Famous sci-fi author Bruce Sterling said it well in his non-fiction essay The Epic Struggle of the Internet of Things:

The internet of things … is neither about internet nor things,
 but a battle … for the control of property.

Too dramatic for you? Well, it’s prime holiday shopping season, and so I went browsing on Amazon.com for “home automation” products. Of the 23 products that showed up on the first page for me, 17 had, in their very title, a variation of “works with Alexa, Apple HomeKit and Google Assistant”. And exactly none of them indicated that I could open it, take it apart, repair it, connect it to something I could freely choose, or to the rest of my house. Only to Alexa/Amazon, Apple and Google.

This is very different from pretty much anything else I attach to my house. I buy a faucet: it attaches to any pipe I might conceivably have and if not, I can buy an adapter or apply a vise or blow torch to it. I buy a stove, a table, a banister, even a window or a garage door: it’s mine, there are no backdoors, no unnecessary third parties are in the loop, I can take it apart and change it any way I like. Is there any home owner who hasn’t fixed things in a way that the manufacturer definitely didn’t forsee, and probably wouldn’t condone?

Those “home automation” products are a black box, which talk to a third party (Google, Apple, Amazon) that I did not invite into my home, I have no idea what they do, and I have no control over. If I installed them in my house and then said, “Alexa, turn on the heater”, and Alexa responded, “I can’t let you do that, Dave”, there is nothing I can do.

We need to do better. We need internet-of-things products to be like other things we buy for our homes:

• controlled by us, the customers and users, not by some would-be tech overlord in the cloud. This means we need to know what’s inside, how it works, and have the ability to change it.
• not have back doors for third parties we did not invite. Sure, service providers like a maid service are nice, they come and do work we don’t want to do ourselves, but I sure will not hire one that I cannot fire without tearing the house down. Why should technology products be any different?
• not spy on us. No, there is no reason to ship off the data from my thermostat to somebody else’s cloud before I can get at it (if I can get at it at all). What happens in my house needs to stay in my house, otherwise it’s somebody else’s house, not mine, and I’m just a guest. Bruce Sterling quote, anyone?
• make all the collected data easily available to us and to nobody else. This should go without saying, but sadly, it does not.
• be friendly to the do-it-yourselfer, by making it easy to install and connect to anything else I might want to connect it to. Not just products of vendors who have business deal with one of the big platform vendors, and they aren’t currently suing each other. And of course:
• can be supported in the long term. Do we really think that any of those tech giants are going to support their IoT products for the life of my house? If not, why would I attach such a gadget to my house?

I call this better, customer and user centric form of the Internet of Things the “Indie IoT”. “Indie” as in independent, grass-roots, not beholden to some overlord.

Personally, I will never buy “Overlord IoT products”. Their downsides are much bigger than their advantages, in my opinion and the opinion of many people I have talked to.

But I would buy “Indie IoT” products, and I have: for example, I have the excellent IotaWatt Energy Monitor in my garage. It monitors my house’s energy consumption, it’s entirely open source hardware and open source software, and has no backdoors. All the data is easily available to me, and to nobody else (well, unless I explicitly send it somewhere else). And guess what: I can hack it like anything else I buy at the hardware store for my house.

I don’t think the IotaWatt describes itself as an Indie IoT product, but it meets my definition. I would buy other because the IoT could be fun, and useful besides! And I would build more to solve problems that I have in my house, if it was a bit easier to connect them to each other. So I think perhaps it’s time to try and connect some of the developers of Indie IoT products together, and come up with some more integrated solutions rather than point projects/products.

It should not be too hard to come up with some basic agreements that allow for more interesting use cases to be supported. For example, my Raspberry Pi-controlled ceiling fan, as well as my pool timer, could use some buttons or switches to control them in addition to what I can do over the web. There must be somebody out there who likes to build “Indie IoT buttons”, and we should be able to agree on some simple protocols (e.g. via the Web of Things) so we can connect, without either of us having to reinvent the wheel or go through somebody else’s cloud. Or a display that I can mount in the hallway that gives me status information. Or some motion sensor that causes the pool pumps to turn of when somebody is in the back yard. Or …

Anybody up for that kind of thing? Reach out, or better, blog your thoughts and ping this blog.