Google's Nest seems to be setting out a home automation standard, partnering with Whirlpool, Mercedes-Benz, Jawbone, LIFX, IFTTT, Logitech, and Chamberlain. Over the weekend, Quirky announced its $79 Wink hub, partnering with Home Depot, Honeywell, GE, Rachio, and Philips.
The two announcements were obviously timed against each other. It's hard to tell who is reacting to whom, but my bet is that Quirky is hurriedly reacting to a pending Google I/O announcement this week, and getting the word out by giving the NYT an exclusive [1]. It also looks like Nest has an actual, planned microsite vs Quirky, who only has mention of GE on their website (https://www.quirky.com/ge) and doesn't have much of an official Wink page yet (https://www.quirky.com/wink is bare).
Add to that Apple's Homekit, announced at WWDC, partnering with Osram Sylvania, Skybell, TI, Chamberlain, August, Honeywell, Haier, Schlage, Philips, Black and Decker, Netatmo, Withings, and others. More brands/logos, but it's early.
Regardless, it's at least a three-horse race and we'll see it get worse before it gets better. All have serious brands on board and consumers will be confused for quite some time. A single, open, interoperable standard is coming but the race will encourage a flurry of product investments to get smarter devices in the initial stages.
I've been in this hobby for a long time (like since X-10 in the 80s, more Insteon now) and one unwanted trend I see is exploding prices. Inflation has pushed your stereotypical X-10 appliance module from $10 to currently $20. I've been mostly Insteon for a decade or so now, but the venerable X-10 appliance module is a good canary as its been around for decades unchanged. Anyway its off to the races with the new breed, hundreds of bucks for a mere thermostat, etc. The days of your linux box turning your aquarium tank lights on and off in a cron job for $30 total are pretty much coming to an end, although it'll still be possible, much slower and harder to automate and it'll cost $750 and have a phone app that demands access to your contact list, but it'll still sorta work.
Another interesting trend in home automation is abandonment of all legacy tech. Ask someone what came before the arduino in microcontrollers, "uh, nothing, they were the first ever". Ditto the new wave of home automation, the zillionth competitor still positions themselves as the "first". Not too impressed with this trend either. If I'm making fun of this marketing scheme, I'm probably not the only guy in the target market making fun of the trend.
A final interesting industry trend is the conspicuous consumption aspect. Nobody seriously tried to claim an early 80s era X-10 wall wart was a designed thing of beauty to show off to your friends, any more than a circuit breaker or a gas shutoff valve. The new wave is all about the bling as the highest priority. Does look pretty nice.
So that's the four trends I've seen in HA, your observation of extensive partnering agreements, the cost is exploding, every copycat is the first product ever, and appearance is the #1 priority. None of the trends could have been extrapolated from the earliest days in the 80s to 2010 or so.
I suspect interoperability is NOT going to turn out any better for HA than it has for home stereo-TV-dvd-xbox remote controls, or kitchen appliances (try swapping egg beaters between competing portable mixers for a good time, if you have a deep pocketbook)
Otoh, home automation might finally go from being a hobby to something normal people actually can use to make their lives easier. I also dabbled with X10 in the 90s and personally I can't wait for apple to come along and do it right. Needless to say, I completely disagree with your final observation.
"normal people actually can use to make their lives easier"
Normal people don't like capital investment of any sort either monetary or time, and newer technology seems not to change the total investment at all, beyond a modest increase, but only adjust the ratio between the two categories.
There is also the stereotypical UI tradeoff that has existed for decades in computers of "make simple things easy" which is almost always in direct opposition to "make difficult things possible". The tradeoff in the new wave of HA products is dangerous because there is the risk of not gaining traction with the noobs because simple things do not help much, while losing all the experts by making the really productive yet difficult stuff impossible.
From a "do it right" perspective, X-10 done the right way is the decade or so old Insteon system, reliable, 2way, giant address space, fast (or, faster than X-10)... "Home Automation" done the right way seems an undefined problem, and that has to be done before the solution can begin.
On the contrary, most people with the income to do so are in fact delighted to invest in things that simplify their lives and save them time. It's one of the pleasures of the modern era.
Regarding this claim about UI tradeoffs, you clearly do not remember the past seven years. Were I to travel back to 2006 and hand you an iPhone 5, you would think it was a gift from some distant alien civilization. The point being that great design continually defines upward what "make simple things easy" means. It is not diametrically opposed to making the difficult things possible.
I really, really, wish we could skip the whole vhs/betamax stage for once :-( Not only is it tiresome, it inhibits adoption across the board, we've all been there too many times before.
It's probably unavoidably a time of turmoil - with so much at stake there's a disincentive for all companies to universally "place nice".
A cambrian explosion of standards and partnerships from which most likely one or two durable standards will emerge, hopefully the "best", is natural - but painful if you're impatient to just get something done.
I'm glad you mentioned HomeKit, because at this point, I'm really starting to fight my desire to go out and hook all of these things up to make sure they work with HomeKit. Apple doesn't exactly have a history of really releasing the standards they say they will (looking at you, iMessage), but I'm all for the idea of standardized home automation.
I'm misremembering - it was FaceTime, not iMessage. Whoops.
Unless I'm continuing to misremember, there was some patent issue that blocked FaceTime from working over cellular on the first version, but it was cleared up when AT&T finally accepted that they were the only carrier not supporting it.
The patent issue is a DoD-connected company forcing Apple to abandon P2P and break the end-to-end encryption. Not unlike what happened with Skype, though quite a bit more hostile.
There are standards for Home Automation communication out there - ZigBee and Z-Wave. These are the low power communication protocols that home automation devices can use to talk to each other (WiFi is too power hungry and heavy weight). If you take apart a Nest thermostat you can find an unadvertised ZigBee radio buried inside.
ZigBee is an open standard developed by an alliance of device manufacturers. Z-Wave is a proprietary standard developed by Sigma Designs. Both are able to create a low power mesh network for communication between home automation devices. These protocols have been around for a while and are supported by 100s of device manufacturers.
That said, I have no idea what Google/Nest or Apple plan to do. They may be creating their own standards for communication, or they may build a communication protocol on top of ZigBee or Z-Wave.
If you're interested in working with an interoperable standard now though, you can buy off the shelf ZigBee or Z-Wave radios, devices that speak ZigBee/Z-Wave, and build your own HA solution. However, I've found through personal experience that interoperability between devices using the same standard is not always perfect. I'm looking forward to the heavyweights getting into the ring and hopefully cleaning up the industry and making it more interoperable.
It's very, very competitively priced! Nest, in the UK at least appears quite expensive. I also dislike how currently the intra-operability is provided via cloud services, from both a privacy and reliability of my internet perspective.
I'm talking about the electronic version of breaking and entering, which would be incredibly difficult to prove someone did. And if falsely accused, would be incredibly difficult to disprove.
sure, kinda like if someone picked your door's pin tumbler lock... or if someone accused me of picking theirs since they knew i carry a pick set w/ me.
I'm really curious how this will pan out. In Europe (or at least here in The Netherlands, not 100% sure about the other countries) we can't use the Nest thermostat because our central heating boilers and thermostats use the OpenTherm protocol. (http://www.opentherm.eu/)
Also, most boilers in Europe are of the modulating kind (not a simple on/off switch, but a more graduate gas consumption. Side note: I really don't get why they don't have this in the US. It's so much more efficient...).
So, we can't use Nest. Apple and Quirky announced they'll be working with Philips and Honeywell, two companies that do offer modulating, OpenTherm thermostats. No word on if they're actually going to integrate OpenTherm, though. I'll be following this closely!
I'm curious as well. It's very nice that the Mercedes integration is in theory just an app, but like you I don't have a house that's compatible with Nest.
It would be really nice though if there was an open protocol for this stuff. I'd like my car/phone to tell my house that I'm on my way, even if I'm not using anything Nest-like.
Sadly, a sense of realism tells me open protocols aren't really what any manufacturers have in mind. So far it's all tightly integrated walled gardens.
I think they had to change the design for the UK (compared with the US model), but it was more to do with the electricity supply than the boiler design.
It gets pretty interesting when you connect all the pieces Google now has.
You car has build in Android device. Your home is monitored and controlled with Nest devices. On your wrist you carry Android wearable. Google Glass tracks what you see. For online activities you use Google products (Chrome, Gmail, Calendar). You consume your media via Android table or Chromecast.
Google knows about your your car. It knows how you are doing, how much and when you exercise, when and how you sleep (vital signals via wearables). It could know who you are meeting with or what products you are browsing at grocery store or what you are reading from news paper (Google Glass). It knows and controls your home. It knows about your planned trips, about stuff you have ordered online (Gmail, Calendar).
One major area that is lacking is nutrition, monitoring what and how much I eat. Google Glass might help here (or products like Vessyl).
Interesting to see what they can make out of this.
Given Google's fondness for shutting down APIs and services, I'd be very disappointed if they monopolised all of these markets. They'll be the Microsoft of home automation. Using it to trap people in their monolith (and Android is just a vehicle for them to deliver ads, folks), and then letting it stagnate or shutting it down if the ad delivery doesn't prove lucrative.
I've been burnt so many times by Google I've lost count. Shutting down the Finance API, banning me from Adsense without explanation, shutting down Reader, cutting email deliverability across gmail from my server. Tying the wrong product reviews to my website on Google search.
I don't want one company to rule the internet. I don't want a MS-esque services monolith. I want a few product-focussed companies like Nest to exist and compete indepently using open standards.
>and Android is just a vehicle for them to deliver ads, folks
They are dealing with that 'problem' by steadily de-opensourcing all of the components. You don't use the email app, you use gmail. You don't use the camera any more, you use "google camera". You don't even use the keyboard any more, you use "google keyboard". A lot of the new android functionality is piled into 'google play services'.
I don't think the day is that far off where the 'next' version of android will be all closed source, and if you want all of those closed source apps that you depend on to continue working, you'll need to install it.
At that point they can start charging fees for android's use.
While cbeach's comment that you replied to is trollish nonsense, in response to your reply-
You don't use the camera any more, you use "google camera". You don't even use the keyboard any more, you use "google keyboard". A lot of the new android functionality is piled into 'google play services'.
Or you use other email clients, alternative cameras, or alternative keyboards. That's kind of the point of Android, and in splitting their own apps off Google is simultaneously making the platform better by opening the APIs required by them. For instance the camera API has gotten dramatically better - http://source.android.com/devices/camera/camera3_3Amodes.htm...
At that point they can start charging fees for android's use.
Google already charges partners for Google apps. Samsung is paying Google. This has always been the case with Google. Further most of those apps have been separate concerns since something like Android 1.5, and the platform is dramatically better for it (e.g. the whole fragmentation thing is far less of a concern, because the core platform shrinks to just core services, exactly as it should be).
Just wanted to say I agree 100%. I expect that Google is going to do very well with all the building blocks they have in place, but I view that prospect with a great deal more trepidation than excitement. (Although I wouldn't say none of the latter; things like self-driving cars are certainly cool, and having Google behind that sort of project makes it much more likely that it will happen in the near term.)
Maybe they are looking new revenue sources and don't want to be just in the ad business anymore. With the Nest line or products I could easily imagine them extracting over $1000 from a household that goes all-in with their their product offering (thermostat, few smoke detectors, few Dropcam devices). And there's also good potential for selling add-on services for a monthly fee.
Something completely new business might be available on the "smart grid" side. If you are controlling the heating and air conditioning you have pretty strong influence on the amount of electricity the household is using at a given time.
I don't think this can exist without at least some outside intervention. The monolith has lower coordination costs, potentially better usability (at least at first) and extra opportunities for profit by violating your privacy. So it's more likely to succeed in the market. And once it has dominant market power, we're back to "embrace, extend, extinguish".
Combine this with their push into machine learning and AI - they have more data points into people's lives than anyone else and will be able to do more with that data than anyone else.
And sell more ads to you in more places than anyone else, I guess.
Yet another hammer looking for a nail. There is no killer app in home automation. Hackers like this stuff, because we like technology for technology's sake. But everyone else has a higher standard for adopting new gadgetry. It has to actually solve a problem, or otherwise provide value. Even the best use cases are largely novelties that will get about two days use. Turning on lights remotely? How often do you need to turn on lights when you aren't physically in the space being lit. Starting a washer/dryer/oven/dishwasher remotely? How will you load/unload the clothes/food/dishes remotely? Unlocking your door with a phone instead of a key? It looks cool the first two times, but novelty isn't enough.
Everyone seems to be worried about the competing standards and walled gardens, which assumes that this space will take off to the point that it even matters.
I'm JUST getting into a little home automation, and here are some great use cases that aren't just a novelty:
When my wife and I both leave the house, the front door locks itself. We never have to worry about if we locked the door or not. We can also leave and enter without having to carry keys, which is great. I can also unlock the door remotely when my in-laws stop by unexpectedly to drop something off.
I have a room in the basement where the light switch is on the far side of the room, so you previously had to traverse a pitch black and somewhat hazardous space to reach the light switch. Now it's all run on a motion detector that turns the light on and off as needed automatically.
And we just got a Nest. My wife, a non-tech person, loves the front door lock magic and the Nest. As more stuff comes out, and I add more stuff to the system, the utility will only increase.
I'm looking forward to things like smart security systems that know when I'm home, away, sleeping, awake, have company, etc... And interact with tokens (for instance if it's 2 AM and someone pulls into my driveway, I want to be woken up and alerted. If it's 2 AM and I'm pulling into my driveway I want my wife to be able to sleep). I'm looking forward to music that follows me through the house, and turns itself down when I get a phone call.
I'm optimistic about nest - they seem to understand this, their thermostat is a great user experience. You hook it up and use it just like you would use a normal thermostat, and it uses that info to make intelligent decisions to do actually useful things - like override the schedule and turn the heat up when you take a day off work, or turn the heat down when nobody is home, all with no user interaction.
Remote control isn't the killer app for home automation, no control is. I don't have a need to turn lights on in rooms I'm not in, but it would be nice if my house could turn the lights off for me when I forget to. Or if I drive off and leave the garage open, it could close it for me so I don't have to call a neighbor and ask them to check it.
> "like override the schedule and turn the heat up when you take a day off work, or turn the heat down when nobody is home, all with no user interaction."
Useful sure, but is it that useful? Which is to say, will people choose it over walking over and twisting a knob?
The same argument can be made for motorized blinds. Sure, yeah, it'd be kind of cool for your blinds to raise and lower themselves with the day/night cycle. Or I can just yank that cord over there.
Ditto the garage door thing. It'd be nice to have a system that just closes the door for me, but I'd call my neighbor, what, once a year to check on my garage? Worth a $100 system + installation + integration into some larger, overarching home automation network?
That's the problem with home automation, it's a bunch of janky, expensive, high-maintenance tech that solves a "problem" that nobody was really that bothered by. It's an incredibly minor incremental improvement that solves a pain that was never actually painful, and it has historically required a lot of infrastructure/installation in the home.
The new stuff is lowering a lot of the installation/infrastructure pain, but it's still relatively expensive, and for the most part their improvements are so minor, and the original problem so inconsequential, that I don't think they'll see mainstream market success. IMO none of this will take off until they find the "Killer App", the one that solves a problem that really bothers people.
I don't know how everybody else gets their nests, but I pay flat rate for power and my power company gives them out for free because it works out cheaper for them in the long run. So no, I probably wouldn't pay extra for a nest thermostat and the small amount of convenience it offers, but I like having it.
Once the Nest API gets more mature, watch power companies give these out with the ability to remotely turn your A/C up if the power grid starts to sag (that whole "rush hour" concept they're pushing).
Remote control isn't the killer app for home automation, no control is.
I think you've got this exactly backward, at least as it applies to the Nest thermostat. I find the automation aspects to be pretty gimmicky, but turning the A/C down before I head home from work or turning it off completely just before I leave town on an airplane are killer features.
I've had a Nest thermo for about a year. For me the #1 thing about it was the smartphone app - in particular the ability to tell it to turn on the heat if we were coming home earlier than normal (or the opposite). Our radiators take a while to warm up the house and this allowed me to keep the house fairly cold when we weren't there w/o any discomfort when we came home.
That said I was disappointed with its learning capabilities in terms of our schedule. Sure, it picked up on trends but it just started slapping temperature change event on the schedule instead of adjusting things. So what happened was that it would get a couple of pinch points where it'd set to 65 at 10, 67 at 10:02, 66 at 10:05, 68 at 10:10, etc and I ended up cleaning it all out by hand anyways. It'd have been nice if it recognized we probably wanted it around 66-67 around 10:05 for instance.
I'd argue there is huge value in app controlled 'home environment' devices (light color & brightness, music, air/heat/fan, etc). When the software controls are sufficiently convenient, adjusting the environment becomes an automatic behavior like scratching an itch or shifting in one's seat.
Of course they look good. They were designed by the Nest team, not by Google's own designers. Nest was an acquisition, and I can't see those guys getting involved in the huge challenge of making Google's other properties look good.
Yea, they bought it. Doesn't that give them the right to call it their own product? Will that always apply to future Nest products, or is it only for the ones that existed at the time of acquisition? If 'we' acquire something at work, it becomes ours. We incorporate its results and start being measured by its performance.
Well nest is still independent from what they want people to believe at least, so it's actually unlikely that the nest designers will help any other Google products.
It's like saying Tina Fey is part of Comcast's acting talent because of her role in 30 rock.
True, but in this case the design is actually really good as well. Nest has the best unboxing and set-up experience I've ever seen. I was blown away by how much thought they'd put into it, and how easy they made everything. It's also really well designed from an on-going experience perspective... but I never really have to touch it now that it's installed and programmed.
There's a huge support cost in providing self-host, together with an impact on image (ease of use, etc). With the tiny proportion of people who actually would self-host, I don't think it makes any business sense.
I want a self-host, and generally resist on any technology unless I can, but at the same time I don't expect any company looking to sell products to actually do it.
So self-host generally only goes hand-in-hand with non-commercial projects.
My primary concern is my house becoming part of Google's walled garden. "Oh, my garage door doesn't work anymore as it is controlled exclusively by 'Works with Nest.'"
Could you imagine the PR fallout if that did really happen? I'm pretty sure Google is going to be smart here and have graceful fallbacks when the interconnects don't work (for whatever reason).
Self host and maybe I'd care about it. I could put it on it's own segregated wifi connection with no access to the rest of the world and the data would be mine.
I have a Nest. Nest doesn't allow you to access your own data that they are no doubt collecting, and their app only shows your energy use for the last 10 days. This makes it absolutely useless for trying to figure out how to optimize your energy usage on a month-to-month or season-by-season basis.
For a while, you could pull that data through a private JSON API, the same one the iPhone app used. I'm assuming something changed a few months ago though, because the application I wrote to chart out that information stopped receiving those data points. This was probably part of the transition to a public API.
Using the API, I was able to see that running just the fan in my house didn't appear to do anything for the temperature of the house, despite the temperature difference between the two floors. It was interesting charting internal temperature against outdoor temperature as well, giving me a visual reference on the efficiency of the system.
One of the prohibitions when signing up for the developer program is this:
> [Prohibited] Collect, aggregate, re-syndicate, retain, log or store Customer Data (as defined below) received via the Nest API beyond 10 trailing days from the date when the data is received.
I wonder if the privacy "advocates" (and I really want to use the word "whiners" here) are to blame for that particular restriction. I can't see a business case for Nest to disallow that.
We're looking at the first top-tier market adoption of a home automation system, albeit it amongst two, soon to be three, products and the opening salvo that'll really stir the pot with other players (Honeywell, Apple, et al.) and open wide the automation market for maturation.
I worked developing interfaces for bespoke home automation systems with a small company out of Georgia in the early 2000s (Listman, if anyone's even familiar with them; they were a blip on the radar later bought by Leviton but quite advanced for their day) but all of these were closed-circuit systems with a then-innovative iPaq to control their environment from afar.
How far the industry's finally become since those days with this announcement; actual platforms are finally here.
In other words, Google wants to turn Nest into a platform for the house, some kind of Android equivalent. Looking forward to seeing their next steps. If I'm right, some kind of appstore I would imagine.
So far they've focused on big brands, if this fails it would be because of insufficient trust from developers, but Android shows people can look the other way if there's enough of an incentive.
To be fair, at the time we were terrified the much more closed Apple platform would win.
To be honest this time I'd rather there was a monopoly, as at least it is then completely clear the only viable competition is open source, so that's where you have to go.
With the government, every tech company and their mother tracking us 24 hours a day 7 days a week, why should we continue to add our personal data to the pool?
Extrapolation != predicting the future. You can't just download an android app to fluff your pillows before you get home, it's about integrating the house with new devices.
Yeah, but mine's got a crack on the screen. I stopped caring to notice it. It's a couple years old, hasn't had a new app installed on it in months, and is fulfilling many utility roles in my world.
The original Nest device is so popular among the general population because of its UX, display and colorful feedback.
"Works with Nest" is amazing for us techies, who know the technical feat behind all this interoperability, but to the average consumer transparent technology and complex automation is a really hard sell.
The mass market consumer still want to touch, feel and see the things they pay for. However as a 'shop hub' for great designed devices that play along, "Works with Nest" is great. Also I am sure "Works with Nest" is a great vehicle for corporate deals of Google, since it allows them to do co-promotions (such as Mercedes) as side-projects which are usually embedded in such agreements.
Popular amongst whom? They have sold only a million of their thermostats since they came out. if anything their marketing and web facing has conveyed them as more popular than reality.
As interesting as they are they are priced out of the market for far too many people. Considering many homes come with programmable thermostats, in some areas new homes have to have them, this becomes close to a frivolous purchase.
If anything it should tell entrepreneurs that there is a golden opportunity for a low cost home automation solution.
Oh, its UX, colorful feedback, etc... uh I cannot recall the last time I ever cared about my thermostat. Its set and forget.
I am a little bit unenthusiastic about Nest integration into the Mercedes Benz where it tells Nest to warm up the house prior to arrival. What a waste of energy! Is it so difficult for physical human body to endure some discomfort while the house warms up? Imagine thousands of household doing something similar.
OK, there could be a use-case where the elderly is due to arrive home and needs the house to be at the right temperature for health reasons. But this is clearly not the use case being put forward by Nest. It uses s luxury Mercedes Benz and is spinning it for luxury market.
Wasting/saving energy is not as simple a calculation as that. When energy is at peak use (say in the winter, around 5:30/6:00 when everybody returns from work), it becomes more expensive for energy companies to procure the requisite energy needed. During these peak times, they often have to buy energy from other companies at a higher cost or find less efficient means of generating it (e.g. coal) to meet the demand. Getting an accurate picture of estimated energy demand can actually save a lot of money/energy for customers.
I work at Nest and it may be a little unclear at the moment (we're working on it), but The Nest API does not guarantee that if you send an ETA of 15 minutes (docs: https://developer.nest.com/documentation/eta-guide), the home will heat to the appropriate level in 15 minutes. It's actually filtered through our algorithms to determine the best course of action to not waste energy but also provide the appropriate level of comfort.
I looked into the nest but I don't think it would save me any energy. And I'm wondering if that's actually true for lots of people.
I have a modulating (i.e. change flame level) condensing boiler (the highest efficiency available). I programmed it to run as close to 24/7 as I could, because that way it uses the lowest flame, which is the most efficient flame.
When I tried a setback thermostat when the boiler attempted to rewarm the house it shifted to a higher and therefor less efficient flame. So I gave up on that and let it run on low all the time. So there would be no peak usage by me.
As people shift to more efficient ways to heating I suspect that this will happen to everyone. Not just modulating boilers, but also multi-stage heating with a heat pump. By trying to rapidly heat the home you can't use the heat pump.
It's also true with an A/C - the faster you are trying to cool the larger temperature gradient you need, and therefor you lose efficiency.
It's also true with an A/C - the faster you are trying to cool the larger temperature gradient you need, and therefor you lose efficiency.
I'm not sure that's always correct. The Coefficient of Performance (heat moved/work done, higher COP is better) of an A/C is:
COP = T_hot / (T_hot - T_cold).
T_cold is the inside temp, T_hot is outside of the heat exchanger. The narrow the temperature range, the better the COP. Thus the A/C will actually be more efficient in terms of heat moved per watt while it's cooling a warm house down than when it's already cool.
There's one other factor that applies. The rate of heat transferring into the house is proportional to the temperature difference. Once the house has warmed up, you reach equilibrium. The net heat flow stops, so the total amount of energy you need to remove later stops rising. If you run the A/C all the time, heat is transferring in all the time. So the total amount of heat you need to move back outside in a day increases significantly if you leave the A/C on.
> I programmed it to run as close to 24/7 as I could, because that way it uses the lowest flame, which is the most efficient flame.
The energy loss from your home increases as the differential between inside and outside increases, right? So at night or when you're out, if your house were kept colder, then you'd lose less energy to the outside.
So by keeping your heating running 24/7, you're wasting some energy in the extra energy loss when you're out, right?
How does this compare to the efficiency gain by running your more efficient flame? Have you calculated or measured that you actually save energy overall?
> I work at Nest and it may be a little unclear at the moment (we're working on it), but The Nest API does not guarantee that if you send an ETA of 15 minutes (docs: https://developer.nest.com/documentation/eta-guide), the home will heat to the appropriate level in 15 minutes. It's actually filtered through our algorithms to determine the best course of action to not waste energy but also provide the appropriate level of comfort.
So you're accepting the ETA request not as a demand to obtain a certain temperature, but as a request, and then still taking utility demand information into account. Excellent.
I am not talking about wasting money, I am talking about the actual energy itself as money is more renewable than energy. Once it has been used, it is extremely difficult to get back.
Firstly, the premise of that scenario, is the house is cold while the person is gone. That means SAVING on energy costs by not heating the house unnecessarily while no one is inside, which is the current de facto standard. (Do you turn off the heat when you go to work? I certainly don't)
Secondly, it's silly to object to innovation that allows humanity to be frivolous or wasteful according to their desires. Where do you draw the line? Why have central heating in the first place? Is it that difficult to wear a sweater indoors? If people want to be warm, they will utilize technology to be warm. And if that technology progresses to allow this process to be simple, efficient, and automated, people will use it. And not just those in the luxury market.
Yes I have the heating on a timer and its off when I am at work. I turn it on if its cold and I working from home or its the weekend. I believe this is normal...
Any HVAC system installed or revamped in the past 15 years is likely to have a programmable thermostat. Whether people are using these _correctly_ or not (which as others have touched on here is not necessarily super intuitive, part of the problem Nest is meant to solve), I think attempts of some sort at saving energy with these is pretty common.
Many years ago I instrumented my house using a wireless network of 20 motes we had spare from a research project. One of my conclusions was that the house warmed up and cooled down quite a bit slower than I had assumed. Our heating at the time was on a timed thermostat. I ended up moving the "on" times 30 minutes earlier and the "off" times an hour earlier. We saved energy and had a more pleasant environment.
This was using dumb thermostat technology. If we were back later than usual, the house still got warm. If we stayed up later than normal, the house started to get chilly. Smart thermostats that know what you're doing can help even more, but the less predictable you are, the more the gains from the thermostat being linked to other systems to monitor what you're doing. If you're home the same time every day and go to bed at the same time, the gains will be less.
BTW, I also discovered that our dishwasher warmed the kitchen by 2C. The obvious conclusion was that it was best to use its timer to start it at 4am, so the waste heat has already warmed the kitchen by the time we're up for breakfast.
Millions of thousands of homes today simply leave their thermostat at the same temperature all the time, failing to optimize for the large portions of the day when nobody is home because it's simpler (and they probably don't have a timed thermostat, or they can't figure out how to set it up.)
Imagine thousands of households doing something smarter.
Not sure this is so common in the uk. My heating is almost always off, and if it gets cold we have a timer and a thermostat, so it comes on for an hour in the morning and a couple of hours in the evening.
I have noticed a trend among our friends, (not technical folks), who actually turns off their heating when they go out. It is as common as turning off their lights when they leave the room.
I guess Nest is heading towards this direction, but pre-warming a house very "first-world".
Not to keep harping on the subject (see my comments above), but pre-warming or pre-cooling houses is actually a really great solution to solving our energy problems with algorithms. It's not strictly a "first world" problem.
Imagine a neighborhood of 100 homes where you know all 100 people will return approximately at 6pm in the winter, and they all want 72 degrees. Taking into account the insulation of each home, if you could pre-heat the homes on a rolling schedule (say, heat 25 of them to 75 degrees at 3pm, accounting for 1 degree of heat dissipation per hour, heat another 25 to 74 degrees at 4pm, etc.), you will actually in aggregate save a lot of energy and cost because of inefficiencies in the way we deliver and generate energy during peak times.
That is an interest way of distributing loading of the peak heating requirements. Does Nest have a ability to interact with the other Nest in the neighborhood? Or is it all controlled from a central server?
Not (yet) with other Nests that I'm aware of, but they're making deals with the power companies. Users can get discounts on their bill for allowing the power company to bump energy use down a degree or two during peak times.
Owning a house with central heating / AC that can be controlled by a Nest Thermostat via your household wifi from your smartphone while driving home in your Mercedes is a pretty "first-world" thing.
Especially given that this works best for second homes like a weekend house in Tahoe where it might take a full day to warm up if you left the heat off all week.
When your house is not insulated properly and you live in a cold climate, it's actually better to not turn the heat off completely. It takes more energy to warm a stone cold house than to keep it at approx. 62 degrees.
A simple, programmable thermostat will make up for the purchase within the first year of service.
This is not true. It might seem that way because of how long it takes your house to warm up. There is no momentum or stiction with heat transfer. When the inside of your house is warmer than the outside you are losing some amount of heat to the outside proportional to the difference. Any heat you add to maintain that difference when you are not home is wasted. As long as you can tolerate a stone cold house when you return, your best efficiency is to just turn the heat off. This is assuming you have an on-off heating system, someone mentioned a throttling system is more efficient at a low flame rate so that made it better to leave the heat on low all the time. I am skeptical of that claim but don't have enough information
It seems intuitive that heat exchanger eff increases with decreased flow rate. Also air handling is always best done as slowly as possible, because you're trying to move air, not make noise and turbulence. There doesn't seem to be any intuitive reason to think eff would be higher at low rates unless its doing something insane like varying fuel/air ratios to get a desired heat flow or some kind of slow PWM with repeated ignition losses. Real furnaces don't work like that, so...
My otherwise boring furnace is variable speed and the longer its on, the faster it runs. It's really roaring after about 30 minutes. Its more than a decade old, this is pretty old COTS tech. My first zillion BTU cost me somewhat less than my last zillion BTU in any heating cycle.
I live in a cold northern climate with great insulation and low heating bills and a variable speed furnace and the house is basically never empty (spouse and I work at home often enough, etc). The whole "nest" idea doesn't work financially for me although it is interesting to watch.
That was me, with the throttling system. It's called modulation, and is very common in Europe.
While there is no momentum in heat transfer, the walls of your house store heat. If you're away long enough to allow your house to cool down it will take more fuel than it would to keep it at a reasonable temperature. Also, this is much, much more comfortable than to arrive in a stone cold house, only to open a window 30 minutes later because of the blazing heat.
I'm not saying it's always better to keep a certain temperature, though. It depends on the time that you're away, the insulation of your house, the outside temperature, the efficiency of your boiler, etcetera. But, having lived in houses with ordinary thermostats and non-modulating boilers and houses with programmable thermostats and modulating boilers I can tell you this: I'm not going back to the old stuff. The indoor climate is so much better now!
We have a thermostat in our house that has a weekly programme: it turns the heat on in the morning, then off at 7:30 when everyone is at work/school, back on at 16:30 so that the house is heated when we get back home from work/school, and off again at 22:30 around bedtime. During weekends it heats during the day and is off during the night.
There was never any question on using this system; everyone I know has this, unless they live in a really old house or an apartment block with common heating. It's a standard for heating systems called OpenTherm.
Is this really that uncommon, or is this a European thing?
We have similar in the UK.
I bought the Nest thermostat originally but could not see how to make it work for me.
I like the shineyness and am a big home automation fan (I have mostly LightwaveRF 433mhz sockets and switches and some hand rolled code connected to google calendars https://github.com/pauly/lightwaverf) but am going to wait a while before buying into this.
I'm in the same boat as you (Western US here). I've had a programmable digital thermostat for at least 10 years, and many newer apartments have them already installed.
I can't believe more people don't know about these things. They predate the Nest by 5-10 years.
People tend to waste energy when they're cold - it's the false savings of turning off the air-conditioner.
You think you can bear it - until you absolutely can't, and immediately turn it to the highest setting you can because damn it you want it warm now.
Then it's warm (or too hot) and you turn it off to cool down faster and...rinse repeat.
For most people it's far more efficient to turn the air conditioner on to a given temperature setting, so when they enter a room they don't have any immediate reaction to "make it warmer or colder". We are exceptionally bad at judging these things in the moment.
For summer, we're lucky enough to live an a near-coastal region where 90 degree days are usually followed by 60-degree nights with decent evening winds - we used the AC 3-4 times this summer instead of the whole house fan (to avoid pulling in neighbor's BBQ smoke), but for the rest the fan is adequate.
In Winter, we use space heaters (new ceramic ones that start outputting heat in 10 seconds that are kid safe to touch) in the bedrooms and a very low central heat.
Of course we have over-standard insulation and trees nearby - anyone in a top floor of an apartment complex will most certainly be using AC almost every summer day (but probably not shivering at all during winters).
> Is it so difficult for physical human body to endure some discomfort while the house warms up?
A person driving a Mercedes-Benz is probably not one to care about saving the half a dollar in energy consumption to pre-heat the living space. So, it's not difficult, but not not-difficult, either.
"Interesting" that they're avoiding supporting HomeKit. But given that the majority of Nest owners own iOS devices, it sure feels like a slap in the face.
Supporting HomeKit would allow Siri integration, but I suppose Google probably wants voice control to themselves.
It was Apple that chose to make HomeKit a nonstandard, proprietary protocol for the purpose of locking people into iOS. You simply cannot fault Google for failing to fall in line with Apple's play for the market.
Putting HomeKit behind MFi is a barrier, but hardly one that should put Nest off the platform. The majority of their installed user base uses iOS. I kind of expected my Nests to last me 10+ years, but now I'm left wondering if I'll be replacing them sooner. (It's not clear to me how Nest's API is any more standard than HomeKit.)
Of course, from a business perspective this all makes sense. But as a customer, it's disappointing.
It doesn't feel coincidental to me that Nest will support geofencing and native voice control on Android while doing the same with an iOS device is likely to require the Google app and a Google account. You might see that as a win against "nonstandard" proprietary protocols but I don't see any upside for me.
Choosing to install a Google app on an Apple device is a much lower barrier than there not even being an Apple app for a Google device.
I'm no lover of walled gardens but one that allows you entry without buying a $700 smartphone is the lesser of two evils.
I was discussing the cost of laser barcode scanners with someone the other day and the few models that are iOS certified are significantly more expensive due to MFi.
One requires you "buy entry" with a $700 smartphone. The other requires that you "buy entry" by giving access to your personal data to a company that has demonstrated that it doesn't care about your privacy. Neither option is objectively "less evil", it depends on the priorities of the user. I for one will only willingly hand Google the keys to my data when hell freezes over or Apple and Microsoft go bust.
I think there's a certain amount of histrionics here what with bringing up the whole 'evil' thing and your phrase "doesn't care about your privacy'.
I would have bought a more reasoned argument - getting into bed with Google demonstrably involves a tradeoff in terms of privacy - but I feel we're just a small step away from the days of 'Micro$oft' comments on Slashdot now.
I don't think "Google doesn't care about your privacy" is histrionics. Let me remind you of a few examples:
1) http://gizmodo.com/what-the-google-street-view-wi-fi-decisio...
"Google did this to enhance the accuracy and precision of its location based services. But it also captured "payload data," or the actual data transmitted through the Wi-Fi networks, including emails, usernames, passwords and more."
Google might have access to WiFi passwords used by every single Android user, a new report suggests. That is a whole lot of WiFi passwords -- maybe most of them in the world.
“If an Android device (phone or tablet) has ever logged on to a particular Wi-Fi network, then Google probably knows the Wi-Fi password,” Computer World's Michael Horowitz wrote last week. “Considering how many Android devices there are, it is likely that Google can access most Wi-Fi passwords worldwide.”
3) https://www.eff.org/deeplinks/2009/12/google-ceo-eric-schmid...
Yesterday, the web was buzzing with commentary about Google CEO Eric Schmidt's dangerous, dismissive response to concerns about search engine users' privacy. When asked during an interview for CNBC's recent "Inside the Mind of Google" special about whether users should be sharing information with Google as if it were a "trusted friend," Schmidt responded, "If you have something that you don't want anyone to know, maybe you shouldn't be doing it in the first place."
4) http://www.theguardian.com/technology/2013/mar/06/google-gla...
Google doesn't want to discuss these issues. "We are not making any comment," says a company spokesperson. But other sources suggest that Google's chiefs know that this is a live issue, and they're watching it develop. That's part of the plan behind the "Glass Explorer" scheme, which aims to get the devices into the hands – or rather, on to the faces – of ordinary people (and which enabled one member of the trial to putatively auction their Glass).
This is a long term pattern - I've got to the point where I just expect Google to not worry about my privacy as a default position, until such times as they get significant pushback, which may be too late for some users. Call that histrionic if you like, I call that learning from experience...
I wasn't arguing that Google has a whiter than white reputation on the privacy front and I'm aware of all the issues you listed.
I would argue however that they vary in severity and also in terms of intentionality. The most severe transgression is probably the wifi sniffing but there's no strong evidence that this was a deliberate policy decision at a high level.
The wifi password issue has got a fairly plausible technical/UX explanation - which you may or may not disagree with - and no evidence other than speculation that there is anything nefarious going on.
I could go on. I think reasonable people agree there have been some genuine abuses from Google but munging these together with speculation, insinuation and things Eric Schmidt should have thought a bit harder before saying is where I get a whiff of M$-style conspiracy thinking.
Huh? My claim was that Google doesn't care about privacy. I backed up that claim with a bunch of examples where Google rode roughshod over the privacy of individuals. There was no speculation or insinuation, there were just clearly presented examples.
What you seem to be missing is that all of those examples I gave point to a culture at Google where privacy is a secondary concern. Yes, things Schmidt said as CEO matter when you're talking about culture. The other examples are things that just would never happen at a company where privacy is taken seriously - you don't see Apple swiping passwords, or recording wi-fi data, when they could easily have done so.
Like I said - I agree with you that Google has crossed the line on more than one occasion but I still think some of your examples don't carry enough weight to warrant saying that Google is a lost cause with regard to privacy:
1. This one is pretty bad but my gut feeling is that it wasn't a high-level nefarious decision but a mid-level manager or engineer who went too far.
2. Devices needs to store the wifi keys in the clear, device settings are backed-up in the cloud. Storing them encrypted would require an extra password. There's probably a solution to this but are you suggesting Google has a plan to illegally login to private Wifi networks? That would be a criminal offence and definitely in tinfoil hat territory.
3. Yeah - I think that was a dickish comment and I wonder if he regretted it the minute it was out.
4. What did you expect them to say?
5. Dumb move but more a part of their desperate need to defend against Facebook. It's hard to see this as anything other an incompetent attempt to grow Buzz quickly.
Interestingly you missed out one of the stronger cases you could have made - the nymwars - which I think were an attempt to push the public/private boundary in their favour.
(Sigh. Replying to myself with a better formatted version because I can't edit the post and I never remember the arcane markup rules round here)
Like I said - I agree with you that Google has crossed the line on more than one occasion but I still think some of your examples don't carry enough weight to warrant saying that Google is a lost cause with regard to privacy:
1. This one is pretty bad but my gut feeling is that it wasn't a high-level nefarious decision but a mid-level manager or engineer who went too far.
2. Devices needs to store the wifi keys in the clear, device settings are backed-up in the cloud. Storing them encrypted would require an extra password. There's probably a solution to this but are you suggesting Google has a plan to illegally login to private Wifi networks? That would be a criminal offence and definitely in tinfoil hat territory.
3. Yeah - I think that was a dickish comment and I wonder if he regretted it the minute it was out.
4. What did you expect them to say? 5. Dumb move but more a part of their desperate need to defend against Facebook. It's hard to see this as anything other an incompetent attempt to grow Buzz quickly.
Interestingly you missed out one of the stronger cases you could have made - the nymwars - which I think were an attempt to push the public/private boundary in their favour.
Google cares deeply about your privacy. They desperately hope that the concerns of people like yourself never become mainstream - or that they never push their luck too far and the population at large becomes as cautious as some of the more tech-savvy already are.
Because almost their entire business strategy depends on people trusting them just enough.
You are a canary in the coal mine and I have no doubt that Google is very interested in your feelings about privacy for just that reason.
It's a pity Google acquired Nest. If not, it probably would have supported both Apple and Google's control solution.
HomeKit is mostly just an addressing system allowing the house to be described in config. It's not a huge loss for consumers if both Apple and Google have their own flavours of this.
Provided the product companies producing hardware remain independent rather than being swallowed up by Apple or Google.
I'd be equally annoyed if Apple started acquiring companies like Withings.
Rogers also sees the Nest Developer Program as "complementary" to Apple's recently announced HomeKit project. While HomeKit is going to allow users to control smart home devices with their iPhones and will supposedly make it easier to connect those devices to your network and to each other, Nest's API focuses on making those devices work in tandem once connected.
...
"The home shouldn't be a platform war," Rogers said. "We're going to be neutral. Switzerland."
I can see the benefit if everything was connected and it was thoughtless integration the whole way through. I can't help but think about what it will take to get your own personal Internet of Things hooked up. Interoperability and a set of standards is pretty much the only way this vision will meet mass consumer adoption. If that is the case its too early to tell if Nest will be the winner, but it is clear they have a head start.
The demo video may have revealed a little bit about Nest's strategy here. The video showed Whirlpool relying on the Nest thermostat to be the smart hub and help control the energy usage.
Nest seems to be selling the user experience right now (Thermostat, API etc) - but they are also positioning themselves to be at the center of the "smart home"
Really cool seeing a kickstarted project (LIFX) working better with others than Philips (Hue). It'll be really interesting seeing how LG and Samsung does.
An unsettling feeling that Home Automation™ is not actually about making the home and its consumer electronics more integrated, easier to use, and providing more free time and control to the individual. Rather, it is about making things harder to use for the average consumer without machine or professional assistance, requiring more hardware bundles and software services to get the "total experience", and control and data slowly taken away from the individual. Of course, this play for command and control is for the innocent sake of predictable behavior that can be optimized (and the steady revenue and growth that implies).
Self optimizing systems' predilection to predictability is quite predictable (whether machine or human).
While I believe Nest and HomeKit will likely be successful due to these optimizations they can potentially bring, I consciously will take the less optimal route this time, and have fun DIY'ing my automation instead.
I'm at least a little excited to see this. I've been reading about home automation for decades in the likes of Popular Science, and now, to perhaps see it really happen, finally. It makes me feel old, feel young.
The nest learns how long it takes to heat your house so if you wake up at roughly the same time every day it will start heating the house 30 minutes before you wake up.
e.g., you always wake up at 6am on mon-fri and it takes 30 mins to heat the house then it starts heating at 5:30am, but on saturday you sleep in until 9:30, and the sun has already raised your homes temp a few degrees and it only takes 25 then it starts it at 9:05am.
It'll be interesting to see how they plan to market this to people, because I have to tell you... "your thermostat integrates with your car, washer and jawbone bracelet" is damn confusing to me as a value proposition.
Yeah I see the potential, but even though I'm a programmer, I see a lot of random shit put together, and no clear reason why I should go through the effort of building a mental model around it in order to understand their vision.
HomeKit is just a set of APIs right now, so we have yet to see if it has legs, but it centers around a versatile device I always have with me - my smartphone. It makes far more sense to make smartphones the hub, so focus it around Android, not the thermostat...
The Google-Nest relationship is a bit toxic right now and people are all scared that their data is going to leak the Google way. But we all know it's going to happen, why delay the inevitable with this forced, confusing vision of the home?
I've seen such integrations fail before. Apple had those weird wireless iTunes / iPod / Hi-Fi integrations in the early 2000s and it never caught on, because it was damn confusing (and plugging the speaker audio cable in your iPod was way simpler).
This stuff needs to seem simple and inevitable, so people say "but of course I want that". And this is not it. It looks complex and arbitrary and their own ads videos say "people might say, why are we doing this". If you anticipate people might say that, you've failed.
I expect they will market the dishwasher as saving a bit of money when you have the thermostat. Similar with vehicles, it's a feature to check off when you are at the dealer.
Those home automation products are marketed to the people who have the money to buy such fancy home automation widgets.
Those people also typically want to "save the environment", but I doubt they want to save it so much, that shaving a fiver off their power bill means they get a thermostat tell them when to wash their dishes and do their laundry.
Putting the dishes and clothes in and out is still a manual activity, so this makes absolutely no sense to be scheduled by a thermostat, unless you have waaaay too much free time on your hands.
Right. I'm saying that I expect the marketing to be at the point of sale of the dishwasher and to be rather mild, so that it mostly targets people that already own a compatible thermostat. Or in press releases and the like.
I think home automation is mostly composed of things that sound neat and provide modest utility; Nest might pay itself off faster, but most buildings would see more benefit from increased insulation. Something like programmable lighting is fantastic, but it probably isn't going to save much money or energy, and switches just aren't that burdensome. I'm glad there are people trying to figure it all out, at some point the cost might get low enough where "neat" is enough justification.
Meh. Already possible with systems like TheThingSystem or Ninjablocks - there are rough edges, but there's a convergence of protocols and platforms taking place in the open source space.
I predict that these sort of things may be slicker initially, but open platforms focusing on one problem (ie: nodered or Huginn for rules, Owntracks for presence, TheThingSystem or Ninjablocks for pushing a bunch of readings to an MQTT hub from multiple protocols) will win out.
Right now I can drive both airconditioners through either a simple on/off 433mhz plug, or for the samsung, via wireless/direct commands.
I can trigger rules to disable when I leave the house.
I can control my lights when motion is sensed.
It's going to take a long time; but the open rules solutions already are approaching competitiveness with IFFFT or Zapier; and the NodeJS based lower level stuff is all focused on MQTT; so interoperability is highly likely.
The two announcements were obviously timed against each other. It's hard to tell who is reacting to whom, but my bet is that Quirky is hurriedly reacting to a pending Google I/O announcement this week, and getting the word out by giving the NYT an exclusive [1]. It also looks like Nest has an actual, planned microsite vs Quirky, who only has mention of GE on their website (https://www.quirky.com/ge) and doesn't have much of an official Wink page yet (https://www.quirky.com/wink is bare).
Add to that Apple's Homekit, announced at WWDC, partnering with Osram Sylvania, Skybell, TI, Chamberlain, August, Honeywell, Haier, Schlage, Philips, Black and Decker, Netatmo, Withings, and others. More brands/logos, but it's early.
Regardless, it's at least a three-horse race and we'll see it get worse before it gets better. All have serious brands on board and consumers will be confused for quite some time. A single, open, interoperable standard is coming but the race will encourage a flurry of product investments to get smarter devices in the initial stages.
[1] http://www.nytimes.com/2014/06/23/technology/quirky-hopes-wi...