4 ms·

I just built a vacation home from the ground up and specced the "smart" features myself and did so very carefully. I've learned the hard way because I was alrea

I just built a vacation home from the ground up and specced the "smart" features myself and did so very carefully. I've learned the hard way because I was already "orphaned" when I bought my current primary residence which had a high-end 'state-of-the-art' home automation system that was designed into the walls by the original owner. Unfortunately, it was cutting-edge tech for 2001. The original manufacturer of the system as well as the local dealer/installer both changed owners and product lines multiple times and were out of business over a decade ago. Fortunately, the hardware was pretty well-designed industrial grade gear consisting of cards mounted to back plane slots inside wall-mounted racks in the basement and I like working on 8-bit era arcade and pinball machines which shares many similarities. The saving grace is that the system is just old enough to predate being internet-connected, so the firmware is in EEPROM and was designed to stand alone forever if necessary. In my new place I was determined to avoid the same issues 25 years down the road so I've selected devices that run open source firmware on widely available, multi-source commodity hardware using long-lived open protocols. For example, the 62 in-wall dimmer switches have ESP8266 boards (https://en.wikipedia.org/wiki/ESP8266 https://en.wikipedia.org/wiki/ESP8266) and I installed the mature Tasmota open-source firmware (https://tasmota.github.io/docs/ https://tasmota.github.io/docs/) in them so they don't require cloud connection or try to phone home. They function normally as light switches with no connection of any kind but can be optionally controlled via wifi by a Raspberry Pi 4 running Home Assistant open-source software which hosts modular integrations with hundreds of different devices, protocols and standards.

Spivak 5y ago
I think your solution is good but paranoid in the wrong areas. You’re probably gonna get bit by 25 years from now is the dependency on current gen Wi-Fi. The thing you have to own is the transport layer and all the devices that directly interface with it. Then all you will ever need to replace is the bridge to your “control plane” — apps, HomeAssistant. The golden solution (for me) is RF. 100 years from now you will still be able to find trancievers and the software to use it. It penetrates walls effortlessly and light switches don’t need high bandwidth. Best part is they’re super cheap. Better living through 1900s tech. You can own the whole stack, have it be self-contained, never need to update. No trillion dollar corporation is gonna get distracted and say that Wi-Fi 420.69x-qwerty is so last year and force you to change anything other than the one little segment of your stack.
- est31 5y ago
  I put wi-fi, at least the earlier generations, up there are the list of consumer technology that will always be supported somehow, simply due to its extreme popularity. I can still buy floppy drives even though they became outdated decades ago, just as I can buy serial ports. USB-A will be up there as well, as will wi-fi and bluetooth, as long as governments don't ban wi-fi/bluetooth devices. Wi-Fi runs in unregulated spectrum, which means that no operator has to pay lots of money to governments in order to use the spectrum. Contrast this to 2G and 3G which network operators over the world turn off so that they can deploy the more efficient 5G, which one day too will likely be deprecated in favour of something even better. Multiple wi-fi networks can coexist. In fact, wi-fi already shares its spectrum with bluetooth and probably a whole range of other applications. Even if a company from say cupertino thinks that Wi-Fi is outdated, the spectrum will remain unregulated, and you will likely have some way to bridge from whatever technology the cupertino devices use to that technology (or well maybe I'm a bit too optimist, that cupertino company has figured out tons of ways to lock you in lol).
  - mrandish 5y ago
    > I put wi-fi, at least the earlier generations, up there are the list of consumer technology that will always be supported somehow I agree with you about wifi. In fact, as you were posting, I was writing my reply (here: https://news.ycombinator.com/item?id=29335111 https://news.ycombinator.com/item?id=29335111) saying much the same. I've been observing the evolution of tech professionally now for over 40 years and as a serial entrepreneur making risky bets on uncertain outcomes with the harsh (yet instructive) feedback of putting my own money and livelihood on the line each time. There are a few tech standards that become so ubiquitous and accepted that they reach the longevity and influence of Roman chariot wheel widths (which many anthropologists believe influenced the gauge of modern railroads). Various forms of early wifi (ie 802.11b) and various flavors and bitrates of RS-232-type serial signaling are two examples. I still have a couple of 802.11b wifi devices running on a sub-net that are ~20 years old and I still network with devices over 20 years old through Cat5 signaling with / RJ45 connectors.
- mrandish 5y ago
  I actually agree with your assessment and did think about the issue you mention. I chose this imperfect set of trade-offs because I didn't want to DIY in-wall fixtures and, after reviewing the current tech and market and which "open" standards were being developed, I decided to go with an inexpensive shorter-term solution (~4 to 8 years). The dimmer switches are only $15 each delivered from Amazon (https://smile.amazon.com/Dimmer-Martin-Jerry-Compatible-Neutral/dp/B07Y53VZ17/ https://smile.amazon.com/Dimmer-Martin-Jerry-Compatible-Neut...) and under Tasmota are highly configurable to the extent of being able to communicate in a switch-to-switch private mesh. Since I had to commit to hardware six months ago, I knew the kind of longer-term hardware I'd really prefer didn't quite exist yet - at least not at the multi-vendor, commodity-cost level. There are some interesting high-end, proprietary, single-source products as well as some cool experimental DIY & low-volume kit-type designs, using stuff like touchscreen displays and ESP32-type CPUs. I think this kind of flexible display/interaction end-point will be the most likely next inflection point but it will first need to go through the early-adopter Darwinian process sorts and then come down the price-curve to reach Shenzen-container, multi-vendor commodity pricing. My choices were to either A. Spend too much now making an early bet on something available today that probably wouldn't be a longer-term keeper. B. Go with a traditional "dumb" dimmer for $6.50 each and upgrade in a few years. C. Pay $15 each today on a shorter-term, standard-sized, replaceable "smart" solution that can fall back to a dedicated wifi subnet for control or, worst-case, be a stand-alone dimmer. I decided to go with C. While I agree with your point that the wireless protocol world will eventually move on, I suspect that wifi as a protocol is so ubiquitous, so standard and so cheap that some form of backward compatibility may survive for decades through bridge devices. A likely contender for the next big multi-vendor standard in home automation is likely to be the ChIP-based standard being pushed by Google, Samsung, Apple and a couple hundred other companies. The standard has been finalized and the wireless will be based on both wifi and Zigbee (2.4ghz and 900 Mhz) as a basic required element, so there's a fair chance my current cheap "Shenzen-specials" will be adaptable through firmware since the vast majority of in-wall wireless stuff installed today is based on wifi or Zigbee. It's okay if they aren't because in my mind they are a short-term disposable solution anyway.

Spivak 5y ago

I think your solution is good but paranoid in the wrong areas. You’re probably gonna get bit by 25 years from now is the dependency on current gen Wi-Fi. The thing you have to own is the transport layer and all the devices that directly interface with it. Then all you will ever need to replace is the bridge to your “control plane” — apps, HomeAssistant. The golden solution (for me) is RF. 100 years from now you will still be able to find trancievers and the software to use it. It penetrates walls effortlessly and light switches don’t need high bandwidth. Best part is they’re super cheap. Better living through 1900s tech. You can own the whole stack, have it be self-contained, never need to update. No trillion dollar corporation is gonna get distracted and say that Wi-Fi 420.69x-qwerty is so last year and force you to change anything other than the one little segment of your stack.

est31 5y ago
I put wi-fi, at least the earlier generations, up there are the list of consumer technology that will always be supported somehow, simply due to its extreme popularity. I can still buy floppy drives even though they became outdated decades ago, just as I can buy serial ports. USB-A will be up there as well, as will wi-fi and bluetooth, as long as governments don't ban wi-fi/bluetooth devices. Wi-Fi runs in unregulated spectrum, which means that no operator has to pay lots of money to governments in order to use the spectrum. Contrast this to 2G and 3G which network operators over the world turn off so that they can deploy the more efficient 5G, which one day too will likely be deprecated in favour of something even better. Multiple wi-fi networks can coexist. In fact, wi-fi already shares its spectrum with bluetooth and probably a whole range of other applications. Even if a company from say cupertino thinks that Wi-Fi is outdated, the spectrum will remain unregulated, and you will likely have some way to bridge from whatever technology the cupertino devices use to that technology (or well maybe I'm a bit too optimist, that cupertino company has figured out tons of ways to lock you in lol).
- mrandish 5y ago
  > I put wi-fi, at least the earlier generations, up there are the list of consumer technology that will always be supported somehow I agree with you about wifi. In fact, as you were posting, I was writing my reply (here: https://news.ycombinator.com/item?id=29335111 https://news.ycombinator.com/item?id=29335111) saying much the same. I've been observing the evolution of tech professionally now for over 40 years and as a serial entrepreneur making risky bets on uncertain outcomes with the harsh (yet instructive) feedback of putting my own money and livelihood on the line each time. There are a few tech standards that become so ubiquitous and accepted that they reach the longevity and influence of Roman chariot wheel widths (which many anthropologists believe influenced the gauge of modern railroads). Various forms of early wifi (ie 802.11b) and various flavors and bitrates of RS-232-type serial signaling are two examples. I still have a couple of 802.11b wifi devices running on a sub-net that are ~20 years old and I still network with devices over 20 years old through Cat5 signaling with / RJ45 connectors.
mrandish 5y ago
I actually agree with your assessment and did think about the issue you mention. I chose this imperfect set of trade-offs because I didn't want to DIY in-wall fixtures and, after reviewing the current tech and market and which "open" standards were being developed, I decided to go with an inexpensive shorter-term solution (~4 to 8 years). The dimmer switches are only $15 each delivered from Amazon (https://smile.amazon.com/Dimmer-Martin-Jerry-Compatible-Neutral/dp/B07Y53VZ17/ https://smile.amazon.com/Dimmer-Martin-Jerry-Compatible-Neut...) and under Tasmota are highly configurable to the extent of being able to communicate in a switch-to-switch private mesh. Since I had to commit to hardware six months ago, I knew the kind of longer-term hardware I'd really prefer didn't quite exist yet - at least not at the multi-vendor, commodity-cost level. There are some interesting high-end, proprietary, single-source products as well as some cool experimental DIY & low-volume kit-type designs, using stuff like touchscreen displays and ESP32-type CPUs. I think this kind of flexible display/interaction end-point will be the most likely next inflection point but it will first need to go through the early-adopter Darwinian process sorts and then come down the price-curve to reach Shenzen-container, multi-vendor commodity pricing. My choices were to either A. Spend too much now making an early bet on something available today that probably wouldn't be a longer-term keeper. B. Go with a traditional "dumb" dimmer for $6.50 each and upgrade in a few years. C. Pay $15 each today on a shorter-term, standard-sized, replaceable "smart" solution that can fall back to a dedicated wifi subnet for control or, worst-case, be a stand-alone dimmer. I decided to go with C. While I agree with your point that the wireless protocol world will eventually move on, I suspect that wifi as a protocol is so ubiquitous, so standard and so cheap that some form of backward compatibility may survive for decades through bridge devices. A likely contender for the next big multi-vendor standard in home automation is likely to be the ChIP-based standard being pushed by Google, Samsung, Apple and a couple hundred other companies. The standard has been finalized and the wireless will be based on both wifi and Zigbee (2.4ghz and 900 Mhz) as a basic required element, so there's a fair chance my current cheap "Shenzen-specials" will be adaptable through firmware since the vast majority of in-wall wireless stuff installed today is based on wifi or Zigbee. It's okay if they aren't because in my mind they are a short-term disposable solution anyway.