Lessons learned from a baby monitor
The internet of things and product design philosophy
You can’t talk about the internet of things (IOT) without talking about a Wi-Fi toaster. It’s a contrived example that is both exciting and absurd. Interconnect all our devices to make them smarter and in turn make our lives better. When does a toaster stop being a toaster? It’s not a trick question; it stops being a toaster when it stops reliably making toast.
The philosophy of Unix software design states “Do one thing and do it well”. I believe this concept holds true when discussing complex hardware devices. Take the iPhone for example, you hardly think about it, but it was built on the foundation of making phone calls. Barring a few minor hiccups along the way it continues to do this extremely well, and all of it’s other features builds on top of that core functionality.
I recently purchased a high end Wi-Fi security camera. As a brand new father and home automation junkie, I wanted the baby monitor of the future. I was excited with the notion I could check on my daughter with any screen in the house, including our television. I was giddy with joy at the thought of receiving real-time push notifications on my watch so I could soothe my little girl without waking up my exhausted wife.
The camera touts a rich features set, which on paper make it THE device to buy. Who wouldn’t want an air quality sensor, If This Then That (IFTTT) support, white noise generator, multi-coloured night light, and a dedicated baby monitor mode?
Unfortunately it doesn’t do any of these things well. More to the point it does them poorly. This might be excusable if the camera itself worked flawlessly, but it doesn’t. Constant Wi-Fi drops, crashes, and a buggy app plague the entire experience.
More than once we’ve woken to find a crashed app and/or camera, and a crying baby in the other room. When does a Wi-fi camera stop being a Wi-fi camera? When it stops reliably streaming images.
This speaks volumes about the mentality of modern hardware companies. If the goal is to disrupt a market, it seems they are attempting to do so without understanding the markets core needs. Baby monitors, like security cameras, smoke alarms, and other safety devices need to work reliably and consistently.
To contrast the experience with the Wi-Fi camera, we purchased a simple inexpensive baby monitor. It uses the equivalent of cordless phone technology to maintain a constant connection between the units. Comparatively, it offers a small fraction of the features. What surprised me is how well all of them worked. The product does one thing extremely well and every additional feature is there to improve that core experience.
Products that “just work” don’t need to be reinvented, they can be incrementally improved. I wrote an article about Apple Homekit hacking which followed this design principle. I took an existing product that worked exceptionally well for controlling power outlets and improved it by adding internet connectivity. A whole world of control opportunities was opened up to me, but at no point did it compromise the core functionality. As a result I will always be able to turn on my lights, even if I don’t have my phone handy.
Calling a product “smart” shouldn’t mean it’s reliant on the internet, it should mean it is enhanced by it.
Our Wi-Fi camera is not without merit. Updates are being released occasionally that improve the speed and stability, but many features are still buggy or outright broken. We will continue to use it, but won’t rely on it as our baby monitor. That job goes to the one that reliably makes toast.
