[a bit of a rant] Starting to consider using Hubitat purely for Zigbee/Zwave and HomeAssistant for everything else

[proffalken]

Hi all,

As many of you know I've had my Hubitat for a while now and it's been great, however as I start to look at integrating it with other devices such as Velux windows and home security solutions, I seem to be relying more and more on the community-written drivers compared with the ones that ship out the box.

I understand that Hubitat is US-based, and therefore is always going to be more US-centric than EU/UK-focused, but the lack of support for some of the top manufacturers (especially the ones like Velux who have opened up the documentation to their API!) compared with the speed at which new components are developed and then brought in to Home Assistant is glaringly stark.

When I blogged about why I was going to use Hubitat, one of the reasons I chose it over HomeAssistant was because Hubitat had a company behind it running dedicated hardware and a warranty as opposed to running my own hub on a Raspberry Pi, however the more and more I get into this stuff, the less and less appears to be supported out of the box.

Maybe I'm expecting too much. Maybe I should have stuck with Home Assistant. Maybe I shouldn't even be writing this post...

The sad thing is that I'm now rapidly getting to the point where Hubitat is going to be a Zigbee/ZWave hub and everything else is going to happen through HomeAssistant because it just has better support for major manufacturers.

Am I missing something? Am I wrong? I hope the answer to both of these questions is "yes", however I suspect Hubitat is turning into a commercial product for which I'm expected to write my own drivers, and at that point I'd rather support Open Source...

Thoughts on a post-card to the usual address

[csteele]

You have to choose your System (total, not just a slice) to meet your needs. Your home, your challenges.

@proffalken

and at that point I'd rather support Open Source...

Just a quick reality check on this.... if You put your code out on Github with an appropriate license, isn't that as much Open Source as anything else?

This idea of having a "modular" or "Component" home automation SYSTEM is, in my opinion, a well worn path. Some people choose Home Assistant as an 'aggregation' node while others choose Node-Red. Probably there's a third group doing both. And there could easily be a group having Hubitat Hubs, Node-Red, HomeBridge, Lutron, Hue and Home Assistant all working together, each doing their best parts better than the others.

As a result, I'm not reading this as a rant, but as a mile marker that you've pushed open a door into a larger world of Automation that others have also found.

[proffalken]

@csteele

Just a quick reality check on this.... if You put your code out on Github with an appropriate license, isn't that as much Open Source as anything else?

Is the hubitat source code available on github? if it is then that's awesome.

I've been working with Open Source for over 20 years now and made a career from it, however as far as I can tell Hubitat isn't Open Source even if the community plugins are?

@csteele

As a result, I'm not reading this as a rant, but as a mile marker that you've pushed open a door into a larger world of Automation that others have also found.

Quite possibly, yes, it's just a bit frustrating that many devices that aren't available here in the EU/UK are supported, and yet so many devices that are available here and in the US aren't supported.

And it's not always complicated things either - I've had to jump through hoops to get album art displayed on my dashboards from my Sonos speakers, that works out of the box in HomeAssistant and it feels to me like it's a pretty obvious thing to want on a dashboard.

I don't mean to come across as ungrateful, I appreciate that my priorities are not the same as other people's, and it may well be that I need to run some kind of hybrid setup as you say, it's just a bit frustrating when I thought that Hubitat would support a lot of the stuff out of the box and yet updates seem to focus on fixing existing drivers rather than adding new ones.

[djw1191]

@proffalken

however I suspect Hubitat is turning into a commercial product for which I'm expected to write my own drivers

I’m a little confused about this statement? Hasn’t Hubitat always been a commercial product? What you describe is how it’s always been so not sure I’d say it’s “turning” into anything,

@proffalken

yet updates seem to focus on fixing existing drivers rather than adding new ones.

I’m also not sure I’d agree with this statement. I think if you scroll through release notes you’ll see new drivers being added in every major release. They might not be the drivers you’re looking for, I’d say they are generally zwave or zigbee and it sounds like perhaps you are looking at either cloud integrations or other sorts of APIs. But a little unfair to say there have been NO new drivers.

Either way, you use the platform that makes the most sense to you personally. If that’s hubitat, cool, if it’s something else, also cool. I don’t think you’re going to find people trying to convince you to stay, simply because it doesn’t impact us.

[marktheknife]

@proffalken

Is the hubitat source code available on github?

No. As @djw1191 noted, Hubitat has always been a commercial, closed source product.

And I would also have to disagree with the notion that platform updates favor fixes to existing drivers over addition of new drivers; it’s usually a mix of both.

I don’t know that much about home assistant because I’ve never tried it. I’m sure the fact that it’s open source is a benefit in terms of adding device support. But I’ve heard it’s more difficult to use, probably because it’s an open source project.

[wayne.pirtle]

@proffalken I agree with @csteele,

@csteele

I'm not reading this as a rant, but as a mile marker that you've pushed open a door into a larger world of Automation that others have also found.

Welcome Frodo.

I came from Smarthings to Hubitat via a dabbling with openHAB. When I got to Hubitat I thought I had found Shangri-La. It tuns out it was only a bus stop on the way to the multiverse. I now have HE, Hue, and HomeKit via Homebridge. I now realize this also is only a pause on my home automation journey.

No system is going to be the best at everything, even the Control4 and Crestrons of the world. They have more customized support you pay for, so they can get closer. For those of us who do this in the DIY word do so because we enjoy the challenge of solving the grand home automation puzzle and appreciate the feeling of success when we do.

I live and work in the enterprise computing world, so I appreciate the nirvana that fully integrated systems (e.g., PeopleSoft, SAP) theoretically should provide, and the reality of the complexity and fragility best of breed systems create in trying to improve on the fully integrated promise. The same challenge applies to home automation.

The goal of being able to use one hub (core system) to support all possible devices using all possible technologies is noble. The problem is it isn't practical with the diversity of technologies available and business plans of the device manufacturers. Every system is going to be best at something that another either doesn't support or is only mediocre at supporting.

In the end we are doing what engineers do for projects everyday: finding the most optimal solution for the project scope with the associated constraints. Each time we want to add a feature to our existing system we reevaluate what we have already put in place based on the new scope and new constraints. Some times this necessitates a change to the fundamental architecture.

I didn't setup Homebridge and HomeKit because I wanted to. It adds complexity and a it more fragility to the system as a whole. I did it because there were products I wanted to include in my system that only talk HomeKit or IFTTT. I have made a fundamental tenet of my architecture a requirement that all devices will be controlled locally. That eliminates IFTTT out of the gate. Is IFTTT a undesirable product? No, it just didn't fit my architectural direction.

So, since Hubitat doesn't support HomeKit natively, do I bemoan my choice to use Hubitat as my central controller (I use Hubitat apps for all my logic, though someday I may decide to add NR.)? No, it just evokes other architectural choices.

Another aspect of my architectural choice to keep Hubitat as the center of my home automation universe is the superb user community and the active interaction between the staff and customers through the community. That is worth gold. Open Source is altruistic in it's proposal. Unfortunately, when I have a problem I need support now, not when the guy who knows how the 1s and 0s of the protocol work gets back from vacation, over being ill, has time at work, etc. Don't misconstrue my statement about Open Source. It is great, but I evaluate the risk and make a conscious choice to be prepared for extra challenges if they arise. I don't expect problems, but I have to think of my customer, my spouse. When something doesn't work, she doesn't really care that it is Open Source and probably saved us money. She wants it to work. My wife is actually VERY understanding of my home automation obsession, but she has her limits.

Sorry about my rant. I'll get off my soapbox.

You have instigated a very good and worthy discussion about architecture, expectations, and what happens over the life of a system. I hope our responses are taken as an effort to help you think through the challenges you are facing, and to understand this is a normal progression in the development of a system. We have all been there, not wanting to make a fundamental change to our architecture and wishing we could find "the One Ring"

"One ring to rule them all,
one ring to find them,
One ring to bring them all
and in the darkness bind them. "

J. R. R. Tolkien

[erktrek]

So why not do that? That's the beauty and flexibility of HE.

I am using a combination of Node-RED and HE for myself and my residential clients. I have minimal apps running on my hubs and it has been working out really well. I have a test HA install running on an old RPi3 and am using an old C-4 Nortek stick (here in the US the C-3 & 4's came with one external stick). It seems to be working well too - I also have that interfaced with Node-RED and maybe will try the HE Bridge stuff.

Note: Currently HA takes too much time to manage properly for my situation. It's easy to do some stuff quickly but once you start to get into things it can get more complicated than I need to worry about/have time for..

On the opensource front, It IS closed source built on opensource technologies which is kinda bothersome however as long as HE is compliant with the appropriate opensource license terms for all the stuff they use then I'm good or more precisely it really does not matter how I feel about it...

[sanewton72]

agreed. nicely put @erktrek, particularly wrt HA.

and on the point of US-centric device support: its to be expected, seeing as Hubitat is a US company, and the US is the(/one of?) the largest addressable automation markets on the planet.
my experience (as a non-yank) has been the Hubitat team are very open to adding device support.
the question could be more about what the community in a given country/region can do to enable them to do it. they can't automagically somehow know/prioritise what devices need doing, and go around buying random products in far away countries.

in my case, i have been proactive about identifying key products that I believe will impact the most users in the aust market, and investing time and $ to send over devices to the Hubitat guys, so they can do their thing (which helps us).

[neonturbo]

@proffalken

but the lack of support for some of the top manufacturers (especially the ones like Velux who have opened up the documentation to their API!)

@sanewton72

hey can't automagically somehow know/prioritise what devices need doing,

I would say this is one of the issues. With a small team, they are not going to be be doing extensive research into what API has recently opened up, or what new device has come onto the market. I think that some of this needs to be on the users to identify what devices have come into play, and put in a feature request in the forum (if not emailing support too) to bring things to their attention.

As to whether they feel it is a worthy thing to add, or if they have time to add it, that is a different issue. But they have been very good in most cases about adding new devices, as well as updating and improving existing items.

@proffalken

compared with the speed at which new components are developed and then brought in to Home Assistant is glaringly stark.

Well when you have 500 or more people doing things, it should speed up the process. But from what I experienced, that speed came at a huge price of instability, and the inability to support things. I got "read the f'n manual" more times over there than I care to remember. If something didn't work, it was on that person who wrote the integration to help you, or to rewrite things, and somehow get it into the update stream.

There is no real official support channel like there is here. And that doesn't even include the horrible setup and all the mucking about in YAML to put some stupid { or indent in the right place else you break the whole thing and have to wipe everything and start over again. So I don't think things are as rosy as people make Home Assistant out to be.

I don't think Hubitat is perfect, but it is a very good commercial product and it does what it is supposed to do.

[proffalken]

Thanks folks, there's a lot of food for thought in these replies and I appreciate the time you've taken to write them.

I'd struggled with OpenHAB and HomeAssistant in the past, and Hubitat seemed to be the "holy grail" in my search for home automation perfection - not too expensive (Control4, I'm looking at you here!), support for a significant amount of devices, and the ability to expand via user-plugins where needed. I think much of my frustration has come from needing to rely on the user-plugins more than I rely on the in-built drivers/apps, and I'd be lost without Hubitat Package Manager!

I totally get the point that as a smaller team, releases are going to be more tightly focused and that (as I mentioned in another comment) my priorities are not the same as other people's priorities, and with that in mind I'm wondering if there's a public "roadmap" or similar available so that the community can see what's coming up in future releases? A simple web-page would be awesome, a trello board would be even better! Something like the Emma Roadmap perhaps?

The ability to feed into that roadmap in a clear manner would also be amazing if it doesn't already exist. DigitalOcean have an ideas site in which folks can add things they want and the community can vote on them to help the developers understand where the demands are. As far as I'm aware, the only way to raise ideas with the dev team is to post on the forums and hope they see it/respond to it, and given the amount of traffic on here it would be unfair to assume they read every message.

In my original post, I asked whether I should be running multiple systems and it looks like the answer is "yes", so that's what I'll do.

I also asked if I should even be writing this post, and I think the answer to that was "yes" as well - it's certainly provided some good responses for me and I hope others who experience the same frustrations in future will be able to learn from those responses too!

Thanks again!

[bf1]

I and others have asked for a roadmap and also the ability to upvote features/integrations but the team have said no multiple times - that's fine.

I've settled (for now) on:

Node-RED - all my logic and the cloud integrations
Hubitat - Z-Wave/Zigbee management only
Home Assistant - UI only
as well as Hue & Lutron hubs.

It works for me.

[neonturbo]

@proffalken

if there's a public "roadmap" or similar available

I am just picking on Velux here as a placeholder name, I have no inside info about any of this. So don't take this as picking on this particular thing. It could have been a Smartthings sensor or a Aeotec switch or whatever.

No there isn't a public roadmap, and I agree with not having one. So many times users have been pissed off when something is promised yet it never materializes. Wink did this to us, "Big changes coming by the Hollidays!" never materialized. I think it caused more ill will than anything. A companies priorities change over time, and devices/integrations come and go. If they said "yea, we want to integrate Velux" and it never happens (maybe not even the fault of Hubitat), what would you think or do?

@proffalken

the community can vote on them to help the developers understand where the demands are.

I had a discussion about this with someone a while back. After much discussion, and some thought, I decided it is a bad idea just like the roadmap.

So you have 500 people voting for "Velux", and 5 voting for a Zwave switch, and the Zwave switch makes it into the firmware (because it is quicker/simpler/fits timelines/skills of staff, whatever) before "Velux", what will those 500 people think? Maybe they can't even do the "Velux" do you get pissed off and move on? Or there is a mob of 500 angry people in the forums and on the support line.

In both cases, there is no winner here. The everyday users lose, the feature voters lose, the Hubitat staff loses, and so on.

[proffalken]

@neonturbo

So many times users have been pissed off when something is promised yet it never materializes.

Yup, and this is what happened for the "proper" driver for the Aurora AONE's, which a random comment suggested was on the way and then it suddenly turned up in a release. I don't expect the devs to come back and comment on every forum post asking for new features, and the "ideas portal" or a clear roadmap would remove the need to do so whilst keeping the customer informed.

@neonturbo

So you have 500 people voting for "Velux", and 5 voting for a Zwave switch, and the Zwave switch makes it into the firmware (because it is quicker/simpler/fits timelines/skills of staff, whatever) before "Velux", what will those 500 people think?

It depends how the ideas platform is used - if there's no response to a feature then yes, people are going to get upset and either rant or move on.

If the idea is moved onto a roadmap backlog, then at least people know it's been selected for consideration and can track the progress.

A lot of my day-job these days is supporting large enterprises in how they improve visibility into their software development cycle and how they can then use that to keep their customers happy.

A huge part of this is transparency - if people can see that something is on the backlog and working its way towards the top, or is in development, they're far less likely to get angry. If they ask for something on a forum and comments are made that appear to be a commitment without any further information given and no ability to find out for yourself without bugging the developers, that gets very frustrating very quickly for all involved (especially the developer who's being hounded by forum posts!)

To come back to your point - if there was a quick justification given for the decision to develop Zwave rather than Velux ("Sorry folks, we just don't have the capacity at the moment", "Sorry folks, we're talking to Velux to make sure we get this right, in the meantime we're going to focus on this Zwave device so we're not holding everyone up") then that would be miles better than just hoping that at some point someone might read a forum post and then a couple of months later publish a driver that meets the request.

[proffalken]

@bf1

Node-RED - all my logic and the cloud integrations
Hubitat - Z-Wave/Zigbee management only
Home Assistant - UI only
as well as Hue & Lutron hubs.

Thanks, it seems like I'll be going down this route as well...

[Inge_Jones]

I don't think any offence is meant in the fact US devices have more support than European. Most people who contribute a driver do so for a device they own and need one for. There are more people in America than there are in England so the probability of finding the driver you need for a US device is higher than for finding an English one. Just like living in some parts of the world you're more likely to go hungry and not get medical care than others, living in other parts of the world you're more likely to be frustrated for technology support than others.

[amithalp]

Europe population is 2.3 times bigger than USA population. Ignoring or investing in the EU should be a business decision that a company should take.
I would love to see HE team pushing more towards the EU market. More users will join and more user development will take place. Currently it seems to me ST is controlling the EU and HE is way behind. I wish this would change.

[Inge_Jones]

Well EU maybe. Won't help us poor stranded souls in the UK now

[marktheknife]

@amithalp

Currently it seems to me ST is controlling the EU and HE is way behind. I wish this would change.

ST presumably has an advantage in international markets since it’s owned by one of the world’s largest consumer appliance/electronics manufacturers. HE has a small team based entirely in the US (AFAIK).

[csteele]

You've been a Community member for 2 years now and that makes me think you saw a big push for UK devices last year during one of the Hubitat Live sessions. Mike had gotten a box of engineering samples from a vendor and got to work on them. I read a LOT of the posts here on the forum and can't think of a missing device.. not because there are none, quite the opposite, I think the process of identifying needed/popular devices and getting them physically from far flung places is more difficult.

I've been here a mere year longer, but I have seen the quantity of supported devices climb very rapidly... a lot based on what devices land in Mike's lap. (Probably Brian's lap now days too.) If I needed something, I know I'd be starting a topic with "I contacted Bob at Blue Acorns and I'll PM Mike and staff with his contact info. He said if Hubitat contacts him, he'll be happy to send out engineering samples and discuss the .... " and so on. (Blue Acorns is imaginary, just as Bob is. )

My point is.. someone around here has better access to the UK vendors, if only from a TZ (and Language ) standpoint and can travel a small distance down those Company's Pre-Sales path for info that can be handed off to Staff.

[neonturbo]

@csteele

and getting them physically from far flung places is more difficult.

Especially this past 18 months or so.

@csteele

If I needed something, I know I'd be starting a topic with "I contacted Bob at Blue Acorns and I'll PM Mike and staff with his contact info. He said if Hubitat contacts him, he'll be happy to send out engineering samples and discuss the .... "