Who designed this UI?

Pardon me sounding facetious, but why is changing “shake to view labels” into some other simple option like a toggle switch or a preference option for just brute force label overly, such a big deal? I honestly ask this with zero sarcasm. I just would like to understand … and would like to empathise with ST and the team.

I know … there have been a dozen forum threads on exactly this topic, but never an answer that explains why it is either controversial and/or difficult.

Once it’s fixed, the explanation will no longer be required or relevant. But in the meantime, you have a very curious Community. I am personally curious, just to help me understand the complexity that a company like SmartThings faces in designing and maintaining just one portion of their product (the Mobile Apps).

For reference:

And older:

And older still… (August 2013!!!):


Actually, I miss the days of @smart throwing a Taylor Swift reference into the fray of discussion about shake for labels :wink:


Don’t feel good these days, my friend being a heart patient and chronic pancreatitis and going thru the chronic pancreatitis attack! So kind of shaky right now! Soon buddy we are going to get it rocking again! :slight_smile: it’s going to be swift! Missed two days of work… And well certainly was not shaking it up with Swift! But Aaah! Now you have my heart beating fast.

1 Like

Ooooo… Get well soon, bud! You’re next smart devices may be new organs! :sunny:


Get well soon with "bud"weiser! You meant right? :wink: but still doing 6-7 miles a day walking with the crappy UP2… So very soon the the whining will start… And @april will get her hands full. :slight_smile:


Hope you feel better soon.


I’ve done software development, and you typically have far more things to fix than you have time. Yeah, this might be a small thing, but you’ve got 50 or 100 small things on your list. So you prioritize. And unfortunately, depending on the code, what appears form the outside to be a small thing just might be a medium or a really big thing. The related problem is that as a system gets older and has absorbed many fixes and changes, it gets kind of spaghetti/rube Goldberg like- hard to understand and complicated to fix simple things. You discover that the product is going a different direction than you anticipated, so your overall design doesn’t quite work any more. So you have to re-engineer everything: now, do you continue to update the existing, or focus entirely on the new, or (usually) some compromise in the middle. So what I look for, is a) is the product moving forward (not looking for everything, but real progress) b) do problems get addressed (big problems more quickly, smaller problems more slowly, no problems hang around “forever”) c) Is there customer focus, including honesty with the customer community. You can have lots of product evolution, but if it’s not what the customers need, it’s not helpful. Caveat: cut developers some slack- sometimes they know things they can’t share. If they’ve been honest all along, you can live with this.
Interestingly, vibrant, critical user communities can sound very unhappy and be very critical, but be very happy when they know someone is listening and responding.

1 Like

I have a similar impression of much need for change, but here are my initial thoughts:

A default dashboard makes sense, but to me, a dashboard should show me status in an efficient way and perhaps give me access to my most common use case direct controls of the system. This “dashboard” shows me a bunch of mutually overlapping categories of things that I can dive on, but tells me very very little by itself. For example, it tells me the number of things I have connected. Is that really a piece of information I’m interested in looking at very often? Almost never. Perhaps I am interested in knowing if all of my devices are currently connected or if something has unexpectedly “dropped”. So perhaps a small piece of real estate dedicated to “connection status” that indicates “ok” or “trouble”. If it’s “trouble”, I can dive to see what unexpected condition has occurred.

The top third of my screen or so is dedicated to an image. Seems like a pretty big waste of valuable real estate in a dashboard.

It would be ideal if my SmartApps could register status indicators on my dashboard. For example, alert indicators for known conditions of concern like closed but unlocked side garage door for more than 20 minutes. Status indicators for the sensors I choose to put on the dashboard. And so forth.

One can dream. :smile:


Bob Florian demoed upcoming “dashboard Solution Module” enhancements (I’m not sure of the proper name) at the last open Developer’s Call (video soon).

Developers currently can’t modify these, but in upcoming release, someday, we can code rich custom pages for the Dashboard.

1 Like

What people are wanting here and what you’re referring to is just fundamental good designing for a user: giving them what they need, when they need it, without having to read documentation.

I’ve always felt SmartThings was developer designed; It never felt like it was designed out from the start to have a great experience, and then developed to fit that design—it’s always felt like the development was done first, and the design tacked on. And great design isn’t a coat of paint.

And I’m not faulting them for it. I’ve been at development-led companies like that, where things are bootstrapped at the start, and they’re trying to get a product out quickly. So instead of spending some time and money designing it out right (with a designer), developers are expected to design as they code, and then maybe a guy who’s good at photoshop is brought in later to “pretty it up”. And it’s products like this that are the result. Great potential, passionate (and patient) users, great developers, functional… but just lacking a really great user experience.

Because it’s hard to design and develop at the same time, or to design after the development has been fully baked and the big decisions have been made. Great developers aren’t always great designers, because developers are great at realizing edge cases and making something do all it could do, instead of all it should do. And when you put those two things together in a design, you get complicated products.

I’ve gotten by with the way the UI is, but it always annoys me, and in its current state there’s no way I’d recommend it to anyone non-technical. I’d love to redesign it.


Quite right, Jaff. To me, ST is Linux, maybe ten years ago: You can argue that it’s better than anything else out there. A power user’s toy par excellence, you can make it do anything you want it to, but only if you’re ready to delve deeply under the hood;. However, if ST ever wants to be embraced by the wide world of people who’ve never needed to deal with a command line and who don’t care to begin now (and yes, that’s almost everyone, folks), they need to show some effort towards optimizing the user experience for those people.


Wow surprisingly, even today in 2018, almost none of the feature OP requested are here in the current SmartThings app… Amazing… particularly grouping devices(not to be confused with room setup, grouping devices is for controlling multiple devices at once), and the “things” page desgin…

I mean how hard can it be to put a sorting and search function to it? Seriously Samsung?

There is scenes where you can control all devices at the same time.

1 Like

That’s not the same though, and don’t forget scenes can only control lights now. What if I want to control multiple outlets? Grouping is still an effective method to do simple task without creating smart apps and scenes for the setup.

1 Like

I agree.

But the greatest thing about SmartThings is it’s extensibility.

Search the Forum for the SmartApp “Trendsetter” (actually… I’ll link it below). It provides intelligent and powerful light grouping tailored to Community user requests; perhaps better than SmartThings would or will ever implement. We recommend Trend Setter to all ActionTiles users.

Thanks a lot Terry! I’m actually using it :grinning::grinning: Just saying native support would he more than welcomed. But yeah trend setter is very robust.

Well I don’t think they’re doing any enhancement now since a UI overhaul is coming (hopefully in months as they said).


Never hold your breath. 2 years is also 24 “months”. There’s no reason to believe SmartThings’s pace of development is going to substantially increase - and for the sake of stability and reliability, it must not be rushed.

1 Like

And then equate that into dog years.


For their record of reliability, I’d rather they stay the way IT IS :pensive::pensive:

It’s gonna happen! You just don’t know when :joy::joy: