Yup I’ve been using ST for a while and I thought they got rid of it because I used the lighting automation wizard and figured it would spawn the lights dashboard tile. A shame they didn’t make that automatic when lights are detected. To my other point… I don’t know if I want it actually…
Just one more layer of things to click.
I don’t think this is actually what the problem is. The problem is that when you finish creating the first automation in Smart Lights, and hit Done, it blows you back up to the Marketplace menu. I’ve made this criticism of the UI in other contexts. In general, finishing an app install blows you up a level in the UI menu structure, when it really should just leave you where you were. The old app left you where you had been (except for shortcut apps). I will send this in to support again.
If they fixed this one issue, it would clear up some of the confusion in the UI.
I agree, I am a novice ST user and only do basic stuff with it, but it took me a while to figure out you don’t need to keep new instances of Smart Lighting. Once I realized you could control all your light automation’s with one instance of Smart Lighting it was very easy from that point on, but getting to that point is not very intuitive for a new or novice user like myself. However, the way the app keeps adding multiple instances of Smart Lighting is counter intuitive and complicates an otherwise very easy task.
As I previously mentioned, Smart Lighting is also not sending the setLevel command in certain instances. It recognizes that GE Smart Dimmers are dimmable, so gives you the option to set the dimmer level, but then fails to send the setLevel command. It does send the setLevel command to my Hue and Lightify devices, and if I include one of those in an automation with GE dimmers it will send the setLevel command correctly, and this is how I’m working around it right now, but it’s not how I want to do it.
I’m a novice user and it did not take me long at all to discover that setLevel is not being sent to GE dimmers.
THANK YOU, this was the piece of the puzzle I was missing! I now understand why it’s called “Smart Lighting” as it’s providing me the opportunity to remove a bunch of discrete smartapps I had for unique purposes (not to mention just one place to go for lighting needs).
I can certainly empathize with your befuddled friends… and I’m not a new user. Under the old method, you could populate a smartapp appearance by selecting it from (for example) Green Living, and then tap OK through all the screens, thus creating a “dummy” or blank instance. Now however, you have to create an actual scenario, which leaves the impression that’s all there is. I had been trying to set a light to come on at a specific time and turn off at 12 minutes before sunrise, and when I couldn’t find a way to do so, never bothered to look further.
I hope that makes sense, and thanks again for the detailed explanation.
I feel compelled to throw in my 2 cents in regards to a couple of points raised in this thread.
SmartThings may not be a “coder’s platform,” but the UI is very clearly designed by coders. As a programmer myself, I have to say that we are almost always terrible at designing usable interfaces, specifically because the interfaces we create tend to reflect the underlying software architecture too well. Every moment spent using the SmartThings UI, I feel like I’m looking at a design document for the code.
This is leading SmartThings to an interface that, while perfect for some, and understandable with a little effort for many, is completely unapproachable for an “average” user that wants their home automation to just work ™.
Let me try to illustrate this with an example:
In the programming world, separation of concerns is paramount. A light is a light, and it has very limited responsibility: to turn on and off and to report its status. A Smart App, Smart Lighting in this case, has one responsibility: to resolve its internal logic into a single on/off command to be dispatched to a light.
However, in the Smart Home metaphor that an average user is building in their mind, those separations prevent them from creating an understandable mental model of their home’s behavior. To them, the light and its behavior are inextricably tied together, as one without the other is useless.
Looking at how this mismatch in models plays out, let’s pretend I need to add a lighting rule.
As a programmer, I clearly I need to go to the Smart Lighting SmartApp because the rules I want to add are obviously implemented in the Smart Lighting class/module. I then have the option of piggy-backing on the existing Smart Lighting instance, or creating a new instance, and they can both live happily next to each other, each doing their own thing. It’s beautiful in its simplicity.
Even the fact that this class happens to be controlling that light, or really any light in particular, is an implementation detail that should of course be abstracted away, and my heart sings at how loosely coupled the logic is to the items it controls.
As an average user, I do not think like this. I want to add a rule to my light, so my first step is to go to my Light. There’s no place to add a rule here, though. But let’s assume I realize this, and find my way to the Smart Apps. I figure out that Smart Lighting is a metaphor for whatever invisible construct is responsible for controlling my lights, so I try to add a new rule there. But now I’m confused because there’s no clear indication of the pros/cons of creating a new instance, nor is there even a clear indication that what I am doing is creating a new instance. But I manage to create a new rule, and now I have another Smart Lighting icon, and there’s no clear indication that either of these are tied to any light at all, much less which light in particular.
I wanted to add a simple rule controlling my light, but now I’m dealing with several levels of abstraction that I am not trained to deal with. I guess I should have read several support documents and browsed the forums in order to understand that my simple mental model of a Smart Home is wrong, and that there are really a complex set of interacting constructs that are responsible for my simple rules.
That SmartThings’ UI is a good as it is speaks volumes towards the programming talent they have. However, specifically because of the UI I would never recommend SmartThings to a non-technical person.
I love SmartThings, and think it has more potential than any other competing device on the market, and it’s going to work wonderfully for many people. However, the Best Buy crowd is not going understand it, and they are going to get frustrated. Without them, SmartThings will at best become a niche product, and at worst be doomed to cede the market share it has to more intuitive products and eventually be shuttered.
Please SmartThings programmers, give some more careful thought to the UI and how the people who are ultimately going to need to be your target demographic are going to respond to it.
Edit: I also have to address one more point: verbiage, verbiage, verbiage. What is a SmartApp? At the end of the day, it’s a set of Rules for dictating behavior. When I do go to my Things, I should not see a SmartApp header. I should see Rules. New terminology should only be used when there is no sufficient term already in the general lexicon. It’s a little thing, but I think significant. Modifying a SmartApp is daunting, and I could probably screw something up if I’m not careful. Adding a Rule is simple and fun.
While I agree wholeheartedly, you must realize that SmartThings’s marketing / strategy department fights this tooth and nail, as they feel unique terminology is valuable essential and fundamental to product differentiation.
It’s why Google calls their Android App Store the “Play Store”, when I haven’t installed a single game to “play” with on my phone.
Similarly, the “programmer” bias in SmartThings is clearly in conflict with their DIY strategy. The imperfect UI was designed by an IoT UX “expert”.
SmartThings’s problems are far more organizational than technical. Poor leadership.
ST was criticized for not using conventional terminology more than once on the Forums, but we’ve never got an official response. So the assumption is that they wanted to be “original” and not bound by existing pre-conceptions. That’s why they had Hello Home “actions” (now “routines”) instead of “scenes” and “smart apps” instead of “plug-ins”.
I don’t think it would be accurate to call smart apps “rules” though because rules assume a simple user interface that does not require learning a programming language or using an IDE.
No disrespect to the individual in question, but there are no IoT UX experts. 
It’s a nascent field that is just beginning to show its real potential and enter the general marketplace in a meaningful way.
That individual is an IoT UX pioneer, and I respect them greatly for that. Part of being a pioneer is taking risks and making mistakes. But they (SmartThings) are making some mistakes, and now is the time to address them. 
@dansor, adding to your point, there is no such thing as an IoT UX itself in terms of Home Automation… cuz lets be honest. HA != IoT… many of the devices CAN connect to the internet, but look at what SmartThings JUST did with V2 Hub… alowed LOCAL connectivity… so its not an IoT platform @ its core… its an HA platform…
And then we get into the discussion on Home Automation vs Home Control… and yes… that’s the balance that SmartThings needs to strike here… Automation of your home… vs discrete control of the items in your home…
Its a tough one, and i don’t envy them… but someone else said… let us make our own Dashboards and we’d be able to strike that balance ourselves… just hasn’t happened yet!
I think this is a super valid point. I find myself (relatively experienced with ST and coding), wanting to be able to add a SmartApp from the device detail page. You can look at the SmartApps already touching that device, why not just add a link at the bottom of that list that allows you to add an app. That link need not be anything more than a link to the Marketplace SmartApps page. That would certainly help complete the concepts you are pointing out.
Coming from Wink to Smartthings I was expecting a similar experience but better, since almost all of the research I had done mentioned how much better and more flexible Smartthings is. And that is true, sort of. I quickly came to realize that the platform itself is incredible. I’m a software engineer and I have a sense of what it takes to create what they have. The ability for users to code and add their own Smartapps is really impressive. So on the one hand I can code myself some Smartapps and do pretty much whatever I want to do. Great!
But I don’t necessarily want to do that. I have less patience for tech stuff at home when I do it at work all day and I prefer simplicity. HA should be about doing less, not fiddling with apps and troubleshooting constantly. When you get to what the average person who can’t code or do all the community developed hack-arounds will be able to implement in the Smartthings app, it is not that impressive. I’d go so far as to say it’s extremely inflexible and things you’d assume would be trivial or options you’d assume would be there are not available. Wink lets you do what their platform is capable of much easier. Even though their platform is not as advanced overall, people will be happy with what they can accomplish with it.
I’m sorry I keep bringing up Wink but that is just where I am coming from. I am hoping the new ST app gets updated and all of these deficiencies get fixed, but my sense from this community is that people have been playing that waiting game for a long time now and it is what it is. ST makes a bad first impression which actually detracts from the accomplishments of the team. They will get a ton of returns at Best Buy. I am sure of that. It’s not ready for the general public. What I wanted was an easy solution for some HA tasks, instead I got a fiddly hobby. I am OK with that. I am keeping mine even though I think the only way to accomplish many of the things I want will be to write the apps for myself. I think it’s awesome that I can do that, but am less enthused that I have to.
Please allow me to add my humble input here. The only reason I haven’t made the jump to ST has something to do with what @JDRoberts explained. Right now, my Wink hub supports my Nest products in an official manner, adding products is a breeze and it works consistently. So why I am following the ST forum and thinking of jumping ship? Wink’s future is on a grey area, speed after I turn lights off/on and power of the app itself.
So why haven’t I switched to ST? Wisdom from the wife. She is pretty tech savvy and likes the whole HA idea. When I explained why I was considering ST she said I’ve been spending way too much time in forums trying to see how it worked. And it was true. I found myself looking at how to add a SmartApp, which SmartApp I needed to use to connect my Nest thermostat, having to use IFTTT, etc. I consider myself a geek and would definitely be able to manage the whole thing, but it’s just the time I have to put into it and then having to teach how to use the app to my wife what’s keeping me away from ST, for now. Well that and waiting to see how the new hub fares. I don’t want to be a “tester” for it and you can’t really buy it right now.
So, I don’t think ST is only for coders, but it’s not for the average person out there. People want for ADT, Control4, Pella, AT&T, Xfinity type of experience. Get someone to hook everything up and show me how it works. Hell, even Wink hooks you up with someone to install your products for you.
I personally think ST is somewhere closer to the coding side though, or at least to the side with people that don’t mind getting their hands dirty to get some of the “cool” features to shine. If Joe “what’s a router” Smith is a 1 and John “I wrote my own router firmware” Jackson is a 10, I’d say ST is somewhere around a 7, probably an 8.
The “problem” is that there are a lot of other products out there and some of the other ecosystems “appear” to be easier to use, or more integrated… I just went to Lowes today and got a rebranded Linear GD00Z-4 garage opener, complete with an Iris logo and “works with Iris”, “requires Iris Hub” and all that stuff, even though it’s clearly a generic item that works with ST.
Getting them to work with ST was a snap. Now getting ST to do the cool stuff like notify me or automatically open/close doors is unnecessarily convoluted, and I think that’s the general idea of this thread.
Ha, I love the reference bar. True, it is around that area.
Yeah, Lowe’s has this sort of things. At Lowe’s you can get the GE Z-wave dimmers usually for less than Amazon. It is the same one, they just put the “Works with Iris” stickers on top of the Z-wave logo. On the side it has the usual Works with ST, Wink, Revolv, etc.
I sure hope ST gets this whole idea and tries to get it as user friendly as possible.
So true. I paired a Lutron Connected Bulb Remote to wink today, and it walked me through every step. It was actually overkill for me, but I think it made the process near foolproof. To do the same with ST (assuming someone here adapts it for our use) I would need to login to the IDE and snag some apps and device types from Github. Another day at the office for many here but very off putting to the average entrant into HA.
This is one of the best worded comparisons I have read… and precisely why I got my brother a Wink system last Xmas instead of SmartThings.
Can it not be both? Not sure why there is so much backlash at calling something IoT versus HA. At its core SmartThings about connecting your things in an way that makes your home smarter. Internet of Things is about connecting your things and making them smarter. Your Smart Home is just one peg in the larger internet of things scheme. If this is strictly home automation, then why connect your car?
Is not whether the platform was designed for or by coders. The issue is that the platform is currently designed for hobbyists. If you pick up SmartThings and go with it today, you are probably and early adopter hobbyist. The concept of just having your things work together in harmony right of the box is still way off in the future. There is not an HA/IoT platform in existence yet that will allow you to walk into best buy, dump some stuff in a cart, bring it home, and have it do everything you want it to all with little to no configuration.
Wink
Wink is easier because it’s a system on rails. As long as you want to do the things they have set up for you, you are good. The minute you want it to do something that they have not pre-programmed, you are going to hit a wall and will be stuck.
SmartThings
Is not as pretty as Wink because it’s not as closed as Wink. Does the design need to improve? Yes, without a doubt. The saving grace of SmartThings is the active community and the fact that SmartThings employees are regularly participating in it.
There are really no experts when it comes to any of this stuff. Sure there are the old guard of systems integrators that could be considered experts in certain areas like home automation interfaces and they will gladly build you a custom solution for many thousands of dollars.
When it comes to a consumer driven solution that allow you to manage and control a wide variety of different devices from different manufacturers; no one is an expert yet.
The platform needs to mature from a hobbyist solution to a developer driven platform. Hobbyists tend to build things for themselves. It’s currently a moderately difficult task to get those creations and improvements back into the system for everyone else to use. If this platform were a developers platform, all consumers would benefit because you would have professional developers fighting to make solutions that everyone could use.
I doubt anyone would call the iPhone a coders phone, nor would they say that about Android. Both of those platforms have made huge efforts to draw in a vary large and diverse following of developers.
There was a time (1975-1979) when “personal computers” were hobbyist (Altair 8800, et al). Then it transitioned to a manufacturer driven platform (IBM), and much later to a developer driven platform (Windows, etc).
We are in early days yet…
That’s true but the speed of innovation has climbed with Moore’s law. It only took Facebook four years to reach a billion users. They have enough developers now to warrant a yearly conference and they have paid out over eight billion dollars to developers.
Granted, smart homes will take longer than social networking to become ubiquitous as there is a higher barrier to entry, but the tools and the developers are already out there waiting to be tapped into.
I agree. I wonder about consumer adoption though. It’s not clear to me that HA as we now know it will become very mainstream, beyond an endcap at HD, irrespective of technology. It will be a long time before the home builder community embraces it (who needs the hassle). Time will tell, but this isn’t like a cell phone or a pc or a tablet.