The Death of the Smart Home
Photo by W A T A R I on Unsplash
Wait a minute. The smart home is dead already? Apparently.
I had decided to look at some ways to “smart up” the household. I thought this would be a fun project to tie a number of existing products I already owned together, maybe do something to pull in stats from our solar panels on the roof, and tie in the new home weather station I wanted to buy. Wow, was I wrong. Granted, this wasn’t an overnight thing, as I’d looked into this topic off and on. But opportunities in the form of industry improvements just weren’t there, despite several years of improvements in nearly every other technological area. Phones improved. Coding standards seemed to rise with the increased adoption of DevSecOps principals. Zero Trust Networking was actually becoming a real thing. There were even pockets of advances in the Internet of Things world. But the smart home is simply not catching on.
The Problems
There seems to be four main lumps of smartness - things that talk to Amazon Alexa, things that talk to Google Home, things that talk to the cloud, and things that were considered smart but might only talk to a phone app. Some things were in more than one of these groups.
There are Wi-Fi, Bluetooth, RFID, NFC, and Zigbee to talk to something, with a lot of devices using more than one of these protocols. Some only speak to a phone running a special app. Some devices required a hub, some do not. There are some devices that can possibly talk to the cloud, but only if you run special code on a dedicated secondary device (think Python or Go on a Raspberry Pi). But should you have devices that have no common protocol and no way to introduce a secondary communications channel, well, then these smart things can not talk to one another.
Some of these devices are security nightmares, a few are fairly locked down. Some allow for a myriad of configuration choices, some allow for none. Some update automatically, some can be manually updated, and some can only be updated by replacing them with a brand new device.
Death Reasons
I follow Stacey Higginbotham, and when she mentioned that the smart home is dead and cited a couple of references (mainly related to vendors re-branding and things simply not getting any better), it got me thinking about some of the reasons for this death. I have a number of non-technical friends who have talked about their love/hate relationships with technology, and based upon their experiences and a bit of research on my own, three coffin nails immediately lept to mind.
Accuracy. There is a distinct lack of accuracy in voice recognition and translation. This has more to do with the use of a natural language interface than recognition of individual words. Most traditional interfaces on an app or browser use a combination of pointing and clicking or selecting items from a pull down menu - maybe a choosing from a list with pictures. In other words, there is typically a choice with finite options, and when we do use natural language we are typing in a single word or small phrase into a search engine. The lack of feedback from using Alexa simply pales when compared to searching and selecting via a browser on a laptop or phone. It’s primitive. Alexa is good for the most simple of tasks such as playing music or answering questions about the weather. If one needs to do anything more complex, one can get it done ten times faster and more accurately from a phone app.
Privacy/Security. I lumped these together because most of my non-tech friends think of them as one thing. Regardless, perception of privacy violation and security issues is fairly well embedded in the consumer mind when it comes to smart devices. This perception is not helped by repeated news stories of devices phoning home with personal data, and tales of engineers listening and viewing audio and video data. We live in a post-Snowden world of compromised nanny cams and smart device botnets, so many consumers will allow for very few smart devices, or just stick to very simple routines they can get a mental grip on - all because of privacy and security concerns.
Complexity. Just getting two devices to talk to each other successfully can be a challenge. Maybe you get device one to talk to device two. Adding device three requires an upgrade to device one, and now device two can’t talk to anything. You update device three and now nothing talks to each other. The rumored next release of device two is supposed to fix things, you check their website and wow, they’ve gone out of business. In a world where sometimes you have two or more remotes to watch television, combining flaky implementations of communication protocols with a natural language interface that can easily get a word wrong, and many people will just say “Never mind, I’ll stick with my old media system and my two remotes, I can’t become an engineer just to live in my own home.” I’ve seen a power outage take out smart home devices for days before the magical sequence of booting, re-initializing from scratch, and various verbal commands gets communication going again. Often a non-technical person will simply give up and remove devices in frustration. Trust me, I have a pile of them given to me by non-technical friends.
There are certainly other reasons as well. I mean, it is hard to convince someone they need a smart water bottle or a smart toaster that prints a weather report on your morning toast, so there is often no good reason to own the tech in the first place, let alone actually integrate it with the same system that controls mood lighting.
The Biggest Reason
Perhaps the biggest reason (IMHO) the smart home is dead is not just because the industry isn’t mature, it’s because there are areas of actual success elsewhere. The simple experiments known as the smart home were experiments with devices on the edge, and when you broke them down into simple and repeatable tasks where a machine processed data and then sent it over a communication channel to another machine (such as the cloud), then you had the perfect environment for this current level of maturity - IIoT, or Industrial Internet of Things. This is not to say that IIoT is perfect, or more secure, or a better performer. But a nice consumer interface on an edge device is not a priority, and they are usually built using an existing model.
For example, in industrial environments there are devices that already do monitoring out on the network edge and beyond, and IIoT allows those devices to lose a few restrictions. Instead of just collecting raw data and uploading, the IIoT device can do calculations on the edge and upload results. The IIoT device can function wirelessly and even operate and gather data while offline, and upload its data when it eventually reconnects. This in itself opens up all kinds of opportunity for the overall application of a system strategy for many companies, and due to lowered cost from construction of the devices with what amounts to off-the-shelf parts in many cases, IIoT has certainly been a success. This is especially true by comparison to the level of adaptation of the full capacity of smart home devices. IIoT devices are not only being used to the fullest, I would argue all of the innovation of smart “things” is occurring at a higher pace in the IIoT world than the IoT world.
In Summary
We’re not there. Smart home version 1.0 was a fail. Maybe the IIoT world will make advances that will provide an innovative foundation for smart home version 2.0, but there are serious changes that need to happen before there is serious full-fledged adoption of smart home technology. Remember, if your smart home interface can be replaced with phone apps and Siri, it is not ready for the smart home world we’ve been promised.
