back to article Nest's bricking of Revolv serves as wake-up call to industry

The extraordinary decision of Nest to brick its $300 Revolv home automation hub has served as a wake-up call to the tech industry. Both customers and the broader internet of things (IoT) industry were appalled when Nest removed all support for the device, making it as useful as a tub of hummus, as one angry consumer memorably …

  1. Doctor Syntax Silver badge

    What sort of wake-up call?

    It seems all too likely that if Nest/Revolv get away with this one others will try it too. Once the money's been taken all those customers are just so many nuisances.

    Alternatively you might wonder if Revolv's intended product was the users. If that was the case then it might indicate that the market for that product wasn't sufficient to make the business viable. That might suggest that other businesses run on those lines could be in trouble too.

    1. tony72

      Re: What sort of wake-up call?

      Yes, I don't see this as any different than some of those old online music services where people lost access to their "purchased" music when the service shut down. It shouldn't be news to anybody that any device that depends on a proprietary cloud service could become a piece of junk in an instant.

      1. Mage Silver badge

        Re: What sort of wake-up call?

        Yes, Open Source won't solve this. HW or SW.

        1. Gene Cash Silver badge

          Re: What sort of wake-up call?

          Actually, my 3M thermostat with open specs allowed me to write an Android app that communicates directly with the device.

        2. Richard Plinston

          Re: What sort of wake-up call?

          > Open Source won't solve this.

          It would if Nest released the source code as open source and it is modified by the community to access different cloud services and can use other protocols.

          1. Anonymous Coward
            Anonymous Coward

            @Richard Plinston - Re: What sort of wake-up call?

            Sorry to bring you down to Earth, but in case you missed it I will remind you and the person who up-voted you that there can be no open source software on closed, proprietary hardware. Use of the hardware by vendors to lock you onto their services is the whole point of IoT, mobile technologies and now being brought to PCs by Windows 10.

            I hate to point this to you but a free user can't be monetized properly.

            1. werdsmith Silver badge

              Re: @Richard Plinston - What sort of wake-up call?

              I hate to point this to you but a free user can't be monetized properly.

              I think what people will be looking for is something like OpenWrt for IoT.

            2. Doctor Syntax Silver badge

              Re: @Richard Plinston - What sort of wake-up call?

              "I hate to point this to you but a free user can't be monetized properly."

              And I love to point out to you something I mentioned at the start of this thread: if monetisation of the users was the point of Revolv it must have failed, otherwise why would they close the server? And if that was indeed the case it might be ominous news for other "services" based on the same premise.

              1. Anonymous Coward
                Anonymous Coward

                @Doctor Syntax - Re: @Richard Plinston - What sort of wake-up call?

                So were they a charity then or they were chums with everybody ?

                It really is ominous news but not for other services, it's ominous for consumers.

            3. Richard Plinston

              Re: @Richard Plinston - What sort of wake-up call?

              > there can be no open source software on closed, proprietary hardware.

              And yet I seem to be running CentOS and Ubuntu on Intel and AMD based systems (and on ARM) which I don't have the masks and circuits for.

            4. Richard Plinston

              Re: @Richard Plinston - What sort of wake-up call?

              > a free user can't be monetized properly.

              a free user, properly, can't be monetized.

              ftfy

            5. Richard Plinston

              Re: @Richard Plinston - What sort of wake-up call?

              > Use of the hardware by vendors to lock you onto their services is ... now being brought to PCs by Windows 10.

              Which is why I have avoided Microsoft products for the last couple of decades.

    2. big_D Silver badge

      Re: What sort of wake-up call?

      The problem isn't whether the device still works or the onboard software is still secure, the problem is relying on cloud services.

      Even if the cloud server software is open source, if the company goes tits-up or just turns off the server, because you are the only person left using the device, you still need to find a replacement for that cloud, which means providing your own cloud based server.

      For the average reader of this site, probably not that big a problem - assuming the IoT device lets a new server address be entered - something usually blocked for security reasons - but the average home user won't have a clue where to start.

      I still haven't seen an IoT device that makes me say, that is more useful than a dumb device.

      IoT toaster that I can turn on remotely? Yeah, but who is going to put the bread in it?

      IoT fridge that tells me what is running out? I tend to buy fresh produce and what is in season or rotate products, so that I don't eat the same thing every week. So a list and looking at the shelves is better.

      IoT lights that I can turn on remotely? Why? If I'm not at home, I don't need light or I'll turn them off, when my family is sitting around the table... If I'm away from home, then I'll put the lamps on a timer - or lower the roller blinds...

    3. Anonymous Coward
      Anonymous Coward

      Re: OPEN SAUCE IosTuff

      as an IoT beginner, forget the $$$ nest stuff - start at http://www.mysensors.org/build/

      decide if you want to use 5volt or 3volt bits (hint: the mesh network radio are 3Volts)

      you need a gateway (Arduino Uno) to send serial data to a PC (or RPi2+/3)

      and remote nodes that sense things

      spend £40 on eBay

      wait a month for all the little bits to arrive

      cut & paste code from MySensors into the free development environment https://www.arduino.cc/en/Main/Software (Mac OSX or Windows IDE suggested)

      Select a port & board in IDE/Tools, Use Tools/Serial Monitor to debug the sensor node data arriving at the Gateway

      you are limited to a few hundred nodes, if you go for the very high power Wi-Fi nodes then you need to start screening things & buying/building voltage regulators, otherwise not.

      If you want reliability & quality then buy original Arduinos from Italy. Otherwise you can get ten nano-arduinos from China for about the same price for development, need a simple programming adapter 'USB to TTL' for these tiny motes.

      it does just work. Google is hardly involved. you don't need to Cloud your data unless you wish to.

      in the IDE with your MySensors library installed, the MyConfig.h text file defines if you need crypto or not; 256 bit hashed/signed data authentication. Change #define RF24_BASE_RADIO_ID ((uint64_t)0xA8A8E1FC00LL) 0xA8.. bit to something else if you want a unique network name.

      some good people have been working for several years, it now literally takes an afternoon to start this IoT stuff, once the China Post delivers. . .

    4. BillG
      Facepalm

      Re: What sort of wake-up call?

      Here's the thing - while many potential Nest customers may not notice Nest's irresponsible shutting down of Revolv by bricking products people paid good money for, you can bet that huge commercial and industrial customers will notice. They will especially notice when a Nest salesperson comes knocking at their door. This is going to crush Nest's future.

  2. Anonymous Coward
    Anonymous Coward

    Cue the teardowns

    It's not useless, it's a SBC that probably has JTAG pads somewhere in it. Speaking of JTAG, I thought FTDI already sent that wake-up call and everyone already knew how rude it was.

  3. Richard Plinston

    Both customers

    "Both customers ..."

    Well, if there are only 2 then I can see why they closed down the business.

  4. allthecoolshortnamesweretaken

    I still feel there is a solution, probably involving triangles.

    1. VinceH

      It probably depends how many Revolv hubs are out there, and whether the number can be used to make an equilateral triangle.

  5. hellwig

    $300?

    Was this thing also supposed to act as a video game console? Could it stream videos to my TV? I still had to buy the products that connected to it, right? Where did $300 of value come from?

    I bought some light bulbs and a Wink hub for like $50 on sale. What more did people get for $300?

    1. Adam 52 Silver badge

      Re: $300?

      The implicit promise of a perpetual server?

      1. JeffyPoooh
        Pint

        Re: $300?

        Adam "...implicit promise..."

        Misleadingly Expensive.

        (Formerly known as 'Reassuringly Expensive')

    2. Dan 55 Silver badge

      Re: $300?

      Speaking about games consoles, it's not as if updates haven't removed features (PS3 Linux) and online services and game servers haven't been closed down. Smart TVs are another fine example of a thing slowly lobotomizing itself as time passes.

      People just complained a bit then shrugged and government consumer organisations did nothing so I suppose Google thought they could get away with it too.

      They might still be able to. It's very doubtful that IoT companies will supply server-side code in an easily installable package for Windows, Mac, and Linux.

  6. frank ly

    IPV6, TLDs, etc

    "That's why we need a DNS for IoT"

    I thought that IPV6 would provide 'space' for many, many devices of whatever kind. Can't the existing DNS system accomodate a massive increase in individually addressable entities given that it seems that any number of new TLDs are now possible.

    1. John Brown (no body) Silver badge

      Re: IPV6, TLDs, etc

      It probably could, if only the ISPs would actually offer IPv6.

      Yeah, yeah, I know, but Joe Average user isn't going to buy into an IPv6 tunnel and set it all up, even if it's not beyond the wit of many readers here.

    2. VinceH

      Re: IPV6, TLDs, etc

      I don't think that's the problem. It's that the devices are hardwired to talk to a particular service/address/whatever, provided by the originating company - so if that company goes away, or pulls the plug on the service/address/whatever, the device can no longer talk to it.

      Nominet seem to be suggesting a system whereby the device isn't hardwired to connect to that company/service, but instead to this new service; instead of phoning home to xyz, it asks the registry/DNS/whatever you want to call it* where home is. So if the originating company disappears or pulls the plug, someone else can theoretically jump in and save the day: the registry entry can be changed to point elsewhere.

      * I'm thinking IDIOTS - ID system for Internet Of Tat Services.

      1. Fibbles

        Re: IPV6, TLDs, etc

        Alternatively when IoT Lightning Co. go out of business you can quickly buy up their IDIOTS address and point it to your own server. A server which just do happens to be running a script which switches everyone's lights on and off in a strobe like manner.

        Bonus points if you also get their wireless speaker system to play obnoxiously loud House music.

    3. Mage Silver badge

      Re: IPV6, TLDs, etc

      Make no difference.

      The problem is gadgets requiring a so called "cloud" service. It could be IPV6 and "open source" and who runs the server it needs if the original company loses interest?

      Google Maps

      Here

      Nokia Maps

      Plays for Sure

      Skype on a TV set

      iTunes

      etc ...

      1. AdamWill

        Re: IPV6, TLDs, etc

        With a system allowing the 'phone home' point for existing hardware to be changed, other people could step in to do so: either as a community effort, or a commercial one. A few hundred thousand (or whatever) people with expensive hardware that suddenly does nothing could be a business opportunity for a savvy person: for $x we'll keep your hardware working for the next X months, or for $x/month we'll keep it working so long as you keep paying and we stay afloat.

        If the 'phone home' point is hardwired into the hardware (and that connection is secured), there's nothing anyone besides the original manufacturer can do to offer support for the device even if they wanted to, short of coming up with some way to hack the hardware and convincing users to apply it.

        1. PNGuinn
          Trollface

          Even MORE security ...

          "With a system allowing the 'phone home' point for existing hardware to be changed, other people could step in to do so"

          Oh what fun ...

  7. Fazal Majid

    Require IoT manufacturers to support cloud services for the life of the device

    And they will swiftly find a way to cut the dependence of the device on the cloud, possibly via firmware update.

    Google is notorious for abruptly discontinuing services, but those people bought the hubs before the company was acquired, so blaming the victim under the doctrine of caveat emptor won't work.

    1. DropBear
      Joke

      Re: Require IoT manufacturers to support cloud services for the life of the device

      "Google is notorious for abruptly discontinuing services, but those people bought the hubs before the company was acquired, so blaming the victim under the doctrine of caveat emptor won't work."

      It might not be that simple. You really can't blame the manufacturer for not being completely forthcoming from day one - when people bought the product, were they aware it's "cloud based"? Yes they were. Now go outside and look up - what do clouds do, huh? That's right. They slowly DRIFT AWAY...!

      1. Anonymous Coward
        Anonymous Coward

        Re: Require IoT manufacturers to support cloud services for the life of the device

        The problem is that many many people really have no idea at all that it's "cloud", or what that means. Fewer still will really understand the ramifications if the mothership disappears

        If the device still worked, but you just lost remote access then that's one thing. But for it to stop working altogether is quite another.

        I'm pretty sure that in the UK, it could be held to be a latent defect, and the retailer would be liable under consumer protection law. Getting the big retailers to refuse to stock stuff that's "faulty" is probably the only way this sort of thing will get reigned in.

        Either that, or a prosecution under the Computer Misuse Act ? Would that work - bearing in mind that the perpetrator isn't in the UK ?

        Second problem ...

        Most manufacturers define the life of a product as "as long as we support it". So they can arbitrarily define the life of the product as something other than the consumer expects.

  8. Alchemi

    This is why I don't trust third parties to process my data. Why does your fitbit need to send data up to be processed? Why do you need a third party to manage the cameras you have in your home and broadcast them back to you? Why does your thermostat need to talk to someone else to set the temp in your house. Why does facebook need your head motions?

    I mean, I ask this knowing the answers. It's easy for the end user. The are hurdles that a vast majority of people aren't competent enough to handle setting up the technical side. But I'll pass on allowing more than background intrusions into my house. I still stand by the opinion that Google is the nicest overlord we'll ever know.

    1. DropBear

      I think you're right, but I also think the raison-d'etre of cloud-services-by-default (well, beyond whatever money can be made off snooping on the consumer) is not so much the difficulty of installing a local server package (come on, most people can perfectly well click "next-next-next" by now...) but the general lack of a dedicated computer powered on 24/7 (= server) in most homes - required for continuous recording / logging / alerts / etc.

  9. Robert Moore

    Nest

    I was thinking about buying a Nest thermostat when I upgrade my furnace this summer.

    That's not going to happen now..

    1. Anonymous Coward
      Anonymous Coward

      Re: Nest

      That's what I was thinking....nobody who reads about this and realises that Nest are run by the same people are going to go near Nest now. Alphabet seem to have shot themselves in both feet with one bullet.

      1. Alan Brown Silver badge

        Re: Nest

        "Alphabet seem to have shot themselves in both feet with one bullet."

        Alphabet == Doubleclick, with all the sleazy stuff that goes with it.

        DC ate google from the inside.

    2. Jan 0 Silver badge

      Re: Nest

      Is your furnace in Port Talbot perchance? Nest may not be your biggest problem. TTFN.

  10. VinceH

    "Nest's bricking of Revolv serves should serve as wake-up call to industry punters - but, sadly, probably won't."

    FTFY!

  11. Someone Else Silver badge
    FAIL

    And this is somehow a surprise?

    And yet again, you expected a large megacorp like Google to give a flying fuck about customers? Customers don't increase Shareholder Value®™©. Especially Google, whose customer focus has repeatedly been shown to suffer from advanced astigmatism.

  12. martinusher Silver badge

    IoT is primarily a gimmick

    The whole concept of the Internet of Things as pushed by 'the usual suspects' is decidedly shady, its got everyone jockeying to monopolize a product space by trying to push protocols that they can own and control. They've been getting away with this by selling product to a not very sophisticated user base, the sort that are gadget users but don't really have much experience with automation in general.

    In the engineering world we've been able to connect anything to anything for decades. Obviously due to technology and cost limitations this hasn't been made into many home products but even if we could we'd run into the problem of "Why bother?". True automation has no interfaces but it requires a level of sophistication from the sensors and processing that little DiY jobs like thermostats lack.

    Anyway, I'm not going to bother much with IoT until the standards are open and are based on suitable protocols. The technology's out there; we just need to stop reinventing the wheel and just adapt well understood industrial automation techniques to this purpose. (....because I can say with some certainty that there's no way that you're going to persuade industry to adopt current IoT methodologies -- its one thing for a porch light to have a 'moment', quite something else when its a hulking great industrial machine).

    1. Anonymous Coward
      Anonymous Coward

      Re: Cart before horse

      All these IoT or wearable tech products are entirely corporate-centric designs rather than user-centric.

      Premise: People's data is valuable in aggregate. How do we ex-filtrate it?

      Solution: Create a bundle of sensors, and build some sort of toy around it to bait consumers.

      Challenges: What is the least amount of functionality we can include before consumers will take these Trojan horses into their homes?

      Maintenance doesn't even come into it. That effort is being spent developing software for the next generation product.

  13. Anonymous Coward
    Anonymous Coward

    That's useful information

    I had a Philips internet radio that relied on a Philips web page. That they turned off the web page, bricking their own devices until someone hacked up an alternative.

    I vowed never to buy Philips again, and have not.

    I regularly look into getting me some of them home automation gadgets, but there are too many companies pushing too many standards and I can never pick one I want.

    Now at least I know one company to cross of that list, and will be sure to pick an Open Standard.

    1. Mage Silver badge

      Re: That's useful information

      Still won't save you. Who pays for the server?

      The solution is things that work without the Internet, where Internet only adds optional value.

      Philips actually sold of their badges. They really only do lamps and health care. Back to the core pre 1926 business!

      1. Jan 0 Silver badge

        Re: That's useful information

        > Philips actually sold of their badges

        Wot?

        1. This post has been deleted by its author

      2. Doctor Syntax Silver badge

        Re: That's useful information

        "Still won't save you. Who pays for the server?"

        If it's open source you can run your own server. Is understanding that really so hard?

    2. martinusher Silver badge

      Re: That's useful information

      Phillips isn't the only company that requires the devices to phone home in order to work. Way back when Logitech made some nice Internet radio kit under the 'Squeezebox' label, it worked very well. Then came the corporate change -- suddenly the Squeezebox was Squeezebox UE ("Ultimate Ears") and it needed to be connected to home to work. There was an attempt to get everyone to 'update' their radios Windows 10 style but I don't think many took the bait.

      The phoning home is a combination of copyright monitoring and trying to sell music services. Its corporate and its silly. We all know that the technology needed to play an audio file is very simple so why do they insist on retrograde steps being improvements?

      I'm more or less retired so my unfulfilled Bucket List consists of collecting together, customizing or writing software that can be used to play audio at home that's "Corporate Free". It might be difficult to do streaming services -- the protocols are gradually being pushed towards proprietary, but what I really want to do is just play music off a NAS device. Nothing fancy, just what I could do without hassle a decade or more ago.

      1. frank ly

        @martinusher Re: That's useful information

        "... what I really want to do is just play music off a NAS device."

        Can't the modern Pi type of devices play your LAN/NAS music files, with WiFi and battery operation into the bargain?

        Going further, you can rent a Shoutcast server for $1 a month by Paypal that will stream at 192Kb/s (or less) to 100 listeners (but you'll be the only one listening to it, of course). All you'd need to do then is run a streaming player on a box at home and maybe develop a remote API for it so you can choose your music wherever you are, all under your control.

    3. Mystic Megabyte
      Go

      Re: That's useful information

      I have a Philips fan heater that is over 30 years old. It has a label underneath that says "Every six months unplug from wall, remove base, clean out fluff and oil the motor bearings". I've been doing that! It's so old that the neon light stopped working, I replaced it with one from a kettle that died in less than a year.

      I also have an equally old Philips digital clock. It has 15mm red LEDs and dims at night. It does not keep you awake like some cheap radio/alarm clocks do.

  14. Anonymous Coward
    Stop

    Simple solution...

    Don't buy IoT crap until you know there will be cross-vendor support.

    (Shame on you, Nest!)

  15. Emmeran

    IoT?

    Are you serious? It's a switch people get off your fat ass, walk across the room and flip it. Why you would ever give control of your daily life necessities to a multi-national corporation is beyond me. Expose my coffee pot to the internet? I think not, coffee is important.

    Forget it, I learned how to work a switch before I could walk with reliable stability I don't need Goog/Appl/Msft to do it for me now some fifty odd years later.

    1. Anonymous Coward
      Anonymous Coward

      Re: IoT?

      HTTP/1.1 418 I'm a teapot (RFC 2324)

    2. martinusher Silver badge

      Re: IoT?

      Browse the Internet for IoT and you get a lot of marketing waffle, all stuff of the "Wow! Its All One, Man!" sort. There are also literally armies of people lined up ready to dogpile on it when it becomes viable, people who can administer, provide ethical and spiritual guidance, possibly a bit of GUI design and so on. They're all waiting for the breakthrough.....it'll be fun to watch (from a safe distance).

      Meanwhile back in the real world I've been hooking crap up to networks since pretty much the day they were invented. Engineers have been doing this for ever, we actually got into networks because it was a whole lot less work than specifying, supervising and testing a cable with a couple of hundred wires in it. Its easy to interface stuff, not so easy to actually do something useful with it. For control centers the function is implied in the name but at home we don't tend to need one -- it was fun to build them when I was a kid but realistically what you're looking for is equipment that is both smart enough and reliable enough not to need constant monitoring.

      1. Anonymous Coward
        Anonymous Coward

        Re: IoT?

        "Engineers have been doing this for ever, we actually got into networks because it was a whole lot less work than specifying, supervising and testing a cable with a couple of hundred wires in it."

        Pure BS. Using "networks" has nothing to do with the number of conductors. There are plenty of standards for point to point data transfer over a few conductors. There are even buses like 1-wire that as the name suggests only needs 1 wire.

        1. SImon Hobson Bronze badge

          Re: IoT?

          > Pure BS. Using "networks" has nothing to do with the number of conductors.

          I think you missed the point, he's talking about replacing the older hardwired control and monitoring systems using hundred of wires for all the discrete connections, with systems where the information is passed over a network connection with just a few wires.

          For example, an electrical switchboard may contain many circuit breakers, and in a substation each of these breakers will have as a very minimum remote trip indication requiring a pair of wires back to a telemetry concentrator to feed the signal back to the control centre. These days they want to be able to monitor status (open/closed/tripped) and control it (open/close) - which would require something like 4 pairs per breaker all wired back to the concentrator.

          By networking it, they can have one network connection to a breaker, it can provide much more information (eg reason for trip - short term fault or longer term overload), perhaps report the actual load, and they can program it remotely rather than an engineer having to visit site to manually twiddle some control or (with newer kit) hook a laptop up to it.

          Large hardwired control and monitoring systems use a LOT of wires, and in a large plant they can be long ones at that. By adding networking, a lot (and in some cases, all) of that can be condensed down to one or two networks - though there are some functions (especially safety critical ones) that will probably remain hardwired for a very long time.

          1. Fatman

            Re: IoT?

            <quote>I think you missed the point, he's talking about replacing the older hardwired control and monitoring systems using hundred of wires for all the discrete connections, with systems where the information is passed over a network connection with just a few wires.</quote>

            and

            <quote>Large hardwired control and monitoring systems use a LOT of wires, and in a large plant they can be long ones at that.</quote>

            I know of what you are talking about.

            Having worked as an electrician in a past life, I was once given the opportunity to pour over the control schematics for an elevator. This was back in the day when the control mechanisms were RELAY based. The cable running from the elevator cab to the control box had more than 70 conductors. There were more than 100 conductors running from the in shaft wiring of the call buttons and direction lights to the control box, and then an equivalent number from the control box to the floor number display over each ground floor door. What made this even more problematic was that all of the conductors were of the same color, with only a printed number to identify each conductor.

            As integrated circuits became cost effective, I could see a transitions from discrete wires to discrete logic performing the same functions. In an elevator those functions would be floor number display, and floor number selection in the cab. On each floor, you have the up/down call buttons and the up/down direction lights.

            Transitioning individual lamps to a numeric LED floor display would be the 'easiest'. You could use simple BCD to 7 segment decoders in conjunction with a 7 segment LED. A single digit will get you up to 9 floors, a 1-1/2 digit will allow up to 19, 2 full digits will get you to 99. All you have to do is just output the floor number as serial data. This would require only 4 wires (power, ground, data+ and data- assuming a balanced data circuit. Hey, doesn't that somewhat resemble a USB 'buss'???)

            How to deal with the call buttons on each floor. simple, IF at each floor, the call buttons had some discrete logic that would pass back the button status, not unlike the way the keyboard in front of you is polled by the computer it is connected to; one could use only 6 conductors to service a stack of call buttons. The electronics located at each floor would be field programmable to respond to a specific 'address' sent serially over a TX pair of wires. The status could be sent back as UP_BUTTON_PRESSED or DOWN_BUTTON_PRESSED over two wires, and then only require power and ground to operate. The direction lights over the doors could be left alone.

            The end result is to reduce the clusterfuck of wires to just a few.

            1. Stoneshop

              The end result is to reduce the clusterfuck of wires to just a few.

              Exactly.

              The home control system* I use started from the requirement that I needed to control two areas with two chains of workshop lighting from three locations each. Conventional switches and cabling would require four (of twelve) switches being quite expensive, and hard to get in a splashproof version. Pushbuttons and stepper relays appeared more promising, but it still didn't grab me as the right solution. The system I use now has allowed me to add movement detectors to switch off the lights in case I forgot, as a supplement to the manual switches. Door and window sensors will tie into the heating and ventilation, the awning windows and skylights will react to outside versus inside temperature, and all of it will depend on anyone being at home or not. But none of this is controllable from the outside, except maybe choosing some particular presets via SMS some time in the future..

              * Two-wire serial+power bus, network connection to the local LAN only is used to display status and allows one particular system to load settings and logic into the sensors and actuators.

          2. Alan Brown Silver badge

            Re: IoT?

            "By networking it, they can have one network connection to a breaker"

            Actually it tends to be one network connection to the bank of breakers and some form of CAN strung between them all.

  16. Kanhef
    Joke

    Unfair comparison

    A tub of hummus is quite useful – and delicious.

    1. Loud Speaker

      Re: Unfair comparison

      A tub of hummus is quite useful

      Not in the cloud.

      The lesson here is that the cloud is "here today, gone tomorrow" and cannot be depended on, even if the service is provided by the biggest company on the planet.

      If you need a server, buy one and put it in a rack. (I suggest buying three is more reliable).

      Giving someone else money because they offer the performance of 3 for the price of 0.3

      is likely to lead to you paying the price of 3 and getting the performance of 0.

  17. Anonymous Coward
    Anonymous Coward

    Here's how the competing IoT market will shake out

    1. an open source solution will be developed, which will eventually fork into several competing versions, but it will be years until it is packaged into products accessible to the masses. Basically it will be the IoT equivalent of DD-WRT and OpenWRT.

    2. Apple will have their own, sort of like iDevices where third parties can make devices to interface with them, but only if you follow Apple rules, and they'll cost more, but it will work pretty smoothly for what it does, even if it doesn't have nearly all the bells and whistles possible in the open source solutions

    3. Samsung will have their own, with an 'S' in the name, that few people will use and be considered irrelevant by almost everyone who lives outside of South Korea

    4. some major US company like GE will introduce something with major fanfare, that will only work in a home that's all GE products so the only time you'll encounter it is in 'spec homes' where the builder is designing it as a "smart home" hoping for a bigger markup

    5. Google/Nest, having lost their trust in the early adopter market by killing Revolv, will fade away but everyone who owns a Nest thermostat will continue to make ridiculous claims that their energy bill is 30% lower

    1. Anonymous Coward
      Anonymous Coward

      Re: Here's how the competing IoT market will shake out

      "2. Apple will have their own, sort of like iDevices where third parties can make devices to interface"

      They already do. It's called homekit.

      "3. Samsung will have their own, with an 'S' in the name"

      They already do.

      "5. Google/Nest"

      Google is pushing a solution called Brillo that is based on android and open source.

      I doubt they will go away. I suspect Google and Amazon will be the two big players in hosting the upstream infrastructure for IoT devices.

      1. Andrew Moore
        Coat

        Re: Here's how the competing IoT market will shake out

        "Google is pushing a solution called Brillo that is based on android and open source."

        So I've I equipped my gaff with this kit, it would be known as a "Brillo Pad"...

    2. localzuk Silver badge

      Re: Here's how the competing IoT market will shake out

      With the "savings" from things like the Nest thermostat, I don't understand how that actually happens.

      Our heating is turned on, manually, when the house is cold. It is set to turn on in the morning, when cold, before we get out of bed to warm the house up, and again at night before we get home.

      I don't see how having a fancy thermostat would reduce the heating bill? It only gets turned on when its cold at the moment. What's it doing that saves money?

      1. Anonymous Coward
        Anonymous Coward

        Re: Here's how the competing IoT market will shake out

        "I don't see how having a fancy thermostat would reduce the heating bill? It only gets turned on when its cold at the moment. What's it doing that saves money?"

        You might be disciplined enough to only manually turn on the heating when it's totally necessary but a lot of people aren't.

        Say Joe Bloggs has his heating on a timer in the winter. Warms up the house for the morning rush, turns off, warms up the house for when everyone is home.

        - Some mornings aren't going to require heating; A smart device could work out if it really has to turn on and only do so if it must. Energy saved.

        - Some evenings no one is home; A smart device could tell if anyone actually came home and delay or skip the heating cycle. Energy saved if the heating cycle is skipped or delayed because they won't be heating the house twice.

        1. Anonymous Coward
          Anonymous Coward

          Re: Here's how the competing IoT market will shake out

          Who has "his heating on a timer" that just turns on at a specific time regardless of the temperature? Is that somehow cheaper than a programmable thermostat, which is $20 or less based on a 2 second google.

  18. irwincur

    This is why I don't ever buy into or get too involved with anything in the Google universe. They are too willing to just end of life services and products.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Other stories you might like