Weekly Update from Alex - 05/05/16

To add to @schapper05 's suggestions, A simpler route is to use the official smartthings/life 360 integration, then just use smarttiles and life 360 on her phone, not the smartthings app itself.

Life360 is a third-party geopresence app. Instructions in the following topic on how to use it with SmartThings:

SmartTiles is a third party customizable Dashboard for smart things that runs in any web browser. Many people use it to give a housemate or child on/off access to specific set of devices while not letting them make any actual account changes. Very popular in the community.

These two together give you the ability to recognize her presence and give her the ability to have some control over devices that you specify without having to give her account manager level access to SmartThings. :sunglasses:

3 Likes

I second what @JDRoberts days about life360. I use it for my entire family and it is honestly very very good.

I have 3 phones on AT&T one on Sprint. The Sprint phones goes goofy about twice a month. The others are rock solid.

2 Likes

Sorry to be negative, but since you started working on latency, my ST light controls have gone from immediate action to a 15-second delay.
This is not acceptable to me, and is pretty much the straw that finally breaks this camels back as to ST and its idiotic remote processing for simple local actions.
The only reason I had held off ditching ST for this long was because of Rule Machine, and now that you people have blown away that superb application, there is nothing on ST that works well enough to make me want to keep it.
I’ll be moving my lighting back to X-10 immediately, and as soon as Keen Vent Hub supports Ecobee temp sensors, I will bin the miserable ST hub. You should look at this as positive, since I will be lightening the load on your apparently-overwhelmed servers.
It’s been a long frustrating experience with ST, and I’m sick and tired of having light-switches that actually work being a time-wasting hobby, instead of an assumption.
IMHO, if Samsung starts building ST into their televisions, the support requirements will probably bankrupt Samsung.

1 Like

Nice job ST folks I’m really impressed how nicely things update in the android app now. I love what you are doing with it.

2 Likes

Hmmm… No update from Alex this week? I hope it means that all the issues have been fixed. :wink:

5 Likes

he heard you…

2 Likes

This definitely isn’t the normal right now, please submit a support ticket and PM me the ticket # and I’ll have someone work with you. :slight_smile:

1 Like

No thanks, Tim. I’ve already wasted far too much of my life trying to get ST to function. As far as I’m concerned, ST is a failed experiment.

My lighting is now back using X-10, and I’ll say good riddance to the ST hub as soon as Keen Vent gets theirs to talk to my Ecobee.

Not to dispute your decision, but even with all the hiccups ST have been having, ZWave is way more reliable than X10. Just saying… :sunglasses:

Yes, Z-Wave is more reliable as a protocol, but when it involves ST it becomes far less reliable than any locally-controlled option.

Lights controlled by X-10 are not subject to bad decisions made by people employed by Samsung, and as long as the power is on, the light switches will work - Internet connectivity is Not required to turn on a light.

Face it - the issue is that the ST hub is nothing more than a router - all of ST’s function is implemented in software in a server thousands of miles away, and that’s just a stupidly-complicated way to turn on a lightbulb.

KISS!

4 Likes

While I get that you are working hard and making improvements, I’m still seeing core breakages in simple, Smartthings-owned apps. (For example, the power allowance/timer trigger hasn’t worked for weeks.) Combined with external apps being broken (IFTTT, SimpleRulesEngine, etc) I’m finally making the very reluctant decision to switch from the pretty-but-broken SmartThings back to the ugly-as-sin, working MCV hub.

I didn’t make this decision lightly (and I don’t expect that my choice will have any real impact on the product) but I wanted to explain why. In some of my earlier posts you can see the automation path that lead me here, from crippled IRIS through MCV and both iterations of the ST hub, and along the way I’ve held on hoping for the platform to achieve anything resembling it’s potential (or it’s advertised features.) I’ve unplugged appliances that it could harm by not turning on and off properly, I bought a clock radio to supplement the broken ‘gentle wake-up’ lighting, etc. I’ve adjusted rules to remove functionality I counted on in favor of bare minimum ‘survival mode’, and I’ve even avoided installing the remaining plugs and outlets to try to keep from further complicating the (fairly simple) setup.

At this point, at least in my house, it is no more functional than the free, crippled version of the IRIS hub I started with. (In several cases, it is much less functional.) If I push a button on the app, it (usually) turns on a switch in the house. But if that was all I wanted, I’d have stuck with my x10 gear back in 2000. I need the automation to work, I need a minimum level of intelligence (timers, triggers) and so I have to leave. I’m not going to get rid of the hardware, so hopefully someday it will be functional, but right now it just isn’t. It is a good product, but it is just not reliable enough to entrust my house to.

I have forum email notifications on, so if you want to clarify please feel free to reach out. Similarly, please feel free to dig through my ST hub profile and see that it is not at all a complex system, and it is just not working. (A simple example: the plug labelled ‘coffee maker’ has a 130 minute timeout - it was turned on at 10am and is still on almost 5 hours later. And we accidentally brewed a pot in the middle of the night the other day because it was still on when it was prepped for morning, despite the timeout, a mode-triggered ‘off’ and 2 time-based off triggers…)

2 Likes

In what way did SmartThings break “SimpleRulesEngine”? Every deprecated platform feature that I’m aware of (and as a SmartTiles developer, I’m quite familiar) was announced with months of advance notice and Developer Advocate support to affected developers.

SmartThings still has to do a lot better for the developer Community (such as providing feedback opportunities for proposed platform changes and a pre-release Beta testing environment), but both parties have to put in cooperative and proactive effort. I believe SmartThings did their part in this case…

I’ll throw in this…

I finally found a use for IFTTT, and yes I’ve been trying, and it works great for me.

Basically using life360 as my presence, IFTTT detects when a family member leaves a certain location. It then tells ST to turn on a switch. Using rule machine, when that family members arrives home that rule is run, but only because she went to and left a certain location.

Also, the system is working… Though it needs improvements with reliability.

I was just very disheartened that two of the core institutions of the platform, routines and smart lighting, are still very very damaged.

2 Likes

In what way did SmartThings break “SimpleRulesEngine”?

I suspect in the same way that the ST-authored trigger apps stopped working. Simple push-button-receive-candy still works 99% of the time, but the “automation” part of “home automation” is flat broken. Even mode changes and simple time triggers fail more often than not. (That is why I specifically mentioned the ST-written apps that failed - any mention of a 3rd party app around here immediately results in “OMG don’t blame Samsung!”)

In a broader sense, SRE shouldn’t BE a third-party app. That level of basic functionality should have been built in from the start (and I had hopes that it would be over time.) I don’t have super complex rules - just simple triggers. For example, here is one that hasn’t worked reliably in months (after working fine for a long time, and functionally identical to the MCV config I had before):

  • When this happens:
    Coffee Maker is turned ON
  • do these things
    Wait 401 seconds THEN Stairway Lights: Turn switch ON

It just doesn’t happen. Even the ST-provided app to turn off the same ‘coffee maker’ switch after 2 hours doesn’t work. The platform is fundamentally broken, and unfortunately I can’t just keep hoping it will get better “someday.” I support the current focus on scaling, but this stuff was hit or miss even BEFORE the scaling issues.

1 Like

Platform / Ecosystem abandonment disorder.

When this happens and momentum builds it can quickly become an exponential effect. RuleMachine left. SimpleRulesBuilder may not care enough anymore to fix their system. (not speaking for them, hypothetical). Users leave because of this. Now less and less devs and users care. Morale internal to ST suffers… etc.

The ecosystem suffers, users bail, devs bail… collapse.

Did ST break SRB… or fail to give them proper and timely information on updates… maybe not… don’t know. Are they free of culpability? Not necessarily. Folks want to invest in those things they can have faith they will get a return on. Users/Devs/Employees. When that faith is shaken…

2 Likes

Finally we found something we agree on, but let me see if I can change that… :wink:

The problem SmartThings faces for the indefinite future is how to keep “platform/product” stability as #1 Priority, while giving major developers nearly equal time, features, and support.

@pstuart just did an ad hoc survey that focused on the experiences of Community Developers. Every developer knows that platform stability is fundamental to their own success, but assuming that all the consumer facing issues are resolved in 6 months, there are an equal number or more of distinct issues that face developers. We can’t wait 12 or 18 months (speaking for SmartTiles…), so the only hope is for SmartThings to solve things concurrently with excellent planning, resources, and management.

Why doesn’t SmartThings have these characteristics after over 3 years in business and nearly 2 years under Samsung?

2 Likes

I guess I’d better get my ST bits and pieces on eBay immediately.

I guess I’d better get my ST bits and pieces on eBay immediately.

I wouldn’t bother. Standards-based kit is going to work with any controller, and the hubs are cheap. (And who knows, maybe in a year or two they’ll have this mess cleaned up…)

They are a newage launch vehicle company. I think it’s a very interesting business model and welcome great success stories from it, but like all models it is not perfect.

The model often means launching with very slim resources, neat ideas, half baked architecture, chocked full of idealistic but not necessarily realistic ideas, as a service, contrived cloud, bigger than life promises, etc.

Often the new model can ignore profitability and revenue based on the idealistic view that if you have a great idea you can figure out monitization much much later. Acquisition is quite often the actual goal. Then the acquiring organization is left to figure out their monitization scheme as founders monitized through acquisition.

If ST’s monitization is through data, this is new territory. Who is the consumer of said data? What do they pay? What motivated them? It could also be a way for Samsung to claim dominance in HA and worry about details later. We can speculate, but we don’t know.

Does Samsung have a strategy here or were they just in acquire or die mode in terms of HA? How are they monitizing? How will they monitize? Right now they either have some very inept leadership, or they are starved for appropriate resources.

2 Likes

Z-Wave network allows multiple controllers and does not depend in any way on the ST cloud. Therefore, you can have full local control, using your favorite Z-Wave remote, regardless of whether ST hub is involved or not. The hub just gives you another control point that can be accessed remotely, which is not possible with X10 without third-party hardware. Just my 2 cents…

2 Likes