The future of YAML

nickrout · April 22, 2020, 9:00pm

I have only been skimming this thread recently, so if this is a re-post I apologise.

templewood · April 23, 2020, 10:39am

I don’t think there is a lot of pushing for 100% YAML. I see that as more of a straw man to retroactively justify a bad decision on the part of the management team. (“It never would have worked anyway, and besides they’re just a bunch of complainers who never contribute anything.”)

But even if that was their goal, there’s a big disconnect between “The door for YAML will be left open for Integrations” and slamming that door shut and driving a nail thru it six months later without any dialog or engagement besides the blog post that announced it. Incongruity breeds mistrust. And preemptively locking the door shut on contributions stifles creativity. Is that what Home Assistant wants the future of their product to be?

Personally, I’m strongly in favor of YAML but not 100% YAML. For integrations where someone might reasonably know the configuration answers in advance, and where those configurations may be relatively static, I think YAML is a better paradigm, and a necessary one for those of us whose networks are professionally maintained (static IPs, several devices of a similar type, etc).

If a contributor wanted to support that text-based functionality, I see no useful reason to prevent them from doing so. And if someone had an idea on how to make that coexist peacefully with a GUI-developer’s work on an integration, I see no reason to prevent that relationship forming either. Unless you’re making decisions based on dogma rather than functionality.

The architecture document should be amended to remove the line about refusing commits and integrations which implement YAML functionality. Instead it should simply state that these are not officially supported and their inclusion in integrations is optional on the part of the contributors who maintain them.

chmihai · April 23, 2020, 7:01pm

I have noticed the still in the “We are still an open-source project”
I am using the HA since about 3 years now and really appreciated the dynamic behind of this project due to the community. Thank you all!
For me the following things are important in respective with HA:

Privacy - I always check the hardware compatibility list of HA that the new equipment has local push/pull support
Stability
Flexibility - can add my set of devices, scripts…adapt the platform for my needs.

I think that last one will suffer as the GUI only approach will be taking over.

I do not care if YAML or JSON as long I can follow and I can adapt things to my wish.
If I break it I know where to make a step back and try again. I know where to look for an fix. Probably that will not be possible anymore with GUI approach
I do not want to be supporting 3000 integrations but want that HA to support top X vendors which agreed to be open and provide an Local API. See counter-example with Smartthings - which do not support Local API even their hardware hub maybe quite powerful.
To much “horizontal” expansion is not desired in a project like this - in my humble opinion.
I will be happy to keep paying NABU CASA as the project will keep the HA system as much as open possible.
In my opinion HA is not for everyone - it should not be tailored for newbies. If you want to make your own custom system then you should pay expertise (someone doing it for you) or you should put your brain at work (and time)

Again, thanks for all you guys you did so far, I really, really appreciate it! Well Done!

Stiralbios · April 26, 2020, 10:06pm

Hello,

I am confused.
Why do the integrator of component have to do more work to integrate the yaml + the ui configuration ? They should just have to integrate the component to ‘home assistant core’ and the core of ‘home assistant core’ should handle both type of integration.
If what is said in the article is true, i find this to be a very odd software design and I think the main issue come from there not from an UI vs YAML thing. Both should be work for every integration without involving devs working on integrations.

Edit: my concern have been expressed there (by someone else and it’s well wrotten) The future of YAML

athome · April 27, 2020, 3:07pm

Does ADR0010 set the stage for improvements on the management of Zwave devices?

“Run commands that affect the Z-Wave network. You won’t get feedback on whether most commands succeeded, but you can check the OZW log to try to find out.”

Until adding, removing, and configuring z-wave devices becomes significantly better, the adoption of home assistant will always be constrained and the issue of UI vs YAML configuration won’t matter.

francisp · April 27, 2020, 3:32pm

Really ? I was not going to reply in this thread anymore, but we should hold a poll how many HA users use zwave. I’m happy using HA as it is, and I would not touch zwave with a 10inch long pole.

cgarwood · April 27, 2020, 5:12pm

Not really related to this thread but no, ADR0010 has nothing to do with management of Z-Wave devices.

There is, however, work progressing on a new Z-Wave component that should solve a lot of the issues the current component has. You can install the alpha/pre-release of the new component as a custom component through HACS, and we’re hoping to get a beta into core within the next few release cycles.

We’re trying to get the core of the new component done, and then an improved UI for managing the devices will come later.

firstof9 · April 27, 2020, 5:23pm

I run zwave, works perfectly fine. The issue people have with zwave is 95% of the time signal strength issues.

Pressing “Add Node” button to add a device too hard to do?

petro · April 27, 2020, 7:45pm

Some devices do not play well with openZwave. But what’s interesting to note, these same devices are hit or miss in other platforms. Just check out the SmartThings forums sometime. Zwave issues everywhere. People on these forums associate Zwaves shortfalls with Home Assistant without realizing that it’s just zwave in general. The device market for zwave is the wild wild west and there’s always a chance that you buy a device from a lazy manufacturer.

athome · April 27, 2020, 7:48pm

cgarwood,

Thanks for the response.

Here is how it is germane. When I want to be sure that a z-wave device will be removed, i unplug the z-stick from my server, remove the device from the z-stick, then edit core.entity_registry and zwcfg*.xml to remove the device. When I add a new device, I Add Node from the Z-Wave configuration UI. Then I edit the files because it is simply easier to edit core.entity_registry and zwcfg*.xml to rename the new device.

I appreciate that you understand.

Super jazzed to hear about the new UI. I’ll go have a look. Thanks for sharing.

cgarwood · April 27, 2020, 8:15pm

The device market for zwave is the wild wild west and there’s always a chance that you buy a device from a lazy manufacturer.

Which is funny because Z-Wave has a certification process for their devices (which is why they’re usually more expensive than Zigbee) but it’s amazing how many Z-Wave devices misbehave or implement the standards very poorly

sparkydave · April 28, 2020, 3:48am

My z-wave network has been rock solid for years now.

MIthras.ftw · April 29, 2020, 11:29pm

I guess it’s time to look for HA alternatives

kanga_who · April 29, 2020, 11:36pm

7 mths between posts. See-ya

kanga_who · April 30, 2020, 1:13am

Thanks for the feedback.

G8YTZ · May 1, 2020, 7:13am

I’m an IT professional, though not a coder, my specialism is telecommunications, but I’ve used this platform for getting on for two years and have therefore done everything in yaml. To remove this now would be a disaster, frankly the software updates and unnecessary breaking changes are taking too much of my time and I am now wondering if I should continue with this platform and throw away my time invested in learning it.

Please can we focus on simple required features like a proper schedular with a simple UI fit for the wife to use rather than playing technicalities all of the time? If I die of Corona tomorrow or sell the house I want a system that will carry on maintenance free and supportable with minimum effort. In effect I’m asking for an LTS release of Home Assistant with a 10 year support window like Microsoft, Ubuntu, Asterisk etc.

TheMikaelT · May 1, 2020, 9:34am

I know this is couple weeks after the discussion but I just want to tell my story.

I am relatively new to home assistant and my setup is extremely simple. I run telldus switches on all bedroom lights (I know they suck but got it almost free so win for me) and one sonoff single wire switch on living room light. I have few Philips hue bulbs and few temperature gauges. I have broadlink ir blaster for projector, screen and tv. I have withings scale to see I have gained weight again. There are two Chromecasts. One for living room and one for sauna. And finally there is Roomba connection to start and stop to vacuum.

My first take on home assistant was maybe two years ago. It ended with me making the statement online that Home Assistant will never be successful and popular and it’s just for people with programming skills and no life. Yeah I was slightly heated

So what changed? Home assistant and me. There are the usual suspects on YouTube. Frenck, DrZzz, Juan… and many more. Suddenly I found a community that actually made even me understand the basics of setting up home assistant. Running home assistant in virtual machine on my Mac Mini server, was so much better solution than raspberry pi. After logging in, it discovered Philips Hue, Telldus and Chromecasts without me doing anything. For a NOT-PRO-USER this was a easy start. Making sure I set up automated backups and I was ready to try again. And I did find completely new world.

I love UI based solutions. If I can set everything up with few clicks, it means that even my elderly mother or father could do the same. It doesn’t mean that I don’t get the thrills when doing something with YAML, and actually getting it working. I love the fact that the basics setup has been done so easy these days, that almost anyone can do it. It’s definitely the right direction. More accessible means more users. More users means more support. More support means more accessible and so on.

I understand the scepticism on changes and some might feel that they may not be done openly enough. But as long as I can TRY to get home assistant do what I want, I don’t really care about how it does it. As long as you never go to, let’s say Philips hue route, you are getting my respect.

At this point I personally think this is the right way to go. It might take sometime to get used to, but if there is not a drop in developer engagement in add-ons etc, then I don’t have thing to worry.

But like I said, I am just a user who LOVES everyone of you who writes and comments on different forums on home assistant or some add-on or integration I am trying to get working. Without you all I couldn’t be a user. But of course I am just one person who has to nothing to give, only to gain.

danielperna84 · May 1, 2020, 10:28am

Then Home Assistant is by far one of the worse things you could use. Especially considering the “selling the house” case. If the buyer needs a briefing prior to moving in, it’s already a broken system to begin with. However. you’ll run into this problem with pretty much any home automating solution. If you’re using HA, but the buyer OpenHAB, then there already is a conflict. So essentially, the used platform for automating the home shouldn’t be a factor for selling a home in any case. The only exception to this, as I see it, are hardwired solutions like KNX, where the people who move into the house have the choice to tap into the system. But these systems typically also are pre-configured by professionals who assign a specific switch to a specific light. This works regardless of any HA-solution. And if I were to sell my house, I would strip out anything that requires maintenance.

G8YTZ · May 1, 2020, 11:01am

My house is a self-build and I put in KNX from day one as I liked the robustness, supportability, quality of components and non-proprietary nature of th technology (nearly 10 years now and KNX has not had one issue).

I do use a commercial product, but that’s 10 years old now and I wanted something that could integrate with many other systems and external sensors. HASS.IO fits the bill perfectly, but it has two major drawbacks:

They screwed top the KNX Lovelace Thermostats after 0.97. Who every thought to do this? Out of touch or what?
The lack of realisation that what’s needed is an LTS version the will just run for years with only bug patch updates.
The lack fo a proper scheduler and suitable domestic UI.

Fix these three items and Home Assistant has no peers, but supportability is a major cost to be factored in, the time I’ve spent sorting out breaking changes and waiting to bug fixes has meant of I factor in my time, I am just better off spending £4k on a Gira Home Server or the like. Its such a shame as the solution has so much promise.

danielperna84 · May 1, 2020, 11:21am

Eventually that will happen some time in the future. Right now Home Assistant can still be considered beta, as it’s not even at version 1.x.