Live data from Hacker News

Home Assistant 2023.11

home-assistant.io

131–140 of 195 posts

Re: Home Assistant 2023.11

#131
The only negative thing I can say about HASS is that getting it running in Kubernetes with a USB Zigbee/Zwave adapter is a real pain. I've got a local cluster and had it working once, but had to rebuild everything for some unrelated reason, and haven't been able to get it consistently working since. There are too many in-built assumptions about how you're running it and the USB access. Documentation assumes that _maybe_ you're using an old version with some Helm chart.

I'm normally pretty good at getting anything running in k8s quickly, but this is difficult!

I'm going to say "screw it" and just run it bare on a Pi soon instead.

Re: Home Assistant 2023.11

#132

Earlier quoted context omitted.

Who cares and there was no such switch away from yaml

https://www.home-assistant.io/blog/2020/04/14/the-future-of-... Just one relevant quote: > Any new integration that communicates with devices and/or services, must use configuration via the UI. Configuration via YAML is only allowed in very rare cases, which will be determined on a case by case basis.

It’s misleading at best. You can configure literally everything via yaml including your dashboard. Mine runs custom js and css for crying out loud. Even if they changed some options that can’t be configured via yaml which I haven’t encountered, what’s the actual damage to you?

Re: Home Assistant 2023.11

#133

Earlier quoted context omitted.

I don’t understand the negativity at all. HA rivals or exceeds commercial automation systems like crestron, RTI, etc. clearly those people don’t have any industry experience. I’ve generally never had an issue with HA beyond my own fault, like a buggy integration.

Home Assistant is an amazing project. I use it a ton, and I deeply appreciate what they've done. HOWEVER, Lovelace specifically is pretty terrible. Having used Grafana extensively, trying to use Lovelace to arrange a UI is so far backwards that it's constantly grating to try to create or modify dashboards. I really wish there was a way to use Grafana as the UI and embed cards as panels. The ability to drag and drop,…

That’s true. Lovelace needs js and css to look pretty. Most dashboards are terrible. I hope they dump it

Re: Home Assistant 2023.11

#134
post #115
post #100

I love HA but hate how manufacturers are going all-in on preventing people from integrating their devices into the platform without an alternative. Chamberlain MyQ did this by turning on CloudFlare Super Bot Fight and 429'ing anything that doesn't look like it came from their app. That's fine and reasonable, but I can't even pay to create an OAuth client, even though I am paying for their Tesla integration, and they…

MyQ was my reason for trying HA. When I saw that the plug-in keeps hitting their API over and over, my excitement faded. I know Chamberlain are the ones causing this issue to begin with (missing HomeKit integration) but DDOSing somebody’s API is not a solution. My experience with HA came to an end when my gate unexpectedly opened on its own - and based on the log in the phone app, it came from HA.

MyQ had a HomeKit bridge... which they cancelled [1]. My experience with it was that, aside from eliminating the Internet/cloud dependency, it didn't make the MyQ experience any more reliable overall and I'm much happier with ATHOM's ESPHome-based kit.

[1] https://www.theverge.com/2022/8/23/23318157/chamberlain-myq-...

Re: Home Assistant 2023.11

#135

Earlier quoted context omitted.

Is there a beginner non-coder friendly HA “getting started” guide that someone here could recommend? I’m interested in investing in an ecosystem of smart switches, blinds, thermostat and music that is future proof and I can take with me as I rent Have some experience with ecobee and smart switches and trying to get onto Airplay 2 currently

HomeAssistants official plug'n play solution is HomeAssistant Green: https://www.home-assistant.io/green It's basically a mini computer (sbc) with homeassistant preinstalled and ready to go. Do expect that HA will expect some attention every few months. It's not a "set and forget" thing.

One of the things I'm really surprised by about HA/Nabu Casa is the complete lack of an LTS branch/releases.

Things related to homes are measured in years or decades. HA has gotten better about it but there are still occasional "cross your fingers and pray" scenarios when doing upgrades because every single release combines new functionality, fixes, breaking changes, and introduces new issues (of course).

People come to rely on things like their lights working and upgrade management is a tough spot - "Oh this integration broke because something else changed an API so I need to upgrade but let's find out how much time I need to update my config and hunt down any other issues".

This is one of the reasons things like Lutron are so expensive. You can have it installed and come back to the house 10 years later and everything will generally work as it did a decade ago.

For things like Nabu Casa cloud and their own hardware I'm sure plenty of people would pay a fairly significant premium for a subscriber-only LTS release that balances HA power, local, and flexible with Lutron user experience.

That said open source monetization is tough, and consumer monetization is as well. Combine the two and it sounds like a complete nightmare to me.

Re: Home Assistant 2023.11

#136
post #96

Earlier quoted context omitted.

> I don't know where all the frustration with HA is coming from Because it's breaking all the time. They have a big surface, and integrations are changing fast, this is natural. But at the same time they are also eager to change and break things on their own software, and this all accumulates to a messy experience for many people.

It’s not natural. I use both hA and Openhab with about 20 integrations/addons and HA breaks more (hello zwave}. Openhab almost never breaks on an addon level. If it breaks at all it’s in the core and patched quickly.

How HA has handled ZWave is my only complaint with HA.

Re: Home Assistant 2023.11

#138

Earlier quoted context omitted.

I don’t understand the negativity at all. HA rivals or exceeds commercial automation systems like crestron, RTI, etc. clearly those people don’t have any industry experience. I’ve generally never had an issue with HA beyond my own fault, like a buggy integration.

Home Assistant is an amazing project. I use it a ton, and I deeply appreciate what they've done. HOWEVER, Lovelace specifically is pretty terrible. Having used Grafana extensively, trying to use Lovelace to arrange a UI is so far backwards that it's constantly grating to try to create or modify dashboards. I really wish there was a way to use Grafana as the UI and embed cards as panels. The ability to drag and drop,…

Lovelace is pretty terrible, 1 hint I can give is to use vertical stacks, they're much easier to deal with for organizing where you want components in a given stack, then if you wanna move a component between vertical stacks go to the yaml view and cut it from source and paste it in the destination.

It's annoying to have to do a workaround like that but it makes it bearable at least.

Re: Home Assistant 2023.11

#139
post #131

The only negative thing I can say about HASS is that getting it running in Kubernetes with a USB Zigbee/Zwave adapter is a real pain. I've got a local cluster and had it working once, but had to rebuild everything for some unrelated reason, and haven't been able to get it consistently working since. There are too many in-built assumptions about how you're running it and the USB access. Documentation assumes that _may…

What benefit would running it in k8s have? I’ve been running it on a bare pi for years without issue and have never once thought I needed abstractions upon abstractions… is this more a “because I can” sort of thing?

Re: Home Assistant 2023.11

#140

Earlier quoted context omitted.

Lots of negativity in here. So this is nice. I use HA at home and it's nothing short of amazing. I primarily use Zigbee devices and they all just work. I pay Nabu Casa for the external access and Google integration. I look forward to replacing my Google Homes with HA as well.

I don’t understand the negativity at all. HA rivals or exceeds commercial automation systems like crestron, RTI, etc. clearly those people don’t have any industry experience. I’ve generally never had an issue with HA beyond my own fault, like a buggy integration.

The negativity is because it's a free project that competes with the thousands of home automation startups.
Post reply on HN