The Things Network launched a Kickstarter campaign last week and looks set to easily meet its funding goal in the first few days. But what’s it all about? What’s an Open IoT WAN and why’s it important? And how can we avoid the trap so many telecoms operators fall into: “Build it and they will come”?
Guerilla Sensor Networks
At Love Hz a few years ago, inspired by Japan’s Safecast radiation monitoring network, I coined a term “guerilla sensor network”. People could connect their sensor devices (flood sensors, temperature sensors, air-quality sensors, bikes, location trackers?) to a free-to-use network as part of a grassroots deployment rather than a managed, organised top-down telco approach. At the time I had TV Whitespace in mind for the network, but the already mature ISM band has emerged as a good bet to getting this off the ground.
The technology for LPWAN is now starting to reach costs which make this possible. This is what you’re seeing in The Things Network. It’s the ideal kind of technology to enable something like Flood Network, which incorporates crowdsourced sensor data to detect flooding and water levels.
Armed with unlicensed spectrum, with a device which can communicate up to 10km and a set of sensor devices what would you do?
An open access low-power wireless Internet of Things (IoT) network is something I strongly believe in. Love Hz was founded on the promise of open spectrum and the explosion in use of wireless devices for IoT.
The Things Network is riding an emerging wave of Low Power Wide Area Network technologies. In their case they chose LoRaWAN, but similar technologies include N-Wave, Weightless-N, and SIGFOX, all designed for securely communicating a few bytes between sensor devices and their applications at low power and long range.
It’s the first time an open Low Power WAN (LPWAN) network for IoT has been coordinated, as opposed to just offering some hardware. I think this is one of the reasons that’s getting a lot of people excited. The Kickstarter campaign aims to create a low-cost gateway to democratise access, but the Foundation also intends to implement the technology stack and infrastructure to support IoT messages from the field through to your applications.
However, the biggest unanswered question remains the same as ever. Who will pay for the infrastructure? How do you make a free network in a sustainable way when physical infrastructure has overheads? Wireless telecoms hardware needs power, backhaul connections and high places. Roofs are off-limits to most building occupants. Landlords and management companies are elusive. Mobile operators have deep pockets to get in there first. Someone has to take this cost on to let end users join the network.
Perhaps there’s a local community funding approach which can help to ensure sustainability. It‘s convenient, then, that radio coverage footprints roughly correspond to the boundaries of communities, and that urban coverage areas of LPWAN technologies probably include enough users to fund their existence.
Ultimately, though, this is not peer-to-peer or mesh, it’s a tree topology where the ownership and resources become concentrated as you approach the core and those costs have to be covered too. Traditionally the small subscriptions of a network operator cover the big costs of a central network’s infrastructure and interconnections over several years.
In a discussion on the ThingsCon Family Slack channel the Guifi model was suggested as an operating model which works. I’m not convinced these networks are the same thing. With Guifi the network infrastructure is the commodity being offered. In The Things Network’s case it’s the ability to connect an application.
Concerns have been raised about tragedy of the commons, lack of SLA etc. I’m not sure we’ll get to the point where we need to worry about those issues in this incarnation of an open IoT network. I suspect this will be the first in a wave of similar ideas. A hybrid model of installing gateways for open connection alongside a guaranteed commercially viable service may help to subsidise the coverage and improve sustainability.
For the time being we can build out some open infrastructure, grow the ecosystem and build support around a standard to see business models emerge.
It’s easy to get caught up in technology lust. We have to remember the applications are the things that will make this sustainable. Ask your average resident if they’re excited about an open IoT network and they’ll probably stare at you, confused. Ask them if they’d like to know about air quality in their neighbourhood, neighbourhood solar generation stats, accurate up-to-the-minute information about floods, or find out which bin to put out tomorrow they’ll more than likely have an opinion.
Get out there and find out what your community care about. Then build that. Then they‘ll come.