Smartphone controlled <i>anything</i> never feels great to me. If I have to type in code to unlock my keypad, then wait for an app to launch, <i>just</i> to change the temperature on the thermostat, it's already over.<p>On a similar note, I wanted to love the Chromecast, but I found the phone to be a terrible controller. Now I use a Roku, which has a great remote. Instead of my girlfriend and I crowding ourselves around a tiny smartphone screen to browse what we're going to watch, we bring up the Netflix app on the Roku and browse on the much more suitable TV.
Instead of a $99 custom button, what about a sub-$99 Android tablet? You can put them anywhere you'd put the button, and leave your home control app running with the screen set to never turn off. Turning on/off a light with a touch will then be no more effort than flicking a light switch -- there's nothing to take out of your pocket, no waiting for an app to open.<p>This is how my home works. I built the app that runs on the screens all the time (a little node.js server and web app), and have a few tablets mounted on the walls. This is a screenshot from October:<p><a href="http://www.dangrossman.info/wp-content/uploads/home2.png" rel="nofollow">http://www.dangrossman.info/wp-content/uploads/home2.png</a><p>Since then, I've added a bunch of new controls for dimmable lights and scenes (e.g. turn off all the first floor lights at once, or dim them all to the same percentage).<p>I built my own, but there are dozens of free pre-built home control apps in the Play Store. I find Wink's very attractive for example: <a href="http://i.imgur.com/DUWIwh3.png" rel="nofollow">http://i.imgur.com/DUWIwh3.png</a>
I have my previous-generation phone that I don't use anymore, plugged in, connected to wifi, on a shelf. It's an integrated combination of screens, speech recognition, sensors, dedicated devices and smart surfaces, <i>and</i> solves the multi-user issue because it's out in the open on a shelf, and it's not my personal device anymore so I don't have sensitive information on it. The apps are right on the home screen, no passcode, just a direct-to-the-app experience.<p>You can find a used nexus one, or a moto e, for example, on Swappa for less than $100.
"We believe that the future of smart homes will not be found in a centralized device but in an integrated combination of screens, speech recognition, sensors, dedicated devices and smart surfaces."<p>I think the latter part of the sentence unintentionally describes smartphones. I would say that today's smartphones are integrated combination of screens, speech recognition, sensors, dedicated devices and smart surfaces. Pretty damn good as well.<p>I take issue with the first part of the sentence as well. I think we do want a centralized device with centralized software. People are sick of fridge operating one way and stoves working another. Take changing clocks on microwave, coffee machine, fridge, stoves etc for example. Don't you just hate doing it one by one? Not to mention the fact that they all have different interfaces and work differently.
What I am missing with all the smart home thing and only the Nest has done right is, that the smartphone is a personal thing. But switch on a light or similar is not a personal thing, if you are living with a family with kids. In particular if you have smaller kids, which don't have a small device and must be able to switch on the light or similar.<p>There are so much levels in terms of a smart home, that a smartphone will never be able to handle, because the later one is only a personal thing.
I noticed a lot of these comments ended up referencing challenges with lighting so I figured it would be good to chime in here (caveat: co-founder @ emberlight.co).<p>Interaction models are something we've been thinking about quite a bit and we've been exploring a number of options. New products like Senic that explore new opportunities are especially exciting for us. Here are 3 rough tenants we've been following when thinking about how people will interact with our product:<p><pre><code> 1) The smartphone is not the solution, it's a stop on the way to something much better. This is basically the point of OP's article and we completely agree.
2) If it's not broken, don't fix it. In other words, if there's an existing interaction model in the home that everyone is already used to (i.e. light switch, tv remote, wall thermostat) seriously ask yourself if that really (really) needs to change before inventing something new.
3) In slight opposition to #2 there are obviously huge opportunities to innovate the way we interact with our home. Some of the models that I'm quite excited about:
-The "dashboard" model that Sentri is building (http://sentri.me)
-The "voice" model: (i.e. Echo, API.ai, etc.)
-The "gesture" model which is what Senic and a few others are building
</code></pre>
We're actively researching many of these and building some very cool partnerships in this space for exactly this reason. I think the end result will end up being some combination of these models and perhaps others that I've missed.<p>Btw, feel free to reach out directly for thoughts/questions/coffee: kevinr@emberlight.co
The connected lightbulb example is great. That's IoT as a novelty; it adds few/trivial new capabilities, and it adds a lot of inconvenience.<p>The flipside of this is IoT as a superpower: finding applications where you can truly use technology to give people a new ability that they didn't have before.<p>From the personal side, this IoT novelty versus superpower discrepancy is something we thought a lot about as we brought Pantelligent to the world (note: co-founder, <a href="https://www.pantelligent.com/" rel="nofollow">https://www.pantelligent.com/</a> ). Our superpower is to let anyone cook great food, perfectly every time, through science. For us, the smartphone was the perfect user interface, because people are already in the kitchen with smartphone in one hand and spatula in the other! But our integration with the Pebble smartwatch hints at an even better future fit; the Pebble is great for cooking because it's waterproof, and it lets us bring real-time cooking data and instructions right to your wrist, even if your hands are busy.
Well, sure, bad interfaces to 'smart home' devices will be unpleasant to use. But most things like Wemo, Hue, etc. still let you hit their wall switch. Their 'smartness' is additive, if you have a screen to control them with.<p>Clearly the reason many connected devices expect you to bring your own screen is cost - Nest notwithstanding, adding touchscreens to everything when you've already got one in your pocket seems unnecessary, and would clearly add to price of these products.
I think its perfectly expect-able that an actual smart home be accessible through a variety of modalities and methods. There's no reason why there shouldn't be smartphone access. But there's also no reason why there shouldn't also be the ability to use voice commands (not through your smartphone), direct wall controls, 'dumb' (relatively) physical remotes, etc etc.<p>For example, I don't always walk around my house with my phone. Sometimes I wear pants/shorts without pockets. Or I have to leave my phone plugged in somewhere cause it's charging. Or maybe cause I'm taking a shower and realized that I may have left the oven on and my roommate is out and I left my phone in my bedroom, and it would be awesome to be able to yell a question to my house and just have to dealt with. Or I have friend over and I want them to be able to play around with some of the coolness, but I don't want to fuck around with getting them to install the app, and syncing the authentication code, etc etc.<p>I mean, we can 'argue' about cost. But seriously, you're wiring up your house for automation and remote access. Enabling additional access nodes is a relatively tiny opportunity cost.
I have a really great UX for controlling lighting in my home. So simple my 2 year old understood it when she was just 12 months.<p>Maybe it's top secret, worthy of some serious VC. But you can probably guess what it is :-)<p>Now smart phones - she gets those too but she finds it so annoying. Adverts on youtube serve to cause confusion. That action bar on the Android is annoying. She says 'waaah' when the screen is locked. Etc.
I agree with the author's thesis, that smartphones, as generic information processing tools, offer poor user interfaces for controlling things in our physical environment.<p><i>Now, the cost and effort to manufacture hardware is dropping and we have the ability to create interfaces that are not designed as a generic device but designed for a specific person or situation.</i><p>This feels very true. The music industry has been taking advantage of this for a while, with all kinds of interesting and innovative tactile controllers interacting with music making software. In that domain companies like Native Instruments are addressing exactly the concerns raised by this article. (Interestingly, the author cites playing music as an example of how an appropriately-crafted physical device (i.e. an instrument) lets the operator focus entirely on the task by minimizing cognitive load.)
You didn't mention the smartwatch at all in your essay. As someone living with one (a pebble) daily for the past year, many of the smarthome needs have been extending into working with the smartwatch, and they don't have the painpoints you describe.<p>Just as an example - for changing volume on my pebble, i do this:<p>1) long-hold the up-button (i have this binded to an app called 'music boss' )<p>2) hit up or down.<p>3) volume changes.<p>There's similar apps for controlling smart lightbulbs, etc, and pretty much expect this to be standard case on android wear / apple watch / microsoft band / etc. in the coming months.<p>Is there a "multi-user" factor missing that makes a case for products like Senic Flow? Yes, but that wasn't addressed in the essay.
I completely agree with the author here, far too much weight has been placed on the smartphone as the control device and UI for everything.<p>This problem has even spread to Sonos which discontinued the dedicated controller a year or so ago. You now have to use either a phone, tablet, or a full computer to select and choose the music you want to play. Granted their apps are well done but the overall experience is nowhere as responsive as the dedicated controller.<p>I really do not understand their decision, and would think it would be possible to offer a dedicated controller device these days at a reasonable price that they can make profitable.
I found myself in agreement with the post but then when I saw the flow it looked far too general purpose. And as a computer input device it looks like a nonstarter (sorta like leap). I was envisioning some very specific tasks.
This article was a great summarization of the problems we face in our startup.<p>We actually haven't yet built a smartphone app, even though our customers have asked for it, simply because we don't know a good way to solve these problems.<p>Do you take the money and build the app anyway, or come back to them with a new and very different solution to what they asked for?<p>Take this problem, and apply it to an entire building of office tenants, and you can see why this is valuable to them. But it's one thing for us to create a product and make the sale. It's more important to me that they actually <i>use</i> what we sell them.
I will say that I think Leap motion[0] is another really well made "controller" that would be great for smart homes. It just "works" with their code out of the box. You can easily tell how many fingers someone has, whether they are tapping, grabbing, a couple other gestures, their 3 dim location relative to the camera, etc. I used one at a hackathon a few months back to attempt a Theremin and it was quite fun.<p>[0]<a href="https://www.leapmotion.com/" rel="nofollow">https://www.leapmotion.com/</a>
It's 2014 (almost 2015). It's time my house knows to turn the hallway light on FOR me because it's night time and I'm walking in that direction.
Well, I don't think anyone ever said smartphone cameras were better than point and shoot from any UX aspect, but that hasn't stopped them from winning.
I feel like I must be missing something about the concept of a "smart home". Why can't I just leave the light switch (or any other controller) how it is and just have an additional Internet-based fancy control for automation or use when I am away from home? I don't see the need for any new control interfaces for things I can already control perfectly well.
><i>Smart homes are not a thing of the future anymore. Right now, 100 “things” per second are connecting to the internet. By 2020, more than 250 things will connect each second.</i><p>Probably unpopular in a programming forum, but, aside for people with physical difficulties, did anyone (e.g. more than a few outliers, not literally anyone as in > 0) ever asked for a "smart home"?<p>And before someone replies with the quote about "faster horsers", did anyone, AFTER shown one, went anything but "meh"?<p>Anybody that show not just as "nice to have, ok, move on", on the level of battery powered toothbrushes, but as something that really impacts your life.<p>Seems more to me, like a few other things in tech, like a solution in search of a problem. In the say way nobody asked or wanted those "if you want to talk about your credit card, press 3", etc, automated speech recognition call services.
Smartphones are not a good way to control things in the real world because there's no tactile feedback. So the only feedback is on the screen, which you have to look at, instead of the real world thing you're controlling.
NFC would be a great intermediate solution to address this problem. Here's hoping that at the next WWDC, Apple will open up their NXP chip to everyone.