If you have a smart home, then you are one of the pioneers of the future and your experience matters!
My name is Sarah, and I am a business student at the Vrije Universiteit of Amsterdam. For my bachelor thesis, I am looking for smart homeowners that would like to share their user experience. The purpose of this study is to gain new insights into the value creation of digital innovation.
The interview will be virtual, for research purposes only and will take no more than 30 minutes. Are you interested in sharing your experience? Please fill out this form , so that I can contact you and arrange a conversation.
My home is plain stupid, I have to give it precise instructions for everything. A really smart home would learn and know when to open the blinds, turn on the lights, based on my behaviour.
right, I think @balloob explained that once ā¦ hitting a switch on the smartphone to turn on the light is just to show off ā¦ not a smart home. Same for āok google, turn on the lightā ā¦
you need those things as a base for automations, but they are not smart on its own.
Iād say that some of my automations are somewhat āsmartā, e.g. my irrigation system measures the amount of rain over the last 24h and checks the expected high temperature of the day. Those two factors combined will determine the duration of todayās irrigation.
Or turning on the light of the bedroom after 22:00 will send me a message on my phone if one of the doors is still open or my media server is still running.
That is true, utimately optimal machine learning techniques would truly improve the home experience. But in the meantime, Iām really curious in what ways you use your devices to operate, as you call it, your stupid home
Thatās a really good point. I think the word āsmartā is often misused for marketing purposes, or what not.
But Iām really impressed with your irrigration system, Iāve never heard of that before! Did you build this system yourself?
I agree with this, though I donāt really see a nearby future for any machine learning for home use. Just because the events we create while being at home is a very small pool of data to work with for machine learning. The amount of times the machine will be wrong in itās āassumptionsā will be so great that itās not practical. Though humans tend to be predictable 80% of the time, the 20% is going to irritate too much to keep going with the machine learning.
True AI is a whole other story though. AI can work (if itās real AI) everywhere. But that is far more complicated and most of the times confused with complicated forms of machine learning and wrongfully named AIā¦
Time calculation is fairly simple, grab max temp: in line 66, and grab the raintime line 72.
There is a lot more, because the pump requires quite a few steps (turn on / wait / turn off / wait) to get solid pressure and after that it will go through the different sprinklers ā¦
But Iāve seen comparable scripts in this forum as well.
JKW
Thank you for your input! This might be a really cool summer project. I already planned on working on my garden this summer. Your code seems easy to follow, but Iāll definitely have a look at some other scripts after I finish writing my thesis. I really hope you are intrested in doing an interview as Iād love to discuss this topic from a user-centric perspective
Thatās a good point. Mainly because the data will be different in every household, and within every house, there may be different users - these circumstances make it extremely tricky to apply machine learning effectively. And youāre right; real AI would be better, but this may take a while until itās well-functioning.
Disclaimer: Iām not an expert, but I assume your devices depend on the user to add the āsmartā element. Which to me, is a very intresting topic to study.
