I know I'm starting off with a question here that most Android fans are going to have a knee-jerk reaction to - "absolutely not, the more Android-powered smart-stuff out there the better." After all, we want to live in a world where our refrigerators know what's inside them, where our laundry lets us know on our phones when it's done, and our cars' infotainment systems aren't so god-effing-awful (even the best ones really are terrible).

Eric Schmidt said he wants your fridge to run Android at CES back in January. Nikon is making an Android-powered camera (and Polaroid already has made one). The next step would be your all-in-one TV (eg, LG's Google TVs), your game console (Ouya), your alarm clock (Archos 35), your watch (Motoactv), your car stereo (Parrot Asteroid), and even your headphones (NOX Admiral Touch). Hell, there's an Android-powered treadmill out there.



It was actually the Nikon camera that got me thinking about this, and the more I thought about it, the more I realized that a lot of these gizmos that have actually been released seemed to have run into mediocre (or flat-out negative) reviews after receiving lots of pre-launch hype. Even the positive review coverage almost always seems to contain the caveat "well, if they could fix this part/bug/feature, it'd be a lot better."

But I think that the real reason behind the critical "meh"-ness these products have come upon is because they've looked to a false "world of tomorrow," where everything is a computer with a touchscreen. A very similar thing happened back in 2004, except with Windows XP (no, seriously, it did - check out this page, and this one). That dream, to an extent, took off for Microsoft. Many point of sale systems and kiosks around the world still use Windows CE, and PDAs had their brief moment in the sun. The thing is, these were pretty strictly business and enterprise products.

But the digital pictures frames, the glasses, the conference phones, TVs, and set top boxes? I think we all know how far those ended up going.

I firmly believe that these Android-powered gadgets are doomed to the same fate, even if the buzzword effect that Android brings to a product keeps them around for a few years (which it probably will). Microsoft, unfortunately (or fortunately), had no such sexiness associated with its brand when Windows XP launched. To understand why these products are destined to die off, though, we must understand why companies make them in the first place. Hint: it's almost always for all the wrong reasons.

The Excitement Factor

Making an Android-powered product is exciting. Starting with the high-level executive who brings up the idea to the boss in a meeting because he heard about an Android-powered <INSERT PRODUCT HERE>, and because someone told him "Android is free software." And free is always a good thing.

Then, when Mr. Chairman Of The Board approves this project for exploration, that excited executive goes to his product development guys and says, "Can we do this?" The software people must absolutely cheer at the prospect of Android-ifying a product. Instead of using the company's crappy, ancient, proprietary (or licensed) software, they get to do something new and fun and sexy. The designers and engineers get equally excited at solving this new problem. And so, the product guys say everyone's "really excited" about the idea, and that they'll get started ASAP. People are working after hours, happily humming away to make this Android-powered thing a reality.

Alternatively, if the company doesn't really have software people for that product division, they hire some, and those people are super enthusiastic about working on making a "smart" version of an analog product that will change the face of the industry.

The marketing team, at this point, is absolutely chomping at the bit to get this thing out in the open for the public to see. This is a chance to reinvent the company's image - to show that Company X, Inc. is with the times, doing new and exciting things with its products. It makes their job interesting again.

And who can blame any of these people? It is exciting to do something new, something different. The problem is that no one ever actually stopped and asked if this solved a problem with the company's products, or if it was the best way to make the leap into the "smart product" world. Most often, these sorts of ideas stem from a desire to "reshape" the company's image, or to get in early on "the cutting-edge" of the market.


Hey, look - a great way to waste a lot of money and quickly regret it!

The end result is either a product that never makes it past the trade show prototype stage, or an unabashed commercial failure. There's a reason that Android treadmill is 50% off on Amazon, guys.

A False Prophet

Android is an unguarded, futuristic oasis in a harsh, scary, highly competitive low-tech wasteland. It makes big promises, the biggest one being "free software." But Android really isn't free, except in the licensing sense, especially if you're putting it on something it wasn't originally designed for.

Making an Android-powered camera, for example, isn't a simple matter of building from AOSP. It takes time - and time is money. The company has to develop (presumably) a much better camera app, make sure the thing actually works, optimize battery life, figure out how to connect up non-standard hardware settings like aperture size or optical zoom, and various other highly specialized considerations. They also have to pick a processor and chipset, and manage to fit that in the camera body along with all the camera bits. It's like making a new camera from the ground up, and a lot of that is because it's running software that the company is probably very unfamiliar with.


And inevitably, much like the first generation of Android phones and tablets, this thing's not going to be very good (the Polaroid Android camera was just terrible - though that is only one example). And it likely won't ever get any real software updates after the first few months. And it'll probably run Gingerbread or some other outdated version of the OS right out of the gate. No one wants that.

And even if the product is good, it's probably going to be expensive, and most consumers will have no idea why they would want such a thing in the first place. What was so bad about a regular camera, or regular headphones, or a regular refrigerator?

Google, of course, politely encourages this sort of tinkering. The more people hear about Android, the more companies become familiar with it, the more ubiquitous it becomes. Even if Google themselves never plan on providing these companies support, or licensing Google services (like Gmail) to them. It's sort of like the crazy-high piracy rate for Windows in many developing industrial nations - Microsoft's quietly not-too-upset about this, because it just means more people are using Windows, and that's ultimately what Microsoft wants: more users.

So, when Eric Schmidt makes an Android-powered fridge sound like a good thing, he's really just waving that Android oasis in the face of various companies who make products that aren't traditionally considered "high-tech." For most of them, the pursuit of that lush digital paradise will just turn out to be a clever salesman's mirage.

"Smart" Does Not Mean A Touchscreen And Apps

"Apps" has become such a product buzzword in the tech industry that I seldom see a PR email without it anymore. Apps, apps, apps. And it's the companies that want to make Android-powered gadgetry who take the popularity of the "app" in entirely the wrong way. No, I don't want my treadmill, my headphones, my camera, or my refrigerator to run apps. That is one of the most ludicrously stupid, technology-for-the-sake-of-technology trends to happen in the last quarter-century.

What I do want is for these things to have apps. Note the subtle difference. I don't want my refrigerator to have a touchscreen and a dual-core processor so I can look at Facebook, I want to have an app on my phone or tablet that lets me set the temperature and do other nifty things with it (like set up a digital pantry) from across the house, or 20 miles away at the grocery store. Implementing that requires Wi-Fi or ethernet, some relatively basic firmware logic, and an app available for major mobile operating systems. Not only will your smart fridge end up being cheaper (probably by a lot) to design, it'll be cheaper for consumers to buy. And they'll like it more.

크기변환_smart_appliance_02 크기변환_smart_appliance_03

Left: Yes!, Right: NO. (they're actually the same product, but you get the picture)

And keeping it up to date will be easier. You can add more features in the app, update the lightweight firmware over the internet, and do it all without having to maintain an entire operating system. This alone guarantees the useful life of your smart product will be longer. You do not need to reinvent the wheel to get in on the "smart revolution," and it's best not to.


The real takeaway here is that I don't want the "product of the future" to be some expensive, complex, quickly outdated computer-meets-toaster running Gingerbread. I want to be able to use the extremely capable, versatile, and convenient device in my pocket or on my desk as a very cool remote control and interactive tool in tandem with it. That's it. If you must put a screen on it, please, don't force an entire mobile operating system on there, too.

There are already companies out there doing this - the Nest thermostat is a fantastic example. Parrot's Zik headphones, if they work as promised, are a mind-bendingly cool take on smart personal audio. I can't say I really get the usefulness of the "smart watch" concept, but Pebble is a much better idea than Sony's mega-failure SmartWatch or the Motoactv (which really isn't technically a smart watch, I guess).

I guess what I would say in order to appeal to the business logic of the would-be manufacturers of such Android-powered stuff is: do you really want the responsibility of maintaining an operating system on your product post-sale, along with the product itself? Or, better yet, can you really afford to?

David Ruddock
David's phone is whatever is currently sitting on his desk. He is an avid writer, and enjoys playing devil's advocate in editorials, and reviewing the latest phones and gadgets. He also doesn't usually write such boring sentences.

  • http://turbofool.com Jarrett Lennon Kaufman

    I think the problem lies not in using Android to run the device, but in expecting the Android UI to remain intact. It seems like a lot of these hardware makers are relying on the familiarity of the Android interface, not to mention the lack of effort involved in designing their own, without realizing that that interface doesn't apply to anything.

    What Android could and should be used for, though, is as the backbone of the device. Much like Java and Linux have been in a lot of other devices, and much like Windows CE did VERY successfully in a massive range of devices people weren't aware ran it. Heck, the Dreamcast was running on Windows CE. In all the successful, well-done cases, it represented only the kernel and backbone for a UI designed for the product in question. Android could work very well as the same.

    • http://twitter.com/misterE33 Mr E

      Agreed. Then again, how many people would buy the Angry Birds Refrigerator?

    • http://www.androidpolice.com/ David Ruddock

      That's a point, I suppose, but my argument is more that manufacturers are going about the ridiculous process of putting an entire operating system on something that doesn't need one in the first place. It's cheaper and simpler for the manufacturer to embrace mobile OS's not as the backbones of their products, but as existing tools most consumers have on other devices, and to then make their products integrate intelligently with those other devices.

      Interoperability, as opposed to trying to "revolutionize" an entire product by jamming a touchscreen and a mobile chipset in it.

      • http://turbofool.com Jarrett Lennon Kaufman

        In many cases, I agree. But in many others, such as a refrigerator, I think the self-contained power of a full OS can be very useful. I'm with you that it's being overused and thrown in as a gimmick in itself in a large range of uses, but there's a surprising number of places where the background elements (like the intents system that Eric notes) of the OS could expand the functionality of a device in useful ways. Letting a device manage itself, and then report to your other Android device using standard communication methods IS pretty handy.

        But merely slapping a touchscreen on it and letting you play Angry Birds is NOT the way to do this. It's not actually helpful or revolutionary. It's similar to the effort to throw TVs in refrigerators a while back. It also reminds me of this constant desire on the part of auto makers to make your CAR the digital hub of your life, checking traffic for you, informing meeting attendees that you're going to be late, adjusting your alarm clock if a meeting gets canceled. These are all things your PHONE or COMPUTER can do for you better than your car. Making your other devices cover all these needs is similarly silly.

        But the OS itself is still potentially valuable for this. The manufacturers just need to divorce the image in their heads of presenting Android the way people are used to it, and instead merely harness its underlying power. From the beginning it was claimed that Android would work on a huge range of devices, including keyboard or no keyboard, touchscreen or not, or even NO screen. Let's remember that much like it's Linux underpinnings, it can be ANYTHING. But it shouldn't be an excuse to skip out on designing an interface and innovating. Android for the sake of Android is pointless. Android for the sake of the power it provides to useful new ideas is kickass.

        • http://www.androidpolice.com/ David Ruddock

          I don't disagree that it could (even would) be useful if done right. The problem is absolutely no manufacturer wants to take the time (read: spend the money) to turn Android into something that would be good for a refrigerator. They'd have to jack up the price to absurd levels.

          In the end, it's better (cheaper) for both consumers and manufacturers to let these sorts of products be dumb terminals to the highly advanced, really awesome devices we use on a daily basis. We may not get all the functionality a full-on OS-running refrigerator could provide, but you could definitely get a lot of it.

    • http://www.androidpolice.com/author/eric-ravenscraft/ Eric Ravenscraft

      This man, here. He gets it.

      It's true that a lot of manufacturers do it wrong. Particularly in terms of trying to attach Android branding and UI to an Android-powered product. Having the underlying architecture is great for device and appliance manufacturers. For example, the Intents system is an easy way for an app on a phone to trigger an action on a refrigerator. Great! However, Facebook and Twitter are not needed on a refrigerator.

      I actually think, from the sounds of it, Ouya is approaching this the right way. They're building a game console that's specifically designed for a TV and partnering with major organizations and developers to make sure things run well on it in that set up. They're not, however, just running "apps" on it. Google tried that with Google TV. It was awful. If Ouya can provide a more focused, refined approach while still relying on all the underlying infrastructure Android provides, then it will actually be worthwhile. Not to mention it would be more than just an Android phone plugged into a TV with a Bluetooth controller. Maybe I'm wrong, of course, and they're not going to do the work to make it a tight fit, but that's what it sounds like/what they should be doing.

      • http://www.androidpolice.com/ David Ruddock

        Yeah, Ouya's sort of a special case. A console needs an OS, I suppose.

        • raindog469

          Digital cameras have had operating systems since the first Kodak model I bought in 1999. We couldn't add anything to them, and those early ones might have just had some single embedded program that controlled the whole thing, but they had CPUs and RAM and ROM and disk I/O... of course they needed an OS. And every digital camera since then has had an OS, and different models had different apps installed.

          The fact that their UIs didn't resemble a general purpose PC or phone OS is irrelevant. The Nook Color, Kindle Fire and Pandigital Novel don't resemble Android either, but they're Android devices. Same with the Ouya, at least what they've shown us to date.

          I have an HP printer with an app store. You navigate it with the little 1.5" screen and eight buttons around it. It is terrible. I didn't buy the printer because I thought I would download apps. I haven't, and I probably never will. But it exists. Maybe some people want their printer to wake up at 5am and print headlines and news for them. It'd be a hell of a lot easier if it were running Android than whatever home-baked bastardized Linux fork HP came up with (no, it's not WebOS, either). My TV is the same. The kernel's been released and everything. It is a HORRIBLE interface. I'd love to flash Android on it and run Youtube, XBMC, have a Bluetooth keyboard, etc...

          I never thought I'd see the day Linux would make a good embedded OS, but that happened 10 years ago, even before Microsoft's vain experiments with embedded XP. Android's next. It may not look like Android and after this initial wave, you may not even be able to tell it's Android unless you break out of the camera/printer/fridge maker's jail. But it's happening, and won't stop.

    • Knlegend1

      I agree that's why I commend Amazon. Most people don't even know its Android let alone gingerbread. I think more companies should take that approach.

    • http://pandu.poluan.info pepoluan

      This, this, and this. I do question the suitability of Android for non-interrupt-based devices such as cameras, though... they might be better off with Embedded Linux rather than Android.

  • marcusmaximus04

    I'll agree with your vision of things like fridges et al. There's no reason at all for my fridge to be able to run Angry Birds.

    Things like cameras, on the other hand... I'd rather use an Android interface(yes, even an old one like Froyo) on (insert camera here) than whatever, generally horrible, interface the company that makes it implements. The points about having to write a camera app and such don't really make much sense. They generally have to do that anyway, just on top of a terrible-beyond-reason OS of their own making.

    Also, is that Polaroid android phone even out yet? I looked around and it doesn't seem to be, but most initial reactions seemed to be positive.

    • http://www.androidpolice.com/ David Ruddock

      It isn't. I honestly think it may have been cancelled. I used it, it was bad. Really bad (horribly laggy).

      • marcusmaximus04

        It seems somewhat unfair to judge a pre-release, lacking even a release date or price, device on how smooth the interface is, but point taken.

        That said, I can see Android being a major boon for future cameras. Throw in a halfway decent processor(say, a lower-end Tegra 3) and you can get a really nice interface that's perfectly smooth and implement some photo/video-editing capabilities, mobile printing(partner with, say, printershare) and such. While, sure, mobile phones can do a lot of that already, being able to have that with the higher image quality you can get when not going for the extreme portability and multi-function nature of a phone and things like optical zoom would be great.

        • http://www.androidpolice.com/ David Ruddock

          Why should cameras try to do what the phone in my pocket already does really well? Why not use Wi-Fi to simply automatically push (or remotely view) images from my camera to my phone directly, where I have an amazing, high-res display to review them, a mobile data connection, and tons of apps to send them off to where I want them to go?

          Again, from a simple cost of R&D and manufacturing standpoint, you're creating redundant, more expensive technology. We want a more integrated technological world - not a more expensive, complicated one. It's overengineering, that's what really kills me.

          • marcusmaximus04

            "Why not use Wi-Fi to simply automatically push (or remotely view) images from my camera to my phone directly, where I have an amazing, high-res display to review them, a mobile data connection, and tons of apps to send them off to where I want them to go?"

            For me, at least, it'd be a lot easier to be able to snap a picture and look at it on the camera, with a nice interface to do so, than take a picture, put down the camera, take my phone out of my pocket, look at it and do whatever I need to do with it, put my phone back in my pocket, pick up the camera again and then take another picture.

            It's mostly just about ease of use, and Android provides one of the easiest-to-use UI's for mobile devices(which I would count point-and-shoot cameras as part of). I don't want to have to choose between an excellent UI on my phone and a terrible one on a camera, or to have to switch to my phone any time I want the decent UI, in the case of a camera pushing photos to it. I just don't see any reason why these need to be mutually exclusive.

          • http://www.androidpolice.com/ David Ruddock

            Find me someone who really wants to pay extra for that, and find me a manufacturer that is going to do it right. I don't want a mid-range point and shoot camera with a high-end point and shoot price just because it has a mobile processor and a big touchscreen and apps - I have a phone for that. I want a camera that takes really good photos.

            And if you told a bunch of photographers Canon was ditching the mode wheel and switching to an Android app for taking and reviewing photos on all DSLRs and raising the price of each model by $100 as a result, you know what they'd do? They'd all go buy Nikons.

            While I'm willing to debate the utility of things like an OS-running fridge, Android cameras are a DOA idea. Phone cameras are moving ahead too quickly for them to have a snowball's chance in hell at catching on.

          • marcusmaximus04

            We have yet to see how much running Android would add to the price. I agree with you that I would not be willing to pay substantially more for a camera of similar quality but runs android.

            That said, the manufacturing cost of the Tegra 3 in the Nexus 7 is $21 and the memory was $13.50. Add in a slice for extra engineering(I'm unconvinced this is needed, since, as I said, these cameras already have OS's that need to be coded for, and are likely much more difficult to do so), by one of the big players and I'd certainly be willing to pay $40-$50 more for a camera sporting the Android interface.

          • http://www.androidpolice.com/ David Ruddock

            The display is the most expensive part of every mobile device, and that's where the cost will come from. That, and the trickle-down of R&D so that the MFU can break even. I wouldn't put my eggs in the comparable price-performance basket.

          • marcusmaximus04

            Well, the cameras I'd suggest this for are the high-endish point and shoots that, by and large, already have relatively large, nice displays(and in many cases, touchscreen ones), so there shouldn't be much, if any, added cost there. I'm aware there's engineering to do w/ respect to figuring out the Android OS, but that really shouldn't add much to the cost, given that they already have to develop software for existing cameras and my estimate for a price increase(vs. the same hypothetical camera without the Tegra 3 and Android OS) is, after all, about 1/2 the total cost of the Ouya. So I feel like that's reasonable.

            That said, I'm pretty sure we'll find out in the near future.

  • Knlegend1

    Damn comment section is better than the article lol.

  • http://profiles.google.com/andrewc513 Andrew Chandler

    It's necessary trial and error that leads up to what will be normal in the future. Half of our modern conveniences and technologies started as half-baked ideas for their time.

  • Elias

    "What was so bad about a regular camera, or regular headphones, or a regular refrigerator?"
    I'll tell you what, they don't communicate with each other. They're all dumb.
    I still have to pop the SD out of my camera to transfer photos, while I'd like it automatically done to the cloud and synced to my PC. And I don't want to use a bazillion proprietary services which have different requirements, fees, constraints, terms of service, caveats, availability issues... I want ONE single center for all my stuff. I want all my damn photos to go straight to Google Drive or Google+, just like they already do on my smartphone. People want their cameras to run Instagram, or want much better cameras on their phones (like nokia pureview has shown). Do you really think it's cheaper to develop your own OS and then convince Instagram to release an app for it? Do you think it's cheaper to mantain your own cloud service? Do you think it's cheaper to make your own OS communicate intelligently with every damn device out there, instead of using a universal OS to do the task?
    Making things communicate is the next big step. Integration. That's what everyone complains about. If you get a mac, you'll get an iphone and an ipad because you know they'll comunicate nicely. If you get a windows pc, you may get a windows phone or a windows tablet. That's all about integration and making things work together. If you get every damn device running the same OS, it will have its own integration features built-in, and you can also use these features to communicate with your app. Much simpler than mantaining everything on your own.
    This is simpler than you made it sound. First, you get some hardware able to run Android (aosp-compatible would be a huge plus here). Then, put Android on it. Third, develop your own app for android and keep it up to date. That's it. You can use the same hardware platform a few times more, so engineering costs are mitigated. If you use aosp-compatible hardware, you wouldn't even need major modifications to get the OS running, plus mantaining the OS up-to-date would be much simpler. And you don't really have to keep the Android version up-to-date, as long as your app continues to run on that version of Android. After all, if things don't get updated at all, you get basically the same functionality of a regular camera which you'll never update anyway, but with the added feature of maybe being able to install a few apps, sync automatically and wirelessly, etc.

  • Eric Jones

    I would love to have light switches in my house integrated with android (bluetooth, wifi, whatever is smartest), but having each light switch be an Android phone with a touch screen would be prohibitively expensive, and really stupid. There might be other devices where a decent amount of Android functionality might need to be included, but not many. I'm glad to read a post like this, where the author gets it.

    • Jei Arc

      well this already exist and has been implemented, is a simply control system for your house presented as a GUI to you be it in your computer (desktop, laptop, ect...) and connected via your home network, you can control many aspect of your home electrical system. From here is just a matter of making this more compatible and more accessible.

      Android could provide such versatility but implementation is a the big differentiator. I wouldn't want to be installing apps in my home control system and wouldn't want the system to freely access wireless networks for no reason, this would be a huge security issue.

  • m8000


  • andrew__des_moines

    I think Schmidt had Arduino in mind (or like devices) when he spoke of appliances. As others have mentioned, it is the underlying OS, communications, and now strong chip industry that opens up opportunities in appliances, not a full Android UI.

  • ilike androidfridge

    i really would love a fridge or a watch or toilet seat where i could play angry bird or shadowgun

  • Rob

    Fundementally I agree with this article. Consumer's don't care what underpins their technology. They just want it to work and market forces will then dictate what suceeds. But convergence of technology is inevitable. However there's always been a temptation to adapt a product for something it's not designed for. Personnally I don't want to surf the web on my tv, but being able to stream movies direct or wirelessly play footage from my camcorder is/would be great.

    The technology for a smarter lifestyle is out there. You can already control your lights, electrical devices, heating remotely. Going to be late home, then delay the heating from your phone. You can even turn on the Nissan Leaf's heating and a/c remotely.

  • Jei Arc

    Don't forget the Audi line of cars for 2013 is including google Nav and maps in their infotainment system which also includes a local hotspot. I was trying to see if the system is actually android or not but i guess i will check that out when i buy the Q7 ;)

  • RedPandaAlex

    I think we agree that we want the smartphones we already have to be the hub of our digital life. There are cool things manufacturers can do to digitize their products, like allowing us to automate or prorgam certain functions and feeding us information from sensors in the hardware. It's up the manufacturers to decide whether they need their devices to run an entire OS in order to do that or not.

    The trend that's stupid is trying to use touch screens and apps on the smart devices themselves as a new hub for our attention--because I don't need a car with a processor and OS I'll never be able to upgrade and a 3g data connection to run Pandora. I just need a dock that connects my phone to bluetooth and maybe opens an app with a dashboard of any info the car wants to display. I think when a device is ADVERTISED as running Android, they're trying to sell the idea of a touchscreen running apps, not the idea of integration.

  • http://www.facebook.com/people/Jack-N-Fran-Farrell/100002337622505 Jack N Fran Farrell

    Should your fridge run on Android. Probably yes if you can can sell your solar electric to PG&E for top dollar from 10am to 6pm you should slghtly overcool when rates are low and sell when rates are high.

  • grellanl

    Smart devices that include more and more sensors and connectivity, I think everybody is on board. But nobody wants these devices to act like a tablet - I don't want a touch screen on my fridge, and although I might want a touchscreen on my camera, I don't want it to have the Play Store, for example. These connected devices should have simple intelligence built in, that connects them to cloud services or allows us to use multipurpose computing devices to manage them. They are fixed-function devices, and should be simple and low-maintenance.

    But as others have pointed out, Android is free and could make a decent embedded OS, saving manufacturers a whole heap of effort in creating an OS for their products that runs on commodity hardware -- and also offers all the tools you need for sophisticated UI, touch interfaces, Internet connectivity, cloud sync etc. It just won't look like Android. The Nexus Q is an example of this approach.

  • gadgety

    In order to save battery, the Pebble has a small black and white screen only, and at a poor 139 ppi. That's not good enough for me. I'd rather charge my watch nightly, I'm charging my phone anyway. The benefit of Android is the huge development community.