Nest - The Learning Thermostat | Hacker News Reader

Nest - The Learning Thermostat

nest.com

430 pointserohead200 comments

This is interesting. I've been involved in home automation for about 15 years now, and HVAC controls are one of the top things to control in most systens.

Occupancy sensing is also extremely difficult to do passively in any kind of reliable scenario, especially when your occupancy sensor is limited to a single location in the house.

I've frequently seen two broad categories around regulating HVAC systems: 1) People keep a fairly regular schedule, and a standard 7-day programmable thermostat works sufficiently.

2) People keep a very non-regular schedule, and a standard programmable thermostat doesn't work, nor is there any inferable pattern to their home/away schedule.

In case #2, it is usually more beneficial to get home/away status from something with a more direct output eg: a burglar alarm system, vehicle presence detectors in the garage (photobeams), status of lights, etc.

I'll be very curious to see what the real-world reactions are to the Nest thermostat. If they can come up with some truly creative solutions it could be a really cool device.

Why not setup a system that detects the presence of your Bluetooth phone? Perhaps even look for your phone on the wireless network if a sensitive detector is not sufficient?

It would not be a stretch at that point to have a simple website served from the device that let you see the history and choose some basic night/day setback points.

It takes some time to raise/lower the temperature of the house depending on the outside temperature so learning the rate at which the house warms/cools is required to reach the desired set point at the requested time. Including the local weather forecast as a sensor and being able to see a graph of temps and system demand cycles would be really nice.

Why not simply have it get the data from an online API connected to something like Google Latitude or Apple's Find My Friend system? Then you could even have it make decisions when the users are much farther away, e.g. "he's 100 miles away, just keep the house above freezing" or "he's come back within 10 miles, return the house to comfortable levels."

I know some folks who have a heater in their winter cabin they can activate remotely (via text, I think) to turn on before they arrive for the weekend. An automated version of that would be even better.

I had something a while back... A little script thingy I found online that would lock/unlock my laptop (MBP) based on if it was paired via BT to my phone. The problem with BT occupancy sensing is that the range is kind of limited, so you need several sensors, and the pairing dance that goes with that, it's kind of a hassle.

I would be willing to carry something-er-other in my wallet which would alert my home automation system (which I don't have) to my presence/location in my home. I usually leave my wallet in our bedroom when sleeping, so it would still be accurate at night. My wife might be willing to wear a small ring or something like that, but it would be a harder sell.

The benefits to the phone, you already have it and keeping it charged and in general not forgetting it at home are more or less taken care of.

If I were considering options for additional units to purchase for determining occupancy (perhaps the kids do not have cellphones yet) then a small zigbee type mesh sensor would be ideal. This type of solution would give many options for those with zone based heating and cooling and would lend itself to even more pretty visualizations.

Edit: Getting a post right the first time from a phone is harder than it should be.

Bluetooth drains the battery too quicky. An app running in the background with gps pings would be better.

From what's on the website it looks like they're trying to learn largely from user feedback of the "I wanted it cooler/hotter" variety. Agree that this will be a problem in case #2, without good occupancy sensing, because the prediction problem is inherently impossible to do well with only time-of-day-and-week if there isn't much regularity.

I would guess the real target audience is people who fall into #1 but never get around to programming their thermostat, and maybe don't even have an accurate self-assessment of their home/away or sleep/wake patterns. In particularly warm or cold climates this might also include people who make wrong guesses about latency, e.g. how early before they return the heater needs to kick on for it to be comfortable by the time they get home. I could see it being useful in that case: there's a regular enough pattern to infer a reliable heating/cooling schedule, but the person doesn't want to, or isn't able to manually program that pattern.

On the website "Activity sensors Nest's activity sensors have a 150° wide-angle view so Nest knows when to set itself to Auto-Away."

At $249 though it needs to make some sort of sell that I can recoup more than those costs.

That's why I like my OmniStat with my OmniPro security/automation system, it can change temp based on security system state (away, night, etc.) And if you still wanted occupancy based learning, it can be done with your security interior motions and Omni programming.

Also the Haiku iPhone app for my Omni just added changing flags in the OmniPro based on iPhone location triggers, this would allow proximity tstat changes if you really wanted that.

http://homeauto.com/Products/Omnistat/Omnistat2Products.asp

I have an OmniProII system, but have scaled back what I am using it for. In many cases though it makes a very nice security, HVAC, automation and audio platform.

#2 today could be solved with smartphones that relay their location?

ZigBee wireless sensors can be built rather cheaply if the intelligent components are centralized on the primary controller.

It would be great if this also supported Z-Wave, Insteon or ZigBee. Z-Wave would probably be more advisable since ADT and a few others are jumping on the Z-Wave protocol.

It supports Zigbee ("Show detailed Specs" link at the bottom of http://www.nest.com/inside-and-out/index.html)

They have an iPhone app. If that can pull GPS coordinates then they can know where at least one person is.

Dear people trying to sell me stuff:

Please put something on your website that I can read. I do not want to watch a video, and that's all you have, so now I will never learn about your product.

Dear people trying to sell me stuff:

Please don't put a ream of text on your landing page that I have to read in order to learn about your product. I'm a visual person, and the web has not been a text-only medium for quite some time now. Something like a video, relevant pictures, and big bold text will do just fine thanks.

I'm not trying to play devil's advocate here. This is actually what I prefer, and I think they did a good job.

I think your zeal for snark has lead you astray.

I understand what you're saying--nobody wants to land on a stuffy wall of text. But in this case, there's nothing to find but the window dressing. I'm fine clicking on a "find out more" link that takes me to more detail. But that's missing here entirely.

No, I truly didn't have an issue with the usability of their landing page. There's a clear link to more information in the CTA area (which is a page with video and a breakdown of features). There's also a clear call to action "pre-order" button. Then 3 more links with more info above the fold.

Then if you end up with the living-with-nest page you get a slideshow, video link, clear links to more info, a step by step slick UI describing it, etc.

I'm pretty impressed with the balance of information and usability. If I had to nitpick one thing it would be that the "Meet the Nest Learning Thermostat" link could be a little more obvious (being next to a button hides the fact that it's a link).

I also hate sites where the only information is offered through a video and I would like to have readable information as close to the landing page as possible, preferably on it.

It turns out that Nest doesn't have it smack dab on the landing page, but you just need to make the logical (to me) choice of clicking on the "Meet the Nest Learning Thermostat" link and the information is there. Scroll down at your leisure and skim through the text and the illustrations. It's not even one of those annoying PowerPoint-like presentations; you just need to scroll down.

I liked it, a lot.

Yeah, I read it and want to buy one now. I also learned that I have a two stage heat pump.

This also applies to programming screencasts and slides which are becoming more and more prevalent. I guess it works for some people but when I look at code I like to skim and use ctrl-F...

This page should give you the gist: http://www.nest.com/living-with-nest/ just scroll down.

That should have been the landing page...

Also, when I do watch the video, at least give me a hint of how your product works and what it does. Don't bullshit me with things like "XXX never stops learning" or "XXX saves energy" during the whole video.

In this case it might be justified. The guy is trying to convince you to buy their design, really. How would they entice you into buying a thermostat with words? It would be really difficult.

and if you must have a video, then please at least provide subtitles that are written in non-computer translated English.

No kidding. I wonder if all those video only sites are the result of A/B tests? Then we are doomed...

If you cared to look for more than 5 seconds, the "Our Thermostat" link at the top of the page takes you to a nice page all about the Nest with just text/images.

http://www.nest.com/living-with-nest/index.html

In this particular case, at least, the video is short, informative, and well-made — a great way to introduce the product.

Yeah. I can't skim a video.

Actually you can, just scrobble through it quickly. I just did, worked great.

Pure genius! Just one thing, anyone upgrading from an old mercury switch thermostat will be left with a mercury waste disposal problem. Properly disposing of these can be a real hassle and some not-so-environmentally responsible customers are likely to just chuck them in the trash. Nest could raise the bar of their environmental responsibility another notch by including a postage-paid mailer and some bubble wrap to return your old thermostat to them for proper disposal.

Compare this to the WiFi thermostat that I previously thought was cool:

http://www.radiothermostat.com/filtrete/

In addition to being a touch screen it has hard buttons labeled: PROG, MENU, MODE, FAN, and SAVE ENERGY

It is a total disaster. When I went away for vacation this summer I wanted to turn off the daily program and hold at a max of 85. Instead I put it into winter hold mode and it maintained a chilly 62 for the entire week I was away. The WiFi link failed the first day I was away for some reason and the iPhone app happily reported the last temperature it had received as current so I had no hint there was any trouble.

I've had that particular problem with an old thermostat before. No wifi or anything, but when the batteries would run out, it would get stuck into AC mode, no matter the previous settings. Getting back from vacation was like a meat locker. It seems weird that the "failure" mode of these things seems to be make the plants think it's winter.

Any ui with a "mode" button is inevitably a disaster.

Meet my alarm clock, 12 years and going strong.

Perhaps I should upgrade to something with a touchscreen and wifi.

I've thrown away many watches and clocks with 'mode' buttons after losing its instruction manual.

I'm pretty sure my alarm clock's manual is on another continent.

Then again, it doesn't take me a manual to figure out the "mode" button cycles between showing seconds, day of the week, time in both current and configured secondary time zone, and time and date in secondary time zone. I'm sure it could be made much more user friendly, material efficient, durable, and affordable with a touchscreen.

Curious about how many other cool projects / companies can come out of that line of thought:

"Here's a plain everyday thing that everyone wants to ignore, let's make it a cherished item"

A learning toaster that can find the sweet spot between warm bread and coal. Works if I've put in frozen bread or bread from fridge or bread from the bread bin.

Somebody please make this! How hard could it be? No, forget about that -- I'd throw money at you. Heck, I'd work for you while throwing money at you while coding for you.

I love toast.

A toaster that includes some kind of simple camera to monitor how dark the toast is getting would be great - I'd buy one.

Toasters with a toast color sensor has been around for at least a couple of years if not longer. They just don't advertise them that way.

Thanks - finding one was easy once I knew they existed! :-)

http://www.amazon.co.uk/Philips-Kitchen-Home-Supplier-HD2580...

I remember seeing a guy design a toaster that had a component from a smoke detector in it which (i'm quoting form memory here) monitored the amount of carbon or something being released (???).

The problem, as I'm sure we're all aware, is that toasters are very slow when you start using them, but if you make three or four rounds of toast one ofter the other, each batch toasts quicker. This smoke detector method solved that problem....

Although I've not seen it on the market, so perhaps it didn't work. But still, a novel idea.

There are already "transparent" toasters with windows on their sides. This is low-tech, but I imagine works better and is cheaper than having a camera inside your toaster (an environment which is dark and hot).

Yes, the dark current on that camera would be substantial.

This has sort of existed since the '40s, using radiant heat from the bread: http://www.automaticbeyondbelief.org/

I haven't tested this, but you're probably have issues with e.g. dark rye.

What if you eat rye bread?

Dyson seems to be doing quite well at that game in the household appliances business.

Give me a tap that will expand that sweet spot between way too hot and way too cold.

They've been out for a while. Expensive, and I don't remember the name, but they are there. My shower has "scald control" so if someone flushes a toilet nearby the temperature doesn't change, but I don't think that's what you are after?

Search the usual suspects" Moen, Delta, Kohler, etc.

Here's one: http://bathroom-kitchen-faucets.com/hot-cold-automatic-senso...

Or one with an adjustable middle section.. Gears?

Just use a thermostat on your tap.

Are mixer taps insufficient?

Mixer taps are great, but you still have the problem that a small amount of movement is required to get the temperature just right.

I've used a couple of taps that actually have a nice, smooth gradient between cool and warm, though it would be fun to design a temperature-controlled tap that cranks open the hot water as the water heater tank is drained.

The Learning Toilet.

You laugh, but TOTO is a hugely successful corporation that's spent millions on toilet R&D. If this bathroom doesn't fall into "object of day to day use made into a cherished item" I don't know what does: http://www.totousa.com/Neorest/NeorestSuite.aspx

Swiss rival Geberit also make a badass robot toilet [1], and they invest huge amounts in everything from software to water-flow acoustics in the home.

[1] http://www.geberit.co.uk/web/appl/uk/wcmsuk.nsf/pages/prod-a...

Nearly shit myself when I saw the price tag on some of those! Looks nice, but damn. A $250 thermostat is fairly reasonable, but a $5k toilet is way outside the WAF (Wife Approval Factor).

I think most people get that as a line item on their $450,000 house.

I remember predictions written in the 1990's of toilets that would automatically provide early screening for cancer.

How much of their millions in R&D goes toward that?

Even TOTO's bare-bones low-volume-flush toilets are remarkably effective and well-designed. I replaced some very expensive Kohlers with TOTOs and was impressed.

A normal toilet is a problem for most men: it's either loud to use, or you get some sprayback. A solution to this problem would capture the entire toilet market.

Any water saving toilet solves this problem by giving you a huge ceramic area in the bin to aim for. At least in my experience.

A toilet that tells you if are dehydrated and need to drink more, that'd be neat. Or perhaps automatically checks if your kids are doing drugs (maybe that's too far).

I would think they would start pissing in the sink or going to the bathroom at restaurants instead. But it is an interesting idea.

You're right, it'd only work if they didn't know about it.

Ever seen a Japanese toilet?

Google has them installed in the Mountain View campus, and probably in their other locations as well.

No. Elaborate...

From personal experience... my grandparents had one in their house (I think it must be a decade old) that would wash and dry everything. You could control the temperature and water pressure. It also had temperature controls for the seat. I think there were more functions I didn't know how to use.

It is strange that we haven't made this as comfortable an experience as our modern technologies/engineering allow. It's the equivalent of us still using the same type of mattress they used in the 1920's. I sleep on a "sleep number" and it's WAY more comfortable. I should be able to take a deuce on a way more comfortable toilet. Heated seat, automated flushing and paper dissension, automated seat rising and collapsing. When you are done using it, it should revert to a "closed" or "sleep" mode. The flush should be a pleasant sound enhanced by audio of your choosing etc. It should be a personalized experience.

Takes shit from no one?

A toilet with a built-in urinalysis system.

Suffers from HGTV syndrome at $249.

It isn't going to change the world or revolutionize home automation it's just going to be another "subzero" or "commercial range" as seen on Househunters: Tahoe.

The issue with programmable thermostats is that they are not about temperature but comfort, and comfort is affected by what you have been doing and what you are wearing than a few degrees of setting on the thermostat.

It's the exceptions that make programmable thermostats problematic - the day you get home early from work and the house is 82 F and high humidity in July, or when you are home with the flu and the heat is off...neither is worth $10 a month of possible savings.

We have a non-programmable digital thermostat. It simply has a button with a "$" which bumps the temperature a few degrees toward energy savings based on a user setting.

[http://trane.com/Residential/Products/Thermostats/Non-Progra...]

The bulk of our energy savings since the install probably comes from the fact that the temperature is set digitally rather than with an analog resister - so we tend to adjust the temperature more conservatively and less often.

Yeah, $249 is way too much. I'd be pressed to spend more than $30 on a thermostat when the one I have works good enough. Programming my current one is like pulling teeth, but I only do it once a year or so (if that).

I keep it low and don't have A/C. If I'm hot I'll change into shorts, and if I'm cold I'll put on a sweater.

I'm impressed that Nest fixes and repaints the wall when you remove your old, large rectangular thermostat.

I wondered about that too, but I see that it does come with a bracket that covers the wall if you need it.

That plate looks way too small and square to cover up the holes left by most rectangular thermostats. Look at 0:12 in the intro video, you can see the mounting holes for the "old" device. Interesting that the holes are painted white and have no screws in them.

Some readers complained about the content of the website, I think they are missing the point.

Most people aren't HN minded, they are not really into how Nest learn about your Energy needs. They just want simple and short content that explain in basic words. Nice looking design will give them confidence that this thing works really well.

The Website and the product are very well designed, and this is something that is not quite common in such products. It's really well designed.

This page (http://www.nest.com/living-with-nest/index.html) explain in simple and basic sentences how the device will work out for you. Pretty simple, yet convincing.

Yes, it's missing the hacker part. But I doubt that people are interested in that. I actually wish that more businesses invest in such designs.

  27. Hardware/software hybrids. Most hackers find hardware
  projects alarming. You have to deal with messy, expensive
  physical stuff. ...
  There's a lot of low-hanging fruit in hardware; you can
  often do dramatically new things by making comparatively
  small tweaks to existing stuff.

While Nest doesn't fall in the young startup category, Next is still a great example of what you can do with hardware and software, where "the software plays a very visible role." The design speaks for itself, but the magic is in the software---meaning whatever tricks they pulled from AI, machine learning, control theory or other fields.

Of course, "small tweaks" might not be the best way to describe it from the hackers POV, but it makes sense if you're using this device.

Yes, please, more of these kinds of innovative products!

Edit: Ah, created by an ex-Apple guy behind iPod and iPhone. No wonder it looks so damn slick.

TC link: http://techcrunch.com/2011/10/24/ipod-godfather-tony-fadell-...

Ah, created by an ex-Apple guy behind iPod and iPhone. No wonder it looks so damn slick.

And costs $249.

Consumers have shown that they are willing to pay a premium for a well designed user experience. Problem is, Apple doesn't have much competition.

Worth it if it can save you a few dollars/month and make you more comfortable in the process.

HN link: http://news.ycombinator.com/item?id=3152924

If one wants to build stuff like that -- I wonder - how do you do the low-tech stuff (the hardware that is not electronics but the plastics/aluminium shell, glass, etc.)? What skills/tools are required to actually create that? How do you get started if you can't afford mass-production at the beginning?

Simple: you design your electronics to fit existing cases that can be purchased off the shelf. Here's one of my favorite vendors: http://www.polycase.com

There are many, many vendors of plastic and metal electronics enclosures. Many of them will customize them for you with printing, cutouts for displays, etc. I know Polycase will do so very cheaply.

Replying to my own post because I can't edit it.

If you have a specific application in mind, you can contact me if you like. I have used a bunch of enclosure types and I might be able to make a suggestion.

As long as 'comfort' includes monitoring humidity and adjusting temperature accordingly, I'm in. Did anyone notice if Nest does this?

Yep, at a minimum it claims to have a humidity meter.

Looks nifty. Some feature requests:

1. Should get feedback for actual energy consumption from the meter.

2. Should allow installation of extra sensors that are wirelessly connected to the main unit, in case the main unit is far from the couch/bed.

For 2. this is where the iphone/mobile/tablet apps come in I guess.

I was thinking it would be great to have an extra sensor or two that would help detecting the auto away.

Here's a quick list of some other folks taking a swing at the Internet-enabled thermostat space: EnergyHub, EcoFactor, Proliphix, EcoBee, Inthrma, Suntulit, and GE's "nucleus."

Nest clearly has some cool differentiating features like "time to temp" and some interesting learning / AI technology. If it delivers on claims of 20% energy savings and they can bring the price down (that must be the plan, I'm sure) I can imagine it being broadly adopted. The challenge with thermostats is there are a few different wiring standards, so you often do need a third party to help you, which introduces a pretty big barrier to adoption. They also seem to have a strategy to lower that barrier, which is cool.

As someone who's been in the energy space for over two years now, I'm excited that Nest is bringing some much needed consumer attention our way, and I'm excited to try Nest!

They have a "is your thermostat Nest compatible" guide and video that isnt too unfriendly. http://store.nest.com/#compatibility

Here's what I don't get -- why have the display to begin with? Why can't the whole thing be programmable via API/Web/Mobile App/Watch/Whatever.

Ideally that's what I'd like -- some kind of intelligent device that lives inside of my house that connects to my boiler and AC unit, as well as other major appliances if possible and provides an open API for people to build on top of, and can connect to my router.

If someone wants to make a thermostat that looks like Nest, then, it could be something that you just stick to the wall but requires no direct wiring. But it also means that I never need to buy a new thermostat: if I get tired of the Nest, I can switch it like I can any other app. Or just keep it controlled from my phone or tablet or the web.

"why have the display to begin with?"

My guess is that many consumers have replaced thermostats before and have found the process fairly painless. Or, maybe they understand the idea of replacing one dingus on the wall for another. The idea of cutting wires in one's basement is much scarier. Your suggestion is quite rational but probably hurts sales.

Wattvision is making the 'no display' bet, instead opting for a smartphone webapp, or our full website. Check us out at www.wattvision.com. We're just whole-house energy monitoring, right now.

I'm more impressed by the founding team (with KCB backing) than the product. Can this be the company that finally gives us the smart home we've been promised for decades? I can't imagine they intend to limit themselves to thermostats.

I don't think they plan to. From their website:

"At Nest, what matters to us are homes and the people inside them. We’re changing the way you think about your home and the way your home thinks."

I watched the video and was impressed by how well designed it is. If it operates as demoed it could save a great deal of electricity.

Also I wouldn't be surprised if, like the iPod, it didn't come with a manual because it's so easy to use.

Why, why US only? Please tell me, it can't be because everything on this device is hard-coded for Fahrenheit.

I would not worry too much about that. I think that they will be able to handle the Fahrenheit/Celsius conversion.

:) I wasn't serious about that bit, just wondering why.

The same thing that kills a lot of cool things - regulations.

It could have something to do with power supply. Different countries have different mains voltages. I think it's a lot higher here in the Britain than in the US, which could fry a device like this perhaps.

Or perhaps not. I'm just speculating.

The control wiring for thermostats in the US is almost a common standard (RH/RC/W/Y/G) with 24 VDC. What is it in the UK?

Usually two wires, often AC mains voltage Live and Load but some newer systems use low voltage. My own house has a thermostat that wirelessly communicates with a receiver near the gas combination boiler (water heating for hot water and radiators). The receiver simply closes an electrical contact to signal heating demand.

I wonder if they'll provide an API. Are any companies in this space taking a platform-oriented approach to their domestic gizmos? Someone needs to put my home on the command line.

This thing paired with Siri would be awesome.

"Siri, what's the temperature at home?"

"Home is currently at 65 degrees."

"Siri, set home to 75 degrees."

"Temperature set to 75 degrees, it will take 15 minutes to reach this temperature."

Computer, tell me a joke.

Knock knock.

Who's there?

Orange?

Orange who?

Orange you glad you bought a Power Macintosh?

Oh sorry, I was amused at how Apple is marketing early 90s technology as amazing in 2011.

Apple does well because it presents old technology in a way that is much easier for the masses to grok and use. I don't know why this offends geeks. I'm a geek and I love it when technology is presented in an easy to use way.

I hate TV remote controls, I know how to use them but I shouldn't have to have 5 of them and universal remotes out there are pretty much crap. I have one, but the Tivo remote is so much more intuitive for that interface that we've abandoned the universal all together. That's an example of an industry ripe for someone with the design and tech chops to come in and fix and then jrockway will be talking about how amused he is that fools are so excited by this old technology. :)

I don't care for the 'learning' aspect, although I'm sure it's great.

My boiler has 19 buttons and 3 dials, I just don't have the time to understand it. This I understand.

i've had an ecobee (http://ecobee.com/) thermostat for a few years and it's little more than a way to bump the heat up while i'm lying in bed. i work from my home so i can't do much with pre-programmed schedules, and i have a dog so i can't let the place get super hot or cold if i'm gone during the day.

when it first debuted, they tried to charge a monthly fee for the website (which is a key part of the product, though it can be adjusted from the display itself). my brother works for an hvac company and was having trouble selling any of them because of the monthly fee.

eventually they ditched the fee and the website is free for life, and now they have an iphone and android app to easily see and adjust settings.

i setup a hack that watched my wireless router to see if my android phone was connected, as a way to detect whether i was in the apartment or not and let the temperature slide more.

http://jcs.org/notaweblog/2011/08/30/an_ecobee_automation_ha...

  "i have a dog so i can't let the place get super
  hot or cold if i'm gone during the day."

You do realize that dogs used to live outside? I have a dog and he can play for hours outside in 20 degree or 85 degree day. You are fooling yourself if you think your dog can not handle indoor environmental temp swings from 50 - 85 degrees.

Do you always use your dog to justify things?

I like the design of the case. I wonder if there's any way to display custom data on the screen, and how they're powering it (apart from the rechargeable battery).

I've always said that in order to get home automation into more people's homes it'll have to be "hidden" in something they already buy, like a TV or in this case a thermostat. I'm doing an automation startup[0] that focuses on building more of the behind-the-scenes technology for early adopters and eventual integration into consumer products, but that technology eventually needs to be packaged into something more consumer-friendly.

It looks like their presence sensing is actually motion sensing (the tech specs list "activity" sensors). I'd like to combine this with a more detailed presence sensor like the Kinect or Xtion Pro for more detailed control of temperature, lighting, etc.

[0] Shameless plug: http://www.nitrogenlogic.com/

here's their white-paper PDF: http://assets-www.nest.com/white_papers/efficiency-simulatio...

"The auto-away feature is based on algorithms that interpret occupancy sensor data and provide a confidence determination of whether or not the occupants are away from the home. When the confidence level is high that occupants are away, the autoEaway feature makes a decision to override the existing schedule to save additional energy"

it has sensors for movement, temperature, humidity, and light.

I'm guessing this will have premium pricing, in which case, are the people who care about shaving a few $100 off their energy bill going to have the forsight, disposable income and knowledge of installation to make the purchase?

Cool looking product, but too many barriers to purchase for my liking.

Preorder price is 250. About 100 bucks more for professional installation. 25 extra for each additional terminal. It doesn't seem too bad ... I was expecting 500 after watching the video.

My biggest concern is installation. I live in a rental, so I will have to uninstall the Nest with me when I leave. They have a nifty video plus compatibility checking app. Will try it when I get home tonight.

I did not see the price. Looking at it now, I'd say it seems nice in relation to the iPhone price point, but remember what this is. It's just a nicer looking thermostat.

The learning functionality is next to useless for most as they're in and out during the day/evening, so it needs constant changing just like a regular thermostat regardless. Effectively, the patterns can't be learnt. My thermostat is beside my front door. I enter my home, I set it, then again before bed and in the morning again. There's not much to it and I wouldn't cede control for $250+$100. $50, yeah. Maybe even $100 for a cool looking, easy to install device. But not $250.

Sure ... I don't think this will be useful for everyone. I've wanted a product like this ever since I went for a 2 week vacation in the middle of winter with the heaters on. I actually had a thermostat at that time but it was too stupid-complicated. The ability to control the heating/cooling system wirelessly, and a UI that doesn't suck are the main selling features for me.

Regarding learning, I don't know if it will work or not. I am also skeptical of their claims. In fact, my worry is that the learning system can't be disabled, and it will start to do stupid things! Then again, it might work ... we'll find out soon :)

Why not just use a programmable thermostat? They are down to around $20 these days.

I had a programmable in my last property. Didn't really use it much due to my erratic schedule. If I was a 9-5 it would be super handy though.

Thermostat managed through a hosted web service? I'm sorry but this is idiotic. Looks neat though.

Hard to say. This is the sort of thing people have been talking about forever ("smart homes"), but maybe the timing is finally right.

Quick, does anyone use their iPad to "store recipes"?

Fuck yeah. It's called "googling for a recipe with it". I do this regularly.

Sorry, a simple digital thermometer with a digital push button up/down temperature select is more than sufficient, and cheaper.

Home automation, in general, has a similar feeling to all those weird devices you put in your microwave to make bacon. Buy a damn frying pan.

Sufficient, but clearly not optimal. Temperature regulation requires time-based cycles to minimize power use, a scenario that doesn't tend to be served very well by a manually-controlled system.

For example, it generally makes sense to reduce heating use an hour or two after bedtime, until an shortly before folks wake up - is someone going to wake up to do that?

There's a lot of room here for simple innovations that will save huge amounts of energy when spread out over thousands of homes. Think how much power is saved by motion sensitive lights at offices now.

Ok, I can buy that my perspective is slightly different on these matters. I see people who set their AC to 67 and their heat to 78 -- even with time regulation this isn't going to save energy.

IMO, Blankets and socks are a fairly universal solution, and that coupled with a few button presses saves an equal amount (maybe more[1]) of energy.

I think the main selling point is going to be geek lust and how pretty it is, however. The energy savings is a loose justification for the consumer to spend money on it.

[1] I wonder how much of the energy savings will be negated by people turning up the temperature in lieu of dealing with a little inconvenience and minor discomfort because they figure the t-stat will handle it.

One place it could help is with the problem that most houses have - multiple people playing with the thermostat. It could help find the compromise point where everyone's happy enough that they stop fiddling.

Could this have an API? Thinking about it, some neat things could be done. For example, might it be possible to infer from electricity use if somebody is home or not? Also all sorts of fancy remote control options could emerge (Twitter is a must :-)

Their iPhone app should have geofencing (if it doesn't yet)

If every member of the household has an iPhone, that is.

Pretty safe bet for households that buy this kind of thing.

I hope there's an override for auto-away. Imagine going away over Christmas and your pipes freeze because there has been no heating at all... Not a problem in California where this has been developed but a realistic problem in much of the world.

Yeah, there are some counter-intuitive things you need to think about with baseboard heat and very cold temperatures in New England. Normally it is just fine and dandy to let your fancy thermostat drop the temps to the high 50s at night except when it is 10 degrees outside - in that case you want to turn up the temperature so you don't freeze your pipes :) Sadly I speak from experience.

I suppose it simply keeps the temperature above freezing even when you're away, like most thermostats

That's not sufficient for many home designs, though; you may need to keep the interior temperature into the 40s or 50s to keep the pipes in the wall above freezing, depending on the outside temperature and your house design. One reason the default "away" temperature for many thermostats is actually pretty high, like 55 or 60.

True, and just "learning" your regular day-to-day habits would not necessarily inform the system about an absolute lower bound on temperature.

I think the thermostat control problem is harder than I might have given it credit for. Any time you have to learn in a nonstationary environment (http://mathworld.wolfram.com/NonstationaryTimeSeries.html), nasty surprises can happen. The future may not look like the past.

The manufacturer mentions that it has a link to the web and can consult weather forecasts. This should allow it to kick the heat on if the forecast calls for freezing weather and your not home.

The first section of their information page (click "Meet the Nest..." in the middle of the landing page) suggests that an absolute min/max temperature range is programmed when the thermostat is first installed.

They should consider selling accessory units that plug into the wall and monitor the temperature of a specific spot in the house, syncing with the master control via wifi (instead of only reading temperature/motion in the one place the thermostat is installed).

I'm thinking of a corner in our house that is always cold -- if it detects I'm there, crank up the heat so it is comfortable in that location. When I leave, do the same for the next spot, or revert back to the "master control" unit. I'm not sure how much money this would save, but it would certainly be a valuable feature I'd pay for, and it would knock the socks off what any normal thermostat could do.

Not sure why the focusing on the learning alogs or the utility of the thermostat - my guess is that company really isn't about the thermostat at all. You can almost hear the founders/investors excitedly throwing around the various buzz terms with this one: 'Big Data!' 'Data Exhaust!', 'DataScience!' 'Household Traffic Patterns!' 'Web cookie for the home!!!!' etc. Looks like a cynical play at trying to become a BigData company by collecting household energy consumption and traffic patterns and who knows what else. Doesn't seem like it provides enough value to trade off being spied on in your home.

It would be worth the cash since my parents could actually read the temperature. I do hope Apple adds Siri integration into apps, then my Mom could just tell it what temperature she wanted.

If this thermostat has activity sensors, how can it tell the difference between a pet and a person? Not that I advocate freezing pets while I'm not home, but they don't need a toasty 72.

I also wonder if it can tell the difference between a teenager and the person who actually pays the energy bill.

Ridiculous.

Depending on where you live you can save more money by simply using a good size fan blowing cold air into the house at night during the summer. We did just that this summer and did not have to use the air conditioner at all, even during days where daytime temperatures reached well over 100 degrees. On some nights we could get the house down to 65 degrees overnight.

The $250 fan paid for itself in less than a month! Total cost of running it 6 to 8 hours per day was less than $10 for the month.

This was an experiment to determine whether or not a whole house fan would be a worthwhile investment for next year. Obviously, it is.

Next on the list is converting the A/C and heating system to what's known as a "zone" system. Why heat or cool several thousand square feet of a home when you only need one or a few rooms to stay warm or cold?

Next would be what I would call "intelligent flow" management. This is of particular value if you live in a two story home. How is a single thermostat on a wall in the family room downstairs going to make decisions about what is best for heating or cooling the upstairs bedroom? Well, it can't.

What you can do is have sensors in every single room as well as a way to pump air to and from various locations. Cold air will accumulate downstairs. Hot air upstairs. When the downstairs is comfortable in the winter the upstairs is boiling hot. By using this "intelligent flow" concept the system would simply look for suitable sources of what it needs, say, hot air, and choose to use the least costly source before moving on to a more expensive solution.

If the upstairs has three really hot rooms and only one room is occupied, you can pump that hot air back downstairs rather than engaging the furnace. If the sun is keeping the solar collectors nice and hot, use that instead of burning fuel.

Following that is some combination of geothermal and solar assist. A simple $20 microprocessor board with some relays and a few sensors can easily manage a set of valves to select a heat source or sink that is more energy efficient than using a compressor or a furnace.

Granted, some of the above would cost far more than a $250 gadget to implement. However, it should be obvious that they would more than pay for themselves in relatively short order, particularly the first couple of options.

The greater point, I think, is that home energy efficiency is a much larger problem that cannot be addressed in an meaningful manner with a wifi-enabled gadget on a wall, no matter how cool it might look.

The good news is that, if we take a serious look at how we build our homes and change the approach the savings can be absolutely monumental.

> Depending on where you live you can save more money by simply using a good size fan blowing cold air into the house at night during the summer.

I have a whole-house fan (Western Massachusetts), and the "where you live" caveat is a biggie. It depends on sun exposure, air movement, humidity and local factors (like air flow through the home). We've found that running the fan at night keeps the house cool through mid-morning (on warm days with low humidity) but after that the house warms up very quickly and that's when the A/C is turned on.

I'm happy you had a good experience with a large fan but don't automatically assume that it's a proper solution for other applications.

All of these things can be automated. Since the Nest is connected via Wifi and Zigbee, a future firmware upgrade (or an external controller) could add support for a Zigbee-connected fan controller. If the outside temperature is in the direction of the target temperature, turn on the fan.

It sounds like you're in a desert-like area with wide temperature swings. The learning system when integrated with your zone system could figure out how cold you're willing to get at night to store cold air for the day, or even use unoccupied rooms to store very cold air to be released into the house throughout the day.

I agree that a thermostat alone isn't enough to save significant amounts of energy, but this does have the potential to get get some smart technology into more homes, where it can be expanded later to include everything you describe.

The "Nest" is simply the wrong approach. By itself it is nearly worthless (my not-so-humble opinion) for anyone but the most technologically incompetent thermostat users out there. And even then, you'd have to wonder. Where is the data on how many people can't or don't program their thermostats?

Now, some people might buy it just because it looks cool on the wall, and that's OK.

A real thermal management system for the home does not start with a high-design thermostat on the wall. You need multiple types of sensors in every room and multiple outdoor points.

You need to change the way airflow can be channeled around a home.

You might even need to change the paradigm and heat/cool objects that can provide thermal inertia (like the floor or walls) rather than just air.

You need to be able to selectively inject other heat sources or sinks into the system.

You need solar thermal panels and geothermal heat exchange system in the ground.

As far as control, a very simple embedded PC board running embedded Linux would be the "lazy" way to build it. It would actually be overkill, but there's an argument to be made for the COGS aspect of it.

Any number of relatively low cost embedded boards running very simple embedded software could monitor the sensor array, make decisions and activate valves, fans, etc. as needed.

I guess what I am saying is that you really can't fix the problem without engaging in a fundamental pivot from what we've been doing to thermally manage homes for the last, say, 100 years.

Having said that, it is a lot easier to sell a cool looking $250 gadget than asking someone to spend $5,000 to modify their home. Even though the gadget is very likely to not do much and the more extensive home modification is almost guaranteed to save the home owner thousands of dollars per year.

Talk about an opportunity for government stimulus!

> Having said that, it is a lot easier to sell a cool looking $250 gadget than asking someone to spend $5,000 to modify their home.

I think a little creativity in the payment model could help this problem a lot. The whole point of the system is to guarantee a decrease in the cost of heating and cooling, right? Doing some statistical analysis, it should be possible to estimate how much your customers will save. Tell them you'll pay their bills for them for the next X years. Charge them 90% of their current costs, pay the actual bills yourself, and keep the rest to pay for your system. After 3 years (or whatever statistically works out to be a comfortably-profitable threshold), they own the system free and clear and pay their bills themselves, again. Obviously they can pay off the system in full at any point, to allow for home sales and things.

> I think a little creativity in the payment model could help this problem a lot.

100% agreed.

> By itself it is nearly worthless (my not-so-humble opinion) for anyone but the most technologically incompetent thermostat users out there

Last year, I was living in a house with several people much, much fatter than myself. The living room was upstairs, where the heat collected. My studio was downstairs, partially buried in the ground, which sucked all the heat away. There were heated arguments over the thermostat settings.

Eventually these arguments resulted in a housemate trying to solve the problem by installing a "fancy" new thermostat downstairs - it was a programmable monstrosity with a completely unintuitive LCD display. To this day I am never sure if I ever actually managed to change the temperature it was trying to maintain, or merely dinked around with specifying a temperature that would have taken effect IF I'd put it into the program.

There was probably a manual but I really wanted to talk with the housemates as little as possible to find out where the hell it ended up being stored. I have written fricking 6805 assembly code, I am not technologically incompetent by any means, but I sure as hell couldn't be bothered to try to deduce this thing's operations from first principles.

(Arguably this means the thing did exactly what they hoped it would - make me quit bitching and put on a couple sweaters instead of maintaining a shirtsleeve environment comfortable for a skinny chick in a house largely full of 300lb folks. But I digress.)

> A real thermal management system for the home does not start with a high-design thermostat on the wall. You need multiple types of sensors in every room and multiple outdoor points.

Well, let's see: a Nest has an activity sensor, temprature, humidity, and an Internet connection, which means it can find out what the weather's like outside. Plant one of these in every room and you've got something starting to happen, if you're using local heaters rather than a central system. Admittedly you won't be doing this yet until they get the price down.

> As far as control, a very simple embedded PC board running embedded Linux would be the "lazy" way to build it. It would actually be overkill, but there's an argument to be made for the COGS aspect of it.

Care to lay wagers on what's inside the Nest? I'm betting it's exactly that. You may be willing to spend the time and money to get the thing going and write software for it. But there are a lot of people out there who are not. This is a really attractively simple version of that.

You're describing a level of commitment akin to "I compile my Linux kernels from scratch", this is for the "I open my Mac and it just works" crowd.

Let's see, for $250 you can spend $10 per room in sensors. Let's say you have 10 rooms. The remaining $150 can easily cover the cost of a fairly intelligent controller. That would be a better system with potential for far more intelligence, particularly if coupled with a simple zone system modification to the AC ducting (valves that can route air where it is needed and prevent it form going where it is not).

Something like that would have a far greater effect on the cost and efficiency of controlling an environment than a single expensive thermostat in one room.

That's what bugs me about these kinds of products. They never address the real problem. They are only there to find a niche that you can sell a gadget into with some pretext that may or may not exist, like the idea that thermostats are hard to program. OK then, try this one, it's cheaper:

http://www.smarthome.com/30429/Venstar-T5800-ColorTouch-Full...

The thermostat is not the problem. Homes (and buildings in general) are wasting energy because we are still approaching thermal management they way it was done 50 years ago. Here in the US it is not uncommon to cover the top of a commercial building with black tar (for waterproofing purposes). And then you wonder why it cost so much money to keep that building cool. These outdated ideas and methods need to change.

That's the problem.

Like I said above, you're interested in this at "I compile my own kernels from scratch" levels of commitment. The Nest is for "I open my Mac and it just works" levels of commitment.

You are quite correct in that the Nest is far from being a perfect solution. But you keep on neglecting to figure in the cost of all the time you spend learning about this stuff, installing it, and frobbing it. How many hours have you sunk into improving your home? What's your hourly rate when you're working? If you LIKE doing it then it's easy to discount that.

I'd like to improve my home's energy consumption too, but I'm not interested in spending days and days thinking hard about it, installing ugly boxes on the wall (the Nest looks a hell of a lot better than that box to my eyes, and this does matter to me because I'm an artist), and getting armpit-deep in ducting. Spending a little more money and not having to sweat all the details sounds like a great compromise to me; I might not get the gains you would, but having a smart system I can control from the Internet is still going to make some improvements!

Also, reading the Nest's pages, I got the impression that what they REALLY want it to be is one Nest in every room - the things will network with each other. They're just starting, small production runs are pricey. I would bet that they are eagerly awaiting the day when they can sell 3-packs or even 10-packs for what one costs now.

Ah, the whole house fan. I grew up in an area with 90-100 degree days/50 degree nights without AC, and my parents eventually got a whole house fan.

We had it installed in the ceiling in our hallway in a 4 bedroom house. I didn't care for it for a couple reasons:

It was dusty around where we lived. This means it was really good at sucking said dust into the house. Made everything get dirty faster.

Even when it was on at night, it didn't really seem to cool off the house much. Sure, if you sat in the wind path of your open windows, it was 50 degrees. If you weren't, it was still pretty warm. I'm sure it was really good at cooling our attic off though.

Part of me wonders if the fan was too large for the house to be effective.

Well, I am not doing an off the shelf whole house fan. Clearly some of the issues you mentioned are very real. Dust can be easily handled with filtering. My summer long experiment proved that you can cool a house and avoid using the AC unit for a whole summer. This is at least true where I live.

The next step is to come up with a plan for next year. Since a fan outside the downstairs kitchen window worked just fine, I don't have to rush to put in a commercial system. I can use all of next summer to experiment and find a better solution.

You can cool my house during the summer for about $10 per month. That's the baseline from which I'll judge all other solutions.

Darn. I have a dual-fuel system (heat pump for spring/fall with natural gas for winter), so I think I'm out-of-luck for now. My current thermostat has an exterior temperature sensor and switches between modes when the outside temp drops below a certain level. Heat pumps are energy efficient on mildly cold days but don't work very well in the dead of winter. FYI, I live in Ohio.

I can't tell from the site if they have done away with the whole cool/heat switch but it seems like they have, which is awesome.

It is, if they can truly intelligently control the modes.

In much of the US, we are in that time of year where it can be cold in the morning and hot in the afternoon. A common thermostat on full auto mode will be running the heat in the morning and the A/C in the afternoon. Most people opt to set their thermostat for one mode or the other, usually forgoing the A/C in the afternoon, knowing that it will soon cool back down naturally.

Full HVAC control is one of those things that is really hard to get right with just a mostly passive device like this.

I noticed that the Nest asks you for your zip code. Maybe it uses the Wi-Fi connection to pull weather information so it can make adjustments like this?

You're right, it does check the weather through the Wi-Fi connection:

"Nest uses its Wi-Fi connection to keep an eye on current weather conditions and forecasts so it can understand how the outside temperature affects your energy use."

There's a section in there that talks about switching it from heat to cool when the seasons change. I don't know if that's more of a suggestion to this device or not, but something tells me it's that same damn heat/cool switch.

Seems like a solution in search of a problem.

Well, probably most of us don't see their utility bills as a problem per se but wouldn't mind 50% reduction. And Nest claims to deliver that. Certainly that claim is to be taken with a grain of salt but I don't think that it's a bad idea.

EDIT: No they don't actually claim that, my bad.

Do they actually claim that? Their page claims that 50% of household energy usage is governed by the thermostat, but they are very careful to avoid any specific claims about how much Nest would reduce this usage.

http://www.nest.com/living-with-nest/index.html

$173 in savings in 12 months on an average annual energy bill of $2200.

12.7% in savings over 12 months.

That is correct. Pardon for referring it incorrectly.

Assuming the learning algorithm is perfect and always gives the desired temperature, then a saving of 50% means that your desired temperature pattern is 50% cheaper than the temperature pattern you would program with a regular thermostat. That could be true, or it could even be more expensive.

Also, it might be annoying to have a thermostat trying to second guess what temperature you want through a "learning" algorithm, rather than the deterministic temperature you have programmed or adjusted manually.

If the house is cooling down while you are away - don't you need more energy to warm it up again than you would use to just sustain the temperature? I'd say that an 'intelligent' thermostat has greater chance to balance this right than anyone could with a manual adjustments.

No, I don't think so. Go to the extreme and think about leaving for a week or a month - it is quite clearly more efficient to let it cool and then warm it back up again. On a shorter time scale, your house will lose heat at a faster rate when the temperature differential to the outside is higher, so letting the house cool will reduce the rate of energy loss over the time period the house is cooler. The amount of time you are able to leave the heating off while the house cools will compensate for the higher load when you need to heat it up again.

No, there is no "stiction" with thermal loss. It is most efficient to not add any energy (heating or cooling) when you are not at home. That is as long as you do not mind it being too hot or cold when you first get back home.

My guess is that most of the savings come from the auto-away feature that's apparently backed by some sensors (motion detection?). I don't doubt that auto-away could save energy as most people probably don't either care or remember to adjust their thermostats when leaving home.

I don't think that the auto-adjustment really is that much more annoying than programming a regular dumb thermostat. Of course the end result being annoying depends on the algorithm design more than anything else.

They sense three things: (1) light, (2) motion (close-range) (3) motion (wide field/longer range). Presumably if nobody turns on the lights at night or walks past in a while, it decides nobody is home and gradually lets the temperature drift closer to the "nobody is home" range. Then brings it right back up as soon as it senses activity.

It makes sense to start out selling this as just a thermostat, but I wonder how long it'll be before they add a "burglar alarm" module?

I don't know anything about their algorithm, but I highly doubt that it 'guesses' the desired temperature. It looks like it simply records the users preferences based on time of day and day of week, and then tries to mimic those preferences. It will be interesting to see if they also provide a web based interface that allows users to have more direct control over the algorithm (for example hard coding certain scheduled times for the thermostat to turn on or off)

Not a random guess but a prediction. They do provide a web interface.

"Second guess" is UK/US English expression meaning "to attempt to predict or anticipate", so I think you and the parent have the same understanding.

Disagree. Two problems:

1. High utility bills (any solution that saves you money is a solution to a problem)

2. Global warming

See also: everything critics of the iPad said the week after it was launched

Every time all winter my wife asks me to get out of bed to go downstairs to change the thermostat that's awkward to use and hard to see in the available light, you can bet I'm going to be thinking about how I could have a Nest and just adjust it from bed using my phone. Combine that with plausible claims of it saving us $100+ a year in energy bills, and this is a very appealing product, IMO.

I think the "problem" he's referring to is the learning aspect. Web based thermostats and home automation have been around for over 10 years.

I read the Wired article about this, and was loving it, and then: "Indeed, the Nest will not only do its learning thing in your home, but also report data to a web site."

Really? Does every goddamn thing have to report my habits out to the internet now? Wtf. (Too bad, because I love the idea. Maybe you can turn the spying off.)

Does anybody have any research to suggest that this would result in less energy usage with a boiler (baseboard water heating)?

From what I understand, frequent fluctuations in temp settings with a boiler result in more energy usage, as the boiler requires far more energy to go from 65° -> 68° as it does to maintain a steady 68°.

Thoughts?

Thermal transfer is proportional to the difference in temperature between the inside and outside of your house. For every bit of energy you put into maintaining the difference some energy is lost to the atmosphere. If your system is equipped with an aquastat that turns the circulator pump on only after the water reaches a certain temperature then it is possible to waste energy if the boiler runs for a short time and the water does not warm up enough to start the circulator.

For those using Chrome on a Mac to view this link, dont. The site only shows an egg and a testimonial.

This is way too cool. Wonder if the device can actually detect the number of people in the room and adjust the temperature for that room accordingly. Not sure how this will work in age old central heating systems which contain one thermostat for a 10,000 sq ft building.

This device looks great, but might not be usable in some municipalities. Some building codes mandate that a thermostat must be individually programmable for a different temperature for each of 7 days a week, and specify an energy-saving min-max range.

Interesting. Their pre-order system went down from heavy traffic this afternoon, and now I see a redirect to Best Buy for pre-orders. I wonder if a quick deal was struck between Nest and Best Buy to get the pre-orders back online quickly.

Wasn't one of John McCarthy's original AI papers about an intelligent thermostat?

I thought of this too when I heard about it:

> In ten or twenty years Minneapolis-Honeywell, which makes many thermostats today, may try to sell you a really fancy home temperature control system. It will know the preferences of temperature and humidity of each member of the family and can detect who is in the room. When several are in the room it makes what it considers a compromise adjustment taking account who has most recently had to suffer having the room climate different from what he prefers. Perhaps Honeywell discovers that these compromises should be modified according to a social rank formula devised by its psychologists and determined by patterns of speech loudness. The brochure describing how the thing works is rather lengthy and the real dope is in a rather technical appendix in small print.

http://www-formal.stanford.edu/jmc/little/little.html (1983)

Nitpicking: Does it annoy anyone else that the website always has a horizontal scrollbar? Also not sure what about the layout confused me, but I didn't realize that I could scroll down to find out information.

From what I can tell the "learning" phase is just an extended version of programming your thermostat. I'm curious how this is any different than a traditional thermostat?

As opposed to explicitly "programming" your thermostat, you just use it and it programs itself.

How much does this thing cost? The store seems to be down.

It looks like it's $249 per unit.

I love it, sign me up.

“First they ignore you, then they laugh at you, then they fight you, then you win.” - Gandhi

Reply on news.ycombinator.com