ST App the weakest part of the system?

I have waited to weigh in on this topic to see where it went. It is a bit off-topic now so I will just say. No. The SmartThings app is much better than most other home automation apps out there. If you disagree I challenge you to point me in the direction of the better app. If you think we haven’t seen them all, you may be surprised :slight_smile:

Our app is heads and shoulders above others out there. Prove me wrong.

…gauntlet thrown… :slight_smile:

Ben, I will go out on a limb and say that the HA app for Control4 is miles, and miles ahead of Smartthings.

  • Simple room organization, ability switch to another room to see devices Clearly labeled icons and ability to reorder them by room
  • Ability to trigger custom buttons / actions in one click and by room
  • Media Lists for audio and video sources
  • AV proxies that are consistent across all devices, just need to have a driver that hooks into those proxies Security
  • System virtual panel
  • Windows and Mac App along with iOS and Android
  • Lighting scenes
  • Media scenes
  • Room based audio and video sources and control
  • Ability to see Audio and video sources across rooms

What Smartthings can do that control4 can’t:

  • Add devices (dealers have to add in control4 via a windows application)
  • Add SmartApps (users can add 3rd party apps to flash navigators but no iOS or android support in c4)
  • Code (and I use that term loosely) basic automation activities
  • Understand basic user awareness via presence sensors / phones

In my opinion, Smartthings app is way behind most HA systems that dominate the dealer model. Crestron, Control4 and others. They simply work. Smartthings App isn’t even in the same ballpark.

Smartthings app is ahead of most DIY HA platforms, I will grant you that, but if you want to play with the big boys, you will need to add a ton of functionality and simplicity before I would even remotely consider the apps equal.

In terms of usability vs control… This will always be the argument.

I’m willing to show you my control4 set up any time.

2 Likes

You cannot just say the app is good or bad. There’re different aspects to it. It excels in some areas and sucks in others. If we’re going to rate the ST app, we should define some specific criteria.

As far as real-time control, I’d give it 2 stars out of 5 for the following reasons:

  1. The grid interface is clean-looking, but it’s not suitable for real-time control. It requires too many steps to get to the device control view, for example to change dim level. Trying to hit a tiny ‘gear’ button to open device settings is particularly annoying.

  2. Although ‘things’ have a ‘name’ and a ‘label’ (which is confusing in itself), none of them is displayed on the grid. Someone even suggested creating custom icons with embedded labels to work around this issue, which should be enough evidence how desirable it is.

  3. There’s no ways to create custom views, other than bundle things into ‘groups’. And any ‘thing’ can belong to just one group. At very least, I’d like to be able to view devices by type, for example all light switches, as well as by location, e.g. all devices in the kitchen. Ideally, I’d like to create arbitrary device groups, e.g. ‘my favorite switches’.

  4. Creating scenes by installing a custom app is cumbersome and is rather a ‘hack’ than a feature.

And of course, lack of native iPad app with landscape view is simply embarrassing. I mostly use iPad in landscape mode and having to reorient it every time I start ST app just makes me want to wipe it out.

Well, there it is. Please don’t soot the messenger… :slight_smile:

2 Likes

Could you post some screen shots of your control4 UI?

Here’s just a few threads asking to add labels to ‘things’ view:

Today:

Jun 2014

Jun 2014

May 2013

Feb 2013

I could, or you can watch this video at the bottom of control4’s website…

http://www.control4.com/o/myhome

There is a ton of things I could take a picture of, what specifically are you looking for?

Geko, I just found a stupid simple way to get labels in the things view.

Shake your phone. No seriously. Shake it like a Polaroid picture… Viola… Labels.

Why isn’t this the default view?

Yeah, it’s an old trick. I challenge @ben to show me another app that has the same great ‘feature’. I hope it’s not what puts ST app “heads and shoulders above others out there:smile:

Isn’t Control4 incredibly expensive? I’m not exactly sure it is a fair comparison to ST since ST does not do media, etc. Control4 is in Crestron territory. I think they sell to a different market than ST, Revolv, etc would sell to. All those are actually attainable by the average person.

I think that comparisons to other apps are only good when you are comparing the same category.

I’m also not sure this thread has split into a useful direction. I think the approach here should not be to slap ST down. I think it should be to talk about what we think will make ST better. Some of that might be borrowed form the high-end platforms. Some might be borrowed from competitors. Some might be stuff they come up with (and I’m sure they are some good things coming since they nailed the open aspect of the platform). Some of it might even come from us.

We can meaningfully contribute, as owners, as developers, as users. That is what we should be focusing on. If not, twelve posts from now someone will bring up Nazis. :smile:

Godwin’s Law anyone?

I don’t consider Control4 incredibly expensive. Everything else you connect to any HA system has been expensive, but ultimately, a control4 HC250 (the basic controller) can control a room full of zigbee devices for under $1000 US.

Granted, Control4’s initial focus was on a better remote control and then moved into HA. They are separate product categories, but I doubt the media and consumers will care about that distinction.

I want to see ST move into complete HA space, with mainstream product support and simple / easy to use functionality so it can one day replace Control4 and others in the high end space.

The biggest weakness about Control4 is it is only sold by dealers. Some dealers are awesome, others not so much. But one thing is for sure, it works and is solid (and doesn’t require an internet connection to function).

The place where ST can improve dramatically is the ST App. It is the weakest part of the system because it IS the system or at least the main control for the cloud…

4 times in the last week, support reminded me to use the phone app to make changes and not the developer portal… It’s just too easy to use the dev portal, versus trying to find where that uninstall button is to remove a smartapp, or why I can’t remove a device because its hooked up to smartapps, etc…

I took a few screenshots of Control4 App interface in case anyone wants to compare…

https://plus.google.com/109218861040495449466/posts/J9jXdB7R8pg

1 Like

I think the key point is $1000. :smile:

That is well outside the comfort zone of many people. We use Control4 at work to control our conference room. I really hate it that I have to get the dealer involved when I know I could make a lot of the changes myself - I am an IT Director after all! But, they don’t like to show the process.

I also think that Apple moving into the arena will push Control4 back up to the high-end automation market. Apple will help democratize everything and I think that will help ST a lot.

I still think there are improvements to be made. But we all seem to forget - the app is at version 1.6! You can tell from the posts here that the ST team believes in their product A LOT. Alex is definitely making the rounds on the publicity circuit. Every time a turn around I see him! So, I’m confident I’m on the right bandwagon at this moment.

1 Like

I’ll respond to a few things here, but I do want to take a moment to recognize the subjective nature of user interfaces and that someone who is familiar (and happy with) a Control4 system is NOT the target market we set out to satisfy when we started SmartThings.

That said, here are some thoughts…

  • You can reorder SmartThings device in a room/group context.
  • You can trigger multiple devices from dashboard views if you group them appropriately
  • Groovy is a mature scripting language and you can use it to create a lot of functionality with SmartApps
  • SmartThings is much more extensible and open

SmartThings is affordable. Control4 is out of most people’s price range. Who wants to contact a dealer? That model is soooo dead. Control4 is complicated. Building a system that is at once simple to get started, and powerful enough for advances users is tough. Control4 was founded over a decade ago. We are young, agile and much faster than them. The pace of development for SmartThings and the community has already and will continue to outstrip that of Control4 and other old systems.

Lastly, we are not optimizing SmartThings for control. That should not be what home automation systems do. They should optimize for automation. If you are pulling the app out to routinely perform tasks we are both failing. Your home should respond to and your routines and actions. Pulling out your phone to turn on your lights is worse than flicking the switch, after all.

3 Likes

I don’t think anyone disputes that. The point of constructive criticism is to make the product better by exposing its weaknesses and comparing it to the best products on the market, regardless of their price.

1 Like

Ben, couldn’t agree more. But there also needs to be some user interaction. That UI / UX is the most important part of a system.

The smart home is an awesome dream. But until we can detect a specific user or users in a room, ala xbox one kinect when it says, Hi, Patrick and then it knows I want to watch the soccer game on ESPN. That is a smart home.

How can this be accomplished? Today? Something has to know its me… That’s kinect today. Nothing else knows its me. A presence sensor is a geofence, it only knows if I am here or not. Not what room I’m in.

So unless I’m missing something, I will always have to do something in a room to trigger MY event and not my wife or kids events.

Simplest way is a UI / UX layer. Maybe just a double tap on a light switch, or a keypad, or a remote.

However, I can tell you in my house, everyone pulls out the C4 app and it defaults to their room and they can control whatever is in that room. It isn’t perfect, trust me…

My goal is to figure out ways I can integrate ST into my Control4 system, and maybe one day replace most if not all of it and that isn’t too far off.

Yes, HA systems like Control4 were expensive and out of the reach of the masses. But they are the gold standard other players will be judged. It’s not fair, its not even the same things, but ultimately, that is the ballpark you are wanting to play in.

I believe 100% in what you are trying to do, and am here to help. Platforms and open standards are the key to the future of HA. Guys like Control4, creston and others will be forced eventually to change or die. ST is one of those forces.

By the way, great article in Time… http://time.com/2926400/at-your-service/?pcd=hp-magmod

1 Like

To be fair, I think it is important to define HA. Both Creston and Control4 are integrators; whole house remotes with HA built in. ST is currently not occupying that space.

Price should not be a factor, as something either works well or it doesn’t. I find the ST app to be pretty stable, and does what it was designed to do. I have the occasional missed/distorted event, but can’t attribute that to a bug in the app vs: the cloud.

While I would eventually like to see SmartApps that allow shortcuts to be defined that easily allow any number of device commands to be assigned to them (i.e. scenes, macros, etc.), and pages that these shortcuts can be placed upon to have far more flexibility from a GUI standpoint, I don’t see that being a critical mass requirement at this point.

More supported devices, and more comprehensive documentation is what I would prefer to see. ST is young, and priorities are important. I don’t need another whole house remote alternative to iRule at this point; it does fine. I’d prefer to see more device integration, interapplication communication (intents on Android), and more comprehensive docs. And BTW, HomeKit and whatever Google has planned will most likely supply us with the APIs for IAC, so I’d much prefer to see the ST team on top of that rather than reinventing the whole house remote with an integrated gui.

This just proves that ‘automation’ means different things to different people. Control is integral part of automation. There’re plenty of use cases where ‘automation’ based on timers, presence or other ‘rules’ is too cumbersome and unnecessary and simple remote control is all that’s needed. Any ‘automation’ system must perform simple control tasks well if it aims for acceptance outside of ‘geek’ market.

2 Likes

I have to say that I liked SmartThings before all the actions. It’s almost like another layer of complexity that does exactly what the SmartApps did anyway.

With that said the only reasons I end up using the app as of lately is because the automation for my garage or thermos don’t work heh. If they worked reliably I’d probably never open the app.

I’m going to propose a very controversial idea here - home automation will NEVER be 100% autonomous. It’s a great idea to have your smart home be able to anticipate your future whims. You know what else is a great idea? Communism! Or Plato Republic! Or Libertarianism! Or Free Market. All awesomely great ideas! Too bad reality keeps screwing them up. Reality is just nasty and disorganized like that.

I’m fine with Ben saying “we’re not optimizing for control”. It’s a dumb thing to say, of course :smile:, but I understand where he is coming from. It says “Automation” on the package, that’s what it must deliver first and foremost. BUT! Humans are important nodes in this physical graph as well. Most important, really. If I can’t control my house when and how the hell I please, then I’m being disempowered - surely not what ST is aiming for. Control is a very important part of automation! (And now watch me pull out the most convincing argument of all times :smile:.) I’ve spent half a decade automating huge chains of giant robots capable of destroying hundreds of thousands of dollars worth of product in one swift motion and killing a human or two in the process (semiconductor fab automation, if anybody’s interested) and even there, where there were hardly any humans around sometimes, control was ALWAYS something we strived to get right first.

Imagine a car, if you will. It has no controls, it’s 100% automatic. It drives straight and does right turns just fine. Left turns, however, are not awesome yet. It’s just not something we optimize for. Besides, UPS proved that it is by far more efficient to choose a route with right turns only. You can write your own left turns, though. :smile:

2 Likes

I like this line of thinking, but I don’t think it is what is happening with ST right now. Focusing on automation shouldn’t be the reason to make your app any less intuitive and optimized for control. We still need manual control even with automated routines, because ST cloud wouldn’t be able to anticipate everything that happens in users life, especially with family setting.

If you strive for better automation, create a web or desktop/ipad app with node based visual action-rule builder rather than just programing IDE that only handful of percentage of human population can handle. Let your mobile app a secondary app that is more focused on controls, with ability to open up the bonnet to modify automation part.

1 Like