Why Doesn't Hubitat (or Life) Work The Way I Want It To?

šŸ’¬ Lounge
22

I keep seeing people linking to that silly Josh.Ai thing. It's a whole lot of gravy with no meat or potatoes.

I've watched their videos, it was basically nothing more than what the "Rooms" tab of Hubitat provides, but with a bunch of giant obnoxious stock photos and a ton of "UI shine" added to it. Functionally it seemed significantly worse than Hubitat, tbh. Then the other part of this Josh thing is basically just Siri.

Is Josh very shiny? Yep. Is it good for selling to rich people with more money than sense? Yep. Would I ever want it in my house? Not unless it was about 95% cheaper than it is. It looks like a great product for integrators to sell, but it doesn't look like a great product overall. It's just a big expensive manual control interface.

Wall mounted tablets functionally provide nothing more than a regular wall switch does, and they lack any sort of tactile feedback or ability to be used without looking, making them worse than a regular switch. Voice control is handy sometimes, but that's easily had with Hubitat via HomeKit (or Alexa/Google if you don't mind having all your personal data gathered and sold).

I really don't see the appeal of Josh other than to integrators looking to add $20k to their bill for someone's McMansion home "automation" system. Automation in quotes since Josh seems very very manual control oriented.

If I'm spending big bucks on something, I'll put the fanciest presence sensors available in every room, 'passage' sensors on every doorway, lux/color temp sensors in every room, having my lighting always be the same brightness and color temp regardless of what's being added to the room from the windows, and never ever touching the light switches again. I'm definitely not putting some stock photo filled tablet on the wall that's in every way worse than a light switch from 1900.

23

I feel like you're holding something back... how do you really feel?

:wink:

2 Likes
24
  1. I admittedly haven't read every post here.
  2. What really chaps me is when something works fine FOR YEARS (e.g., a custom driver), no changes have been made on my end other than updating the Hubitat firmware, but doing that renders something unusable.

This happened recently with my Daikin mini-splits. I used the only driver I could find that worked when I started using Hubitat (2021? maybe 2022). NOT a driver from @sburke781, but something (the only thing, at the time) that worked with my units. It was never perfect, but I could turn the units on and off, change the mode, and adjust the temperature. Good enough. Worked for years. Then one day after an HE firmware update (I think when I went to 2.3.8, but it's not like I log this stuff every day. Might have been the horror that was 2.3.6 (sicks) as well.), the dashboard tiles for my mini-splits, on two different hubs in two different buildings, stopped working. Oh, and I think they were fine under 2.3.7. The Rules still worked so I didn't notice it right away, but the dashboard tiles? Kaput. Sigh....... FINE BY ME IF YOU WANT TO JUMP ON THE LATEST AUTOMATION FAD, HUBITAT CREW, BUT HOW ABOUT NOT BREAKING STUFF YOUR EXISTING USERS - YOU KNOW, THE PEOPLE WHO HAVE ALREADY SPENT MONEY WITH YOU - ARE USING, AND DEPEND ON? When I or others have made this complaint in the past, we almost always hear something like, "Oh, that driver used the wrong [whatever] and was broken all along." Well, it worked for years, it's not like YOU were offering something with the "correct" [whatever] Hubitat, and so I used what I could get. I don't do this for a living, but YOU do. If you're going to offer a device that relies heavily on user-generated "software" (apps/drivers) to make it useful, it really needs to be a lot more fault-tolerant than what I've been seeing lately. <OK, rant over. That felt good.> And yes, I know I can roll back to whatever previous firmware. I just shouldn't be forced to do that, and be forced to lose potential security updates.

25

From a non-AV automation perspective Hubitat is as or more capable than dealer systems. And:

Control interfaces are still important to lots of people. It's an area that Hubitat has not focused on. Thus, the plethora of threads on alternative control frameworks. Lots are adequate. None are outstanding. Lots of people use Alexa/Apple/Google for voice. And they all are somewhere between suck and limited. Control4 and Josh.ai have the control stuff down, more or less. Especially AV. It's more than just pretty icons and tablets. It's about how everything works together.

I've got a Control4 system as well as Hubitat. Control4 is very good at controlling stuff, especially AV. They're one of the few options for hard button remotes with Harmony-type control. Their integration framework for controlling AV equipment is really quite good, especially for complex distributed/matrix AV sources. Technically, Control4 is about as much of a legacy system as one can think of, with all the plusses and minuses. Their UI works but is dated. How one builds the layouts is clunky and not flexible.

I don't know the technical framework of Josh.ai. My impression is that the AV stuff is not as advanced as C4, but is growing. I don't know anything about their automation capabilities. But their UI looks really nice and highly functional. The voice stuff seems good.

The dealer systems companies essentially do try to rip people off. The C4 8" tablet is about $1,200 and isn't any more capable than a $100 Lenovo/Samsung tablet. Nobody 'needs' tablets, but it is kind of nice to have a hub, somewhere. The companies add just enough proprietary functionality to get people to invest in the proprietary hardware.

Where these dealer systems struggle is making money. C4 is part of a public company, Snap One. And Josh.ai is VC funded. So they have formal profitability growth targets. They've focused on the narrow high end of the market, which limits growing by breadth. C4 is moving from an optional $150 annual fee (similar in scope to Hubitat's) to a mandatory $250 annual fee. Josh.ai has gone through revenue models and I think is now $720/year for 500 devices. But even the consumer brands struggle to make money, including brands such as Logitech, Amazon, Samsung.

I need some system to control my house, especially AV. I'm OK with C4, but I'd be tempted to dump them for Josh.ai. But not at the price they are charging. And of course, I'd need some way of managing the system myself. C4 offers that opportunity, not sure about Josh.ai.

26

I donā€™t understand the argument in favor of this viewpoint.

The platform develops and changes over time. Itā€™s on community developers to stay current with those changes.

The alternative would be for Hubitat to delay or avoid making updates to their own platform. Not a viable alternative IMO.

3 Likes
27

I was going to have a lot of things to say at first, but then I realized you said Life and not Wife.

2 Likes
28

Uh-Huh.. That'd be great. But too often, they don't. Home automation is largely a hobby because far too often, the latest cool thing fails to work properly a few years later. And when a person complains about that, the fan club boos them down. You don't understand or agree with this view because you're part of that fan club. Doesn't mean either of us are wrong - it just means we're after largely different things. Fear not, though. You will always win this match, because new and shiny almost always beats out ol' reliable.

1 Like
29

@marktheknife has a point though. Often apps will use undocumented features of the underlying host because it's more efficient or allows what otherwise could not be achieved. Or even if all guidelines are followed the author of the host system decides that some aspects may constitute a security risk or will no longer be available (rightly or wrongly). As a result the app breaks, and if the dev has abandoned the app then unless someone takes up the mantle to fix it.

This happens all the time. Google is a major culprit, I've had many things break or are no longer functional due to changes in their software. The reality is also that often things can't be fixed regardless. Having said that I don't recall experiencing anything that actually broke during my ownership of HE hubs.

30

There's a lot to be said for reliability.

1 Like
31

One of my goals has been to get Hubitat working well for AV control. I canā€™t solve the whole thing myself. The number of devices and interfaces is astronomical. But Iā€™ve taken a start on some of the ideas, and made them work for my own devices. Documented here: Experiments in controlling A/V equipment

3 Likes
32

If I were doing this I'd build up a framework in the system that is similar to the way Control4 works. Essentially you build out a virtual AV rack. Outs to Ins and Ins to Outs. The TV is the video endpoint in the room. The speakers are the audio endpoint in the room. The streaming box output goes to the AV receiver HDMI input #x. The AV receiver output goes to the TV. The AV receiver Zone 1 goes to the speakers. Etc. This framework can be made complex with matrix/distributed AV. Once defined you don't have to know/remember any of the complexity of intermediate configuration to control or automate the system. Play the Apple TV in Family Room. When the video in the Family Room is activated turn off the lights. Etc.

1 Like
33

No, youā€™re making assumptions about me because I donā€™t expect Hubitat to cater the entire platform to the needs of specific community drivers or apps. I donā€™t always need the new, shiny thing just because itā€™s new.

For one thing, I donā€™t think that breaking changes with updates of this platform occur very often, particularly compared to a platform like Home Assistant.

But since I have zero knowledge of coding, Iā€™m as dependent on certain community developersā€™ contributions as you are. In the event a breaking change occurs, Iā€™ve found that community devs typically do make needed updates. If it just so happens a dev doesnā€™t, I accept that as a consequence of a choice I made to use code that Hubitat doesnā€™t control, and that I am unable to fix myself.

But I donā€™t blame Hubitat for that. And I think thatā€™s preferable to the alternatives, which could include Hubitat not permitting community developed code to run on our hubs, or curtailing their own platform development efforts.

34

Sometimes I feel like I provide such an open or ambiguous opening statement that people either read past it or the conversation moves on so far it becomes irrelevant... I guess that's part of prompting a broad discussion like I tend to do in these topics.... Or maybe I'm just offering click-bait... :wink:

Hopefully people can still find a way to offer constructive suggestions without denigrating the platform or those involved.

1 Like