Let's talk about Dashboards

The security pad looks good. Not sure if that’s the wall mount tablet you’re talking about.

Home Automation will only work if all of the appropriate triggers have the appropriate sensors. That’s an expensive proposition today. As people are building up a system, a macro executed from a dashboard is easy enough to implemented piece-meal. As new sensors, switches, and activators are added to the system, they can be included in the macro.

Movie time might mean that I want the blinds closed, the lights set to 10% power, the entertainment system set appropriately and the popcorn popper started. My trigger could be either starting the popcorn popper or closing the blinds, but I may not have sensors on them yet (cost). The light level and the entertainment system aren’t good triggers because I may set them to those same values for other reasons (afternoon nap or TV time), so it’s hard to use them as a trigger. But a dashboard with a button can set the levels I have and as I add more controls, my mode of interaction doesn’t have to change.

Coincidentally, this came across my Twitter today.
Biased, of course, but contributes to this discussion.

http://www.control4.com/blog/2015/04/the-benefits-of-a-dedicated-touch-screen

And those are just android based tablets running control4 native android app in dedicated mode.

I have 2 in wall c4 touchscreens. Barely ever use them, but use the app on my phone and tablets all the time.

They also make a desktop adobe air app that is identical to mobile and touchscreen.

Also their controllers have HDMI outputs that shoe the same ui controllable via sr240 or sr260 remotes. All ZigBee or IP based.

At the low end, Oomi is making the same point, touting their tablet controller as a point of differentiation. They’re still in the concept phase, but just as a marketing position it’s interesting to see how reviewers react:

Yup.

I swear I quoted this somewhere already. As long as they can keep it “affordable” for their target market, I think the strategy is excellent:

http://igg.me/at/Oomi/x/3973929

This morning’s dashboard use case:

It takes me about 5 minutes to transfer from the sofa to my wheelchair.
My health aide called and said there was a problem with traffic, she would be late.

10 minutes later, I get a notification on my watch that the front gate has opened.

I glance at the dashboard by the sofa. I can see that:

  1. the gate has closed again (sensor status)
  2. my aide is standing at the front door (camera screenshot)

I use one tap to unlock the deadbolt.
I send my service dog to physically open the door. (He has to wait for a human to undo the deadbolt.)

My aide comes in, and the dog shuts the door behind her.
The deadbolt Autolocks.

So a quick glance and one tap and I didn’t have to get off the sofa. But I had the peace of mind of knowing the gate was closed before having the door open, and I knew exactly who was there.

Note that a presence sensor, even if I wanted to give one to each aide which I don’t, doesn’t solve this use case. I want to be the one who decides when the door gets opened, not the person coming in. And I want to confirm the gate is closed first. :sunglasses:

@JDRoberts, thats a great story. What Dash are you using, SmartTiles?

Yes, I use SmartTiles. :wink:

The new Initial State + SmartThings integration gives you a dashboard as well as other visualizations to view SmartThings events. You cannot control devices through this dashboard but you can see your events/states at-a-glance. You can also view the history of your ST events over time and collect statistics (starting from when you install the integration), which is really interesting.

Full disclosure: I am a founder of Initial State so, I am obviously biased.

Read about it here: http://blog.initialstate.com/tutorial-smartthings-meet-initial-state/
http://blog.initialstate.com/wp-content/uploads/2015/07/Screenshot-2015-07-28-18.11.43.png

“Interesting” is definitely in the eye of the beholder. It’s great to have event analysis & visualization tools, but don’t overestimate the desires of average consumers. :wink:

Here are some insights after nearly a year of full time dashboard use.

  • Doors: Graphing door opening and closings over a period of months has had very little utility. I still want the data to analyze trends and do other things behind the scenes, but I never look at my contact sensors in a graph view anymore.
  • Energy/Power Consumption: This is great in a dashboard view. I can predict my electric bill every month. I know day to day how much I am spending to run various devices in my house. This has had this biggest impact on household behavior.
  • Locks: It’s good to know if the door is or is not currently locked, but this is another area where graphing has had very little utility for me.
  • Lights/On/Off/Dim: Interesting data set for comparing to energy usage patterns. Averaging can show behaviors in the household.
  • Motion Sensors: Another good data set that makes a boring graph by itself.
  • Environmental Sensors: I look at these graphs everyday. We have natural gas and our stove can pump a lot of CO2 into the house. Alerts are nice, but the graphs have given me a method to see when it’s starts to rise, and how long I can cook before I need to open windows or vent the house.
  • Batteries: This is a big one. After a while I knew exactly how long a particular brand of batteries would last in a sensor. The aeon multi sensor chews through store brand batteries.
  • Occupancy/Presence: Interesting data set. Boring graph. This is another data set that will be useful when viewed in comparison to another, but does not provide much on its own.

I look at the batteries chart nearly every day. Sometimes devices that are on the outside of my house will just die. Heat, polling frequency, and other factors can make a set of batteries go from 33% to useless over a couple of days. I am working on a dashboard that will provide more relevant information than just plotting a chart of all the previous events in a linear fashion. I would say that 95% of the data generated by all of sensors is useless at the moment, but there is 5% that is priceless.

Sifting through the noise to find that priceless information is what will make a killer dashboard.

This is pretty much exactly what I’ve found as well. Energy usage and temperatures/humidity (especially when plotted against outside temp/humidity) have been my most interesting graphs. Being able to watch battery levels as well is useful to know when to re-order what types of batteries.

Outdoor illuminance has also been somewhat interesting against the temperature/humidity and light usage graphs.

What devices are you using to get your CO2 levels?

The Netatmo weather station indoor module.

I suspected this would be the most valuable use case.

Do you think, however, that in most homes, once the energy usage pattern is understood (i.e., the most demanding appliances, electronics, and climate settings have been identified) behavior changes will generally make ongoing monitoring have limited incremental value?

In other words, I think folks buying an expensive energy monitor system like Curbed will soon have it gather dust…

I don’t think so. People are only going to be adding more connected devices in the future. Devices like curb are not just for monitoring, but they take proactive steps to adjust energy usage. I see a lot of utility in that. Seasons change, habits change, moods change. People will ask, “did I do better this month than last month” and " did this change save me any money". Those kinds of things won’t gather dust. If it provides actionable information in an easy way people will keep using it.

Also, down the road when there are enough of these devices out in the wild, the utilities may start offering incentives to adjust your usage to a certain time frame. And with solar increasing I think people are going to want to know how much they are generating every month. Imagine having one of those Tesla battery packs on the wall and few solar panels. How much did you generate? How much was stored? How much went back into the grid if they allow it?

@jrbail01 Interesting idea there, I’ve been using Graphite/Grafana to log data myself, but it’s a heavy install for the average user. I’m giving your system a try to see if it can pull of the same stuff. I’d like to see energy meter capability added though :wink:

Best Buy’s PEQ is struggling to get traction, though I don’t have any market numbers. Anyone? I guess I can research. Once again, it’s a “less open” strategy than SmartThings and relies on a $10/month subscription fee.

But their UI is much more “dashboard-like” than SmartThings App (well, it looks a lot like SmartTiles.click by @625alex, but imitation is flattery in either direction).

It is hard to conceive of SmartThings moving to this “rich” of an interface (integrated camera tiles, map view, etc.) in the short term. Long term, sure … if the demand is there. In other words, regardless if we want to permanently mount “dashboards” in our home, or just want a more customizable rich view, then we need to continue to push SmartThings for these features or ask them to open up a true, efficient, and secure “front-end user-interface API”, so that 3rd party developers can design an unlimited number of front-end experiences. The concept of a “dashboard” is bound to be popular if and only if it is highly customizable, dontcha think?

This is a bit like running a custom home screen on Android… though, frankly, I have to admit that the proportion of Android users that actually swap out their phone vendor provided native home screen with a 3rd party App is very small. Yet SmartThings could change this trend.

@bmmiller I am the author of the Initial State + SmartThings integration so I can give you some insight here. The integration originally did include energy meter support, but that was pulled (along with 15 other, mostly less popular device.capabilities) last minute before SmartApp submission because it wasn’t testable inside the SmartThings developer IDE. After working closely with SmartThings on the issue, they informed me that the deficiency is in the IDE device emulators only and not in the actual SmartThings production network.

All this to say, you can follow the feature implementation here: http://support.initialstate.com/forums/252778-feature-requests/suggestions/9182180-ability-to-monitor-aeon-energy-monitor-with-smartt

All of the SmartApp code is all written and implemented, it’s just commented out while the issue was being resolved :wink: Now that it is, I’ll be finishing the testing of a new SmartApp version and submitting the new version shortly thereafter.

I’m not familiar with PEQ offerings, but it would be curious to check them out for inspiration.

I let my daughter tap a tile on my dashboards and the lights went off. She said: “Wow! This is like magic!” Everyone has different needs and preferences, but it’s always good to have options.