i think people who repair and make stuff hate these sorts of things because they introduce a layer of complexity that ruins longevity and predictability. what you call convenience, i call added costs down the line, and unreliability. also, the "time saved" is a bit like saying modal editors are good because you can enter text more efficiently. its all technically true, but seems irrelevant to me (and i use vim exclusively). but i have to concede that this idea of IOT home convenience stuff may be legitimately useful for some people, like yourself. it just seems to be the case that thats only true for some people.
> introduce a layer of complexity that ruins longevity and predictability.
These are more growing pains than anything else. Adding a HomeKit appliance is absurdly simply these days (you just scan it with your phone) and then it just works.
> the "time saved" is a bit like saying modal editors are good because you can enter text more efficiently
It’s not an insignificant amount. If you live in a highly connected smart home, you will definitely notice the saved time and you will definitely notice how you simply have to do less things. There’s a sort of lightened cognitive load that’s difficult to describe. Things just feel more chill when you can adjust conditions to perfection without lifting a finger.
There is a level of internal complexity implied within "it just works," and an added layer of frustration when someday down the line, it just doesn't work. Eventually an electronic component wears out, or the manufacturer shuts down their server, or your phone is no longer supported.
I have always loved the idea of a connected house, but I haven't yet seen a feature that would make it worth the trouble of installation, added cost, and yet another app to control my life.
Curious if, in ten years, the HomeKit appliances of today will still work. Or will you have to find a decades-old phone to control them?
It's always interesting to see vestiges of a home intercom in really old homes, the phone jacks in bedrooms, ladder-style wire for TV aerials mounted in the walls, bits of some kind of home-alarm system....
I don't know specifically about HomeKit, no experience with it, but many of the underlying standards are "old". E.g. Zigbee is 20+ years old and was standardized in 2003.
Most of these systems today also let you put in place one hub per standard you care about, and then just update devices, and you can mix and match. Some, like Zigbee, are standards supported by multiple manufacturers. The radiator thermostat interface I rely on is standard for UK radiators at least, and is physical and pre-dates electronic thermostats by decades - the smart thermostats physically press down on a switch on the radiators; to replace them with another standard - or a manual one - you just twist the old one off and twist a new one one, done in ~2 minutes.
They similarly communicate with a hub, and an optional central boiler thermostat. The boiler thermostat is the only part that took any time to figure out (but there are plenty of installer services available), and the thermostat interface for boilers in the UK at least is built so that anyone can build replacements; you just need something that can switch a current.
Keep in mind that the only thing that is new here is the network enabled smart control - many of the interfaces are already really old and established.
I have ~20 Zigbee lightbulbs. I also have some LightwaveRF switches. As I refurbish and/or as the Zigbee bulbs die, I'll slowly replace the Zigbee lightbulbs with Lightwave switches as they are more convenient. But the difference is small enough that I've not bothered going over the house to replace them all. If either one of them become hard or expensive to come by, I'll just at that point add a hub that can bridge between a new standard and whatever device I use to interact with them. Setting up the Lightwave hub took me less than half an hour. If I have to invest half an hour to add a new type of hub every 10-20 years, I'm fine with that.
In this case, if Lightwave "dies" entirely and becomes impossible to support, I'll lose the smart functionality of those switches, but nothing else is hard-wired (the Zigbee setup is all mesh between the bulbs themselves, and a hub built into one of my Alexas, and they too require no extra wiring that a normal switch doesn't, and fall back to working like a slightly fancy normal switch if the network is unavailable, and someone unaware of Lightwave wouldn't know anything was broken. And that's what I'd face while transitioning to something new.
I don’t know how to measure but my intuition is choice support bias is at play here and from many proponents of “highly connected” homes. HomeKit seems to be nearly as fiddly as roll your own solutions and the cognitive load for additional debugging and layers of complexity seems to outweigh the gains.
But these are just two points of view.
well, what your saying is not insane. but i would think id need to see these products in use for another 5-10 years before id feel like they were worth the trouble. i.e. the end of life cycle is kind of uncharted territory. is apple just going to brick everything like they did with their old phones? are they going to crap out because of power surges? etc...
