everything, even the business plan and our legal documents. So if people want to copy our approach, they can do that. And I think that’s an important way in which we can move forward.