I am the only one who is starting to think that Samsung’s vision for SmartThings is moronic and they are running this platform straight into the ground??
When the app first came out it was great. So much flexibility. A bit of a learning curve, but it just worked.
Flash forward to today, almost 5 year later for me and Smart things is just stupid. There is almost no functionality left in the app. There are exactly 13 smart apps available, 10 of which noone will ever use. It’s a fricking joke. Stupid train is literally off the track.
Please Samsung, fire all your project leads and managers and put some fresh innovative minds in place to bring this platform back to what it once was.
You need to fix this stuff or this platform is doomed. Firstly, why do you only have 13 ridiculous published smart apps in the app??
Most of them will never be used by anyone. Secondly, the stupidest thing ever is why isn’t there some sort of button to search community provided apps??
My frustrations came to a head today, when suddenly my gradual wake up disappeared. It has been running for years and it is literally just gone. Looking through the app, there is literally no way to add it back. I am well aware that I can likey restore this functionality from the IDE, but do you really think that Grandma is going to be able to figure out how to do that? I mean does anyone believe that it’s okay that I can’t gradually fade on might lights(literally the first thing that many people try to set up), but there is an app to turn on a vent when co2 levels are high or operate smart windows that three people in the world own. It’s completely mind boggling just how idiotic the few published smart apps are and just how few useful features are available in the app.
I am seriously concerned. Fix this and fix it now. You’re going to have to fire several people because the people currently making decisions are completely out of touch with the reality of what people want to do with home automation.
If Samsung does not make changes, this platform will be dead in a matter of years.
Thanks for this…it actually is still there so no idea why it’s not working. Also I have no idea why that menu is placed there and not in the automation section. It seems like they just randomly dumped in it the corner.
This changes nothing about my sentiment though. This functionality has been experimental for the entire time I’ve used smart things and I can’t really recall any functions being added over time. Things just disappear.
So now, three of the most useful features (presence and timed light dimming) are hidden in a randomly placed menu and still experimental 5 years after they were published.
Is the plan that they will just be experimental forever? I mean for god sakes, this is 150 lines of code.
The entire approach has changed, so that now the automation creator which is built into the SmartThings app is much more sophisticated than it used to be. You can do all kinds of things with it that you never could with the original smartthings routines, including wait steps, duration checks, multiple Ifs, preconditions, etc. So I think that’s intended to replace the whole idea of smartapps for the typical smartthings customer, Who has, according to previous staff postings, fewer than 15 Devices, and never uses any custom code. And these days, probably doesn’t even have a hub.
In addition, they are providing the new “rules API“ which is essentially a scripting language for smartthings, similar to the community-created Webcore, and provides a much deeper level of complexity for power users.
So basically the need for individual smartapps is pretty much going away. You can do a lot more with the built-in rules engine, and if that’s not enough, they want you to use the rules API.
I’m not saying that will do everything for everybody, I just think that’s the explanation of why the app doesn’t provide much in the way of finding smart apps anymore.
All of that said, Samsung has done a terrible job of providing tutorials and examples to explain to people what is available to them through the built-in rules engine, you just sort of have to figure it out for yourself, or come to this forum and ask. There’s a huge gap between what the platform can do and what most customers are aware of, even customers with strong technical backgrounds. It can be very frustrating.
They should at least provide a “gallery“ like Shortcuts and many other rules engine platforms do with samples of commonly used Automations.
While I do agree that moving to the IFTTT model makes sense, I completely disagree that this functionality wasn’t always there. There is really not anything that I can think of that is actually useful that I can do now that I couldn’t do in the past. It, from a GUI perspective is easier to configure now, but that is about where it ends for me.
I just turn my lights on and off and schedules, have my cameras, harmony, thermostat, and smoke detectors linked. That is pretty much it and I’m betting that is the case for 99% of the community, yet I’m constantly presented with these bizarre smart apps and missing functionality in the routine model.
If they are going to add a gradual ramp up model to the automation section, hey great. It’s about time. I mean when is it okay to take 5 years to correctly implement basic functionality that was removed when they decided that they needed to dumb down smarthings?
Sorry if I was confusing. I agree all of the functionality was there before if you used custom code, I just said it wasn’t in the “routines“ which is what the new automation builder has replaced.
And they’ve done a really bad job of explaining it, whether you are a new customer or transitioning from the older platform. IMHO, of course.
The new app is an embarrassment, so is anyone who spends time trying to defend it. You can’t even see what Mode your house is in, it’s probably in there somewhere, but I don’t need some geek on a fringe forum to explain the correct sequence of button presses to find it. It feels someone is deliberately trying to make it less user friendly.
so i have an old android phone 7 that is my on-the-go cheap (read: if it breaks its no big loss) that is stuck running the previous version of the current app. i find myself going to that vs a new phone running the latest update of the current app. i try to avoid the app right now by using buttons.
Believe it or not, adding your own custom device handlers to the IDE is surprisingly easy these days for someone with general IT background and a little bit of coding. It not the IDE, side that bothers me and ST-anything actually works quite well even though it is a community project.
My big complaint is that basic functionality is missing from the phone app and all kinds of weird niche functions are everywhere. My Second big complaint is that things go into experimental mode and stay that way for 5 years.
Then of course there is the communication (or total lack thereof) of what they are changing. And when you go searching you have to again learn a new language to figure out what they are talking about. A little clarity goes a long way.
For example if I told you 4 new THEN actions are available today, you would know right where to look and what they were for right??
Unfortunately, SmartThings has made the decision to discontinue hosting a public groovy cloud, so that type of custom code will no longer work by the end of this year, and the IDE itself is going away. . They have announced a general timeline for this.
So we don’t really know what the options are going to be as far as custom code for hub connected devices going forward.
I guess they have one chance to impress me. While I said “pretty easy” honestly I have never been a big fan of “groovy”. I don’t know java so groovy is even worse for me, but I can still mad-lib the templates they provide.
It mentions moving to an API as a replacement. I’m not sure how this will be implemented, but if done correctly, would allow you to write custom automations until your heart is content in whatever language you prefer.
Honestly though, I’m expecting an overly restrictive API and no way for an average Joe to turn some ordinary device into a “smart” device. Their claim of optimizing for Java bothers me too. Can we please stop using java already?? If they were trying to modernize, they would be moving to Python. Maybe I’m just being pessimistic but they’ve given me no reason not to be.
They might as well have just added the implied statements at the end that says “in the meantime, we don’t give a ?!&& if your lights stop working or you can no longer do certain things”.
If anyone has any further information in this API model, who will be able to program what, whether there will be any ability to implement say a raspberry pi or a Arduino etc, I’d be interested. I’m going to have to decide soon whether to continue down this path, or to switch to one of the open source hubs or one of the competitive products.
They put out a lot already on how smart apps will work. You can write them in any language you want, but you have to host them yourself. And there is a quite robust API which is published that tells you how to communicate to the smartthings account. So that part seems to be OK as long as you’re willing to put up with doing your own hosting.
They’ve also published a lot on cloud to cloud integrations for DTHs, so again, that part seems pretty clear, but again you have to do your own hosting.
What we don’t have details on yet on how custom DTHs for hub connected devices are going to work.
Details on what has been documented so far is all on the developers site… Note that the “rules API“ is different from the API for communicating from your own smartapp or for communicating as a cloud to cloud DTH.
Just noticed in this post the link to the “rules api” forum. I will look through here and see if I can make some sense out of how to get started with it and what it can do. Just browsing the first few posts is not really inspiring my confidence so far…
I just updated my post to include the link to the new developer site, I think that’s where they want you to start now. But I’m not familiar with the details myself.
Forgive me for having to ask clarification, but what is DTH in this context?Google says DTH is “direct to home”, but it’s not making sense to me in this context.
DTH is “Device type Handler.“ One of smartthings’ many names for a Device driver.
They are also called “device handlers“, “Device types”, and a couple of other things.
The following community FAQ explains the terminology and the concepts. It was written for the 2015 version of the smartthings app so some of the publishing steps are out of date, but it’s a good basic read anyway as far as understanding the approach for adding custom code to the groovy cloud. (The topic title is a clickable link.)
And for giving you some of the terminology that will still be used with the new platform.
Got it…yes I just found it as well. Guess I should’ve searched “SmartThings DTH” instead. I do know what a DTH is and have created/ modified a few myself (working on one for esp8266 & wled very soon actually). I guess I have just not spent enough time in these forums to pick up on the acronyms.
Thanks for the links and helping me find the gradual wakeup “experimental” smartapp.
Samsung honestly still has some proving to do to me at this point. I honestly hope they are able to pull this together and expose an interface that works the way it should.
This “host it yourself” and authentication model talk is interesting. It’s confusing to try to understand whether this is in the context of “I as a device manufacturer” or “I as a consumer” (or maybe both) guess I have some reading to do.
. They have announced a general timeline for this.
