Thread is a mesh network to connect products around the home

If I were to bet, I'd say it's a way of guaranteeing they retain external access to in-home data. Just market it as ensuring interoperability and the apologists will be coming out of the woodwork.

duncan_bayne7• 12 years ago

Read again: to members. You will not be free to implement the technology.

mrbill• 12 years ago

So, this is what, ZigBee+? Any plans on integration or gateways for interoperation with existing ZWave/Zigbee networks?

atonse3• 12 years ago

Today I learned there's a fan company called "Big Ass Fans" - and they're a partner in this!

unchocked• 12 years ago

They're damn fine fans too, proper airfoils instead of mere tilted planes. Not to mention big-ass.

fixedd• 12 years ago

You should really see some of the fans they make, they earn their name: http://i.imgur.com/GxuVHDx.jpg

pizza• 12 years ago

Wow, those blades have got to be really stiff to not buckle. Nice small feat of engineering.

Zuph• 12 years ago

They've recently diversified into lighting, with "Big Ass Lights." The companies are organized under "Big Ass Solutions" http://www.bigasssolutions.com/

chimeracoder• 12 years ago

Time Warner has been creeping into home automation - I've been getting ads for their home automation service for "only" $40/month with my monthly bills for the last few months.

If TWC has their way I imagine this will go about as well as Google Wallet did, which was killed because all the mobile carriers wanted to promote their own version (formerly Isis), and had the power to do so (since they could prevent Google Wallet from working on their devices.

maxerickson• 12 years ago

AT&T has also been advertising home automation and security (on national television).

I think they don't care too much about the technology, they want to market the monitoring and connection services, which I guess means they will still choose whatever favors collecting those monthly fees.

matthewmacleod• 12 years ago

I am super-glad this is finally happening; I was getting close to starting it myself, given the rubbish options which are currently available for building a proper end-to-end home automation system.

However, I'm concerned about the openness of it. I'm like 99% certain that the only realistic way to get a system like this to work is a 100% open, unlicensed, freely-implementable protocol with absolutely no restrictions. It doesn't currently sound like that will be the case, which is going to be a problem.

Perhaps I'll be working on this after all…

bravo228• 12 years ago

It looks like a transport, which has been around for a long time. 6LowPan has certainly been.

From what I read, it doesn't give you things like send command X to dim lights. Zigbee HA already defines those but everyone violates them in some way.

johnvschmitt• 12 years ago

They didn't launch any protocol. They launched a site to get (paying) members to come up with a protocol. Big difference.

I'm not too hopeful yet. Maybe it's just their ".aspx" website that has a proprietary flavor to it, when we terribly need an open standard here.

We do need an open standard, so each inventor just needs to implement their idea, & not spend 90% of their time building the whole supporting architecture, and trying to train the customers in the differences in the setup. The wearables/IoT ecosystem will continue to flounder until the equivalent of W3C arises.

atonse3• 12 years ago

Err, saying that a site has a proprietary flavor just because it's hosted on Windows, is like saying this open standard is proprietary since the documentation was authored in MS Word.

jnbiche• 12 years ago

is like saying this open standard is proprietary since the documentation was authored in MS Word.

And yes, that gives off the same vibe. It's a vibe that says, "we're not really into open source". I say that in spite of Microsoft's recent -- and welcome -- forays into open source. Their technology is still used primarily among people and organizations who are not very welcoming to open source technology.

jamesbritt3• 12 years ago

The wearables/IoT ecosystem will continue to flounder until the equivalent of W3C arises.

The horse may have left the barn already. We have the IETF, and they have offered CoAP. (http://www.utwente.nl/ewi/dacs/Colloquium/archive/2010/collo...)

But the gold rush to claim a chunk of the IoT and wearables product space means a good many people and groups want to invent Yet Another Standard.

With so many out there (some good, some wonky) it wil be hard for any single organization to stand up and declare anything remotely like a de jure standard.

(For myself I'm working on software to at least bridge some of the more sensible ones.)

agentultra2• 12 years ago

And then there were N+1 competing protocols...

Bureaucracies and committees are good at developing protocol specifications. Instead of throwing spaghetti at the wall maybe those of us interested in pervasive network computing should get an ANSI committee (or IETF-like task force) together to develop the de-facto standard language.

I don't want to have to build out an API in front of my cluster that speaks every possible permutation of language from every device and sensor. As a hardware vendor I wouldn't want to lock myself out of an ecosystem because my device only speaks a particular language that only a portion of the cloud speaks (or may become obsolete in a few years).

It's nice that this stuff is happening... but a Cambrian explosion of protocols isn't going to make the IoT very robust, IMHO.

bravo228• 12 years ago

I prefer IETF CoAp standard. Look it up. Much better I think.

jamesbritt3• 12 years ago

Same here. Feels like a mix of OSC (Open Sound Control) and HTTP. As with OSC it runs over UDP, which is handy for low-pwoer devices and/or constrained bandwidth.

However, also as with OSC, you can use TCP for CoAP if need be.

jchrisa• 12 years ago

CoAP is very compelling, here is the spec: http://tools.ietf.org/html/rfc7252

Cool that it maps to HTTP / REST so you can proxy it to the web. All the tricks it gets from UDP + TURN ICE STUN are p2p friendly.

middleca• 12 years ago

+1 for CoAP - We use it!

bravo228• 12 years ago

Google is doing this so they can give you another reason to buy Nest -- it now acts as a bridge!

Yale is doing this because they have locks that already have Zigbee.

ARM is doing this to sell chips. They have their fingers in a million different protocols. I don't blame them.

Funny enough, there are very few things in home automation that actually need battery. Everything from light switches to light bulbs to smoke alarms and security systems are wired.

The better way would be to use BLE which has the same benefits of 802.15.4 stack. BLE 4.1 allows you to do mesh routing, like Zigbee. The biggest boon is that you can talk to the nodes directly with your phone.

wlesieutre4• 12 years ago

Light switches don't need to be wired either, we've just been doing them that way because it's the obvious thing when you're just turning a 120v circuit on and off.

When your lightbulbs become smart devices, they no longer want to be on an on/off circuit. In fact, it doesn't even matter if a group of them is on the same circuit or not. You just need a switch to talk to them wirelessly over Thread, ZigBee, Enocean, or similar. At that point, you don't have to run the circuits to your light switches either.

Enocean's switches are all powered by kinetic energy, but if you don't mind replacing all of your batteries every few years, that can work too.

alexchamberlain3• 12 years ago

From a safety perspective, it's probably better that they are wired, though I'd prefer them to be wired to Ethernet and powered by PoE, rather than with 120/240V.

ddeck• 12 years ago

Light switches don't need to be wired either

Indeed, and sometimes can't be. There are a couple of locations in my home where there's insufficient depth to mount a wired switch and I instead have MK astral (z-wave) low-profile battery powered switches.

bravo228• 12 years ago

You are right they don't need to be, but from a building code perspective you can't bury the controller behind a wall. You have to make them accessible behind some kind of a wall plate so for aesthetics reasons people will likely end up putting switches there anyways.

I think the better option is, as you said, have all switches be wireless, and somewhere like a basement and have a bunch of 'controllers' to tell it to switch on or off.

xexers• 12 years ago

But the software industry already uses the word "Thread" to describe something else. This will get confusing fast.

sangnoir• 12 years ago

The knitting industry also uses the word "Thread" to describe some other thing. This is going to be a nightmare, I tell you.

alexchamberlain3• 12 years ago

This is fantastic; it is just an IP based home automation protocol, with backing from some of the biggest players in the industry.

duncan_bayne7• 12 years ago

... and which can be used only by group members, under RAND-RF terms. It is not an open protocol.

atarkmani• 12 years ago

interesting development. just think about the impact of automation on the hotel industry. Samsung has created a smart room control solution for BLOC boutique hotels in the UK. i wonder if this protocol embedded into the fabric of their automation solution

chubot• 12 years ago

How likely is it that they will create an open source reference implementation (e.g. C code anybody can use)?

SEJeff• 12 years ago

This is basically ontop of zigbee. Perhaps zigbee will finally compete with the likes of stuff like insteon which is cheap and for the most part "just works TM".

pocketstar• 12 years ago

Looks cool! Though someone should have edited their website

"Extensive support for sleepy nodes operating for years using even on a single AA battery" - http://www.threadgroup.org/Technology.aspx

Why is there no easy way to report errors on websites these days?

paletoy3• 12 years ago

In my eyes, the ideal rotocol would be one that fits, but doesn't need any router at all. While not excatly that, lora , a long range(kilometers) low power/cost protocol, seems like the right choice. Just find the right model for personal deployment some open routers(similar to wifi) and let nodes seek those open nodes , and we all win.

But i'm probably missing something here, since i haven't heard of anybody doing this. I wonder what ?

bravo228• 12 years ago

Lora is like 10 bits per second... really slow.

paletoy3• 12 years ago

So do you see some IOT tech that could spread without needing a router in every home ?

JetSpiegel2• 12 years ago

I only need a bit to turn a light on, so we may be onto something here.

freyr4• 12 years ago

The Internet of Things hype hasn't hit HN hard yet, but it seems to be ramping up. Is anybody else skeptical?

The strongest push for IoT seems to be from the wireless hardware manufacturers (Cisco, Qualcomm, etc.) and of course, they have the most to benefit from putting a wireless chip in everything. It's a wonderful solution to their problem of "How do we sell more chips?"

Now Google and Apple are hopping on board. They also have something to gain; Apple sells more hardware, Google collects more information.

But from a consumer's perspective, it seems like a solution in search of problem, and I haven't yet seen a single truly compelling use case that will impact a large segment of the population (turning heating/AC or lights on and off automatically is cool, but not that cool for most people when initial cost is factored in). Can somebody illuminate this for me?

alttab2• 12 years ago

Right now its all buzzwords and vaporware. We are in 2004 talking about the "cloud."

However, its coming, much like the cloud has.

freyr4• 12 years ago

There's a critical difference. Although the cloud and the IoT both began with buzzwords and vaporware, the cloud had clear and compelling potential benefits for customers from the start.

For every "cloud," there are many overhyped technologies that went nowhere, despite industry's best efforts to thrust them upon us.

Someone1234• 12 years ago

A lot of people here are sceptical but I keep looking for products on Amazon I just assume would already exist but don't.

Three examples:

I was recently looking for a door chime ("door bell") and since I have WiFi already all over the house I wanted something that could piggy-back off of that to increase range but also maybe send the notification to my phone or similar. Doesn't exist. Only ye-olde style RF doorbells no different than what they sold 50 years ago.

I wanted a motion sensor that wasn't part of a security system, just a single stand alone motion detector which could send a text message ONLY during set hours (e.g. 11 pm to 6 am) when motion was detected. But nope, also doesn't exist (in fact there is a serious lack of variety in motion detectors, the closest I found was a Honeywell one which set off a doorbell when motion was detected, but cannot be shut off based on time of day).

I want to know when the clothing dryer is done. I often don't hear it. I have WiFi, but yet I cannot receive an update digitally yet. Why?

I cannot speak to everyone but for me personally I WANT this stuff. I'd happily buy this stuff. It doesn't exist. Where is it? Make it and I'll buy it.

bduerst2• 12 years ago

The Internet of Things won't be that useful until it starts tying in several external services. I imagine then there will be enough value added that it will go over the tipping point for consumer recognition.

I'm personally waiting for my refrigerator to detect my food inventory, and auto-order grocery delivery once a week to restock itself. But seeing how Safeway, Shopping Express, Amazon Fresh, etc. don't even have an API yet, it's going to be a long wait.

freyr4• 12 years ago

I'm personally waiting for my refrigerator to detect my food inventory, and auto-order grocery delivery once a week

I imagined this as well, and it would be pretty cool (although I actually enjoy grocery shopping).

As you say, it's still a long way off. In addition to Internet connectivity and an API, fruit/vegetables/meat and all other refrigerated items would need some kind of tracking label (RFID perhaps) so that the refrigerator could automatically track levels. Scheduling delivery of refrigerated food might also be an issue (I'm not sure how existing online grocery delivery services handle this), as I don't want my dairy items sitting on my hot doorstep all day waiting for me to get home from work.

This could be extended to pantry food items as well.

FroshKiller4• 12 years ago

Personally, I'm stoked. The first connected home device I got was an Internet-connected thermostat when I got my home HVAC replaced two months ago. I had no idea I was getting that in the package, and I love it. Looking forward to the day when I no longer have to curse myself for forgetting to start the dishwasher when I'm away from home.

Keep in mind that in terms of factoring in costs, yeah, that will be true at first. But it wasn't that long ago that you had to buy an aftermarket modem as part of the barrier to entry to getting online on your personal computer. In a few years, connectivity will be the new normal, and prices will adjust.

rsync• 12 years ago

"Looking forward to the day when I no longer have to curse myself for forgetting to start the dishwasher when I'm away from home."

Is that a tech prediction, or an optimistic view of your future mental health ?

freyr4• 12 years ago

Perhaps an Internet-connected thermostat is something I need to try to fully appreciate.

No doubt the costs will come down, and capabilities will increase. And as costs and capabilities increase, potential for innovation will increase.

But it would feel much more compelling if they had a strong application out of the gate (perhaps its networked thermostats). It would benefit from some initial momentum. Industry seems to be taking a "if we build it, they will come" approach, but this often doesn't work out in reality.

paletoy3• 12 years ago

But from a consumer's perspective, it seems like a solution in search of problem

I think the company to look at is Philips. They really understand how to create appealing connected things. For example this article[1] that talks about their controllable color lights, cooking assistant and coffee maker, all controlled through an app to great benefit.

And once something is connected, bluetooth becomes a limitation - say you want to choose coffee type at the living room with your guests ? or you want to watch the cooking assistant while being at the kids room ? Considering that the cost of full connection should be similar to bluetooth in the best case scenario , why settle ?

[1]http://gigaom.com/2014/04/04/lights-cookers-and-coffee-a-dee...

MrQuincle• 12 years ago

What's the problem with Bluetooth here?

No mesh networking? That has been in the pipeline for quite some time now, and will hit the market big time:

http://www.csr.com/news/pr/2014/csr-mesh

Disclaimer: I've been part of a Dutch consortium with companies that developed their own mesh standards and did not understand that you need to "board the smartphone train". This is why I'm skeptical w.r.t. anything that does not just interface with smartphone technology.

toby• 12 years ago

I have a few "home automation" items in my house -- lights, thermostats, outlets, switches, etc.

By far the most useful is the Z-Wave deadbolt on my front door. It means I never carry keys, I can give codes to friends or the contractor and get texted when they enter, I can set temporary codes for deliveries, etc. I never have to worry if I locked the door.

I was surprised at the hype around Lockitron, since these have existed for a while. The main different seems to be that people prefer WiFi (since they already have it) even though it's a poor protocol for low-power and low-bandwidth.

jamesbritt3• 12 years ago

The strongest push for IoT seems to be from the wireless hardware manufacturers

Another area seems to be from those offering yet another (non-standardized) protocol or format such that they can then sell you tools and consulting tailored to their software.

[I]t seems like a solution in search of problem ...

Indeed. Most of the demos I've seen (e.g. turn on a light using Twitter) are easily doable in a number of ways, so the requirement for some IoT "solution" is dubious.

But suppose we grant that such demos are mainly done because they are easy to comprehend, easy to show in a bite-size video clip. Still, there still needs to be something that extrapolates to a more significant use case. I'm not so sure these currently exist for most homes.

Part of me thinks people are tagging "IoT" onto anything they can because it sounds sexy. Controlling my living room lights from my phone is not an example of the Internet of Things.

Controlling fertilization for a farm based on smart sensors out in the fields? That I can see.

Actual use-cases in the average home are less obvious, as is why people will want it.

bandushrew2• 12 years ago

I believe that the main reason it is hard to see the benefits is that we dont yet have the interesting infrastructure inside that house that an IoT will allow.

ie, I think it is actually too early in the IoT life cycle to really visualise the impact it will have.

Let me explain - currently our lighting and applicances are designed entirely around the idea that they will be manually operated and standalone pieces of equipment.

Once that assumption is taken away, and houses are built with a different assumption, suddenly magic begins to happen - imagine that instead of a few lights in the lounge, there was a ceiling and walls covered in thousands of LED lights that could be independently controlled via the internet. The average home could literally have lighting shows. An entire wall could be utilised as alternately a television or a photo viewer, even a wall in every room.

etc, etc. Use your imagination here. The point is that the impact of an IoT on the average middle class house as they are built today is currently pretty small because neither the house nor the contents of the house were designed with an IoT in mind. Once we actually have an IoT houses will be build from the ground up with different assumptions .

At least, that is my opinion.

tl;dr: The IoT has a chicken and egg problem.

x1798DE3• 12 years ago

I still don't see a compelling argument here. Who cares about having a "light show" in their house? What does that even have to do with the internet? We already have huge televisions that aren't very expensive. If we want the whole wall to be a TV, we can get a projector. A wall covered in individually-addressable LEDs is probably not going to be cheaper than a wall covered in televisions or just a big projector.

The question the parent was asking was "Why hasn't anyone shown me a concept that makes me think, 'Now I want an internet of things in MY house!'?" Not, "Why haven't I seen quick adoption of these IoT concepts out there?"

_pmf_• 12 years ago

suddenly magic begins to happen

The magic of a tenfold increase in maintenance costs.

dmritard96• 12 years ago

Energy efficiency. IoT can bring more awareness to things for finer grain control which can take systems developed over 60 years ago and make them more aware and targeted.

Shameless plug: http://www.flair.zone We are aiming at lowering HVAC energy consumption by between 20 and 30%. Financial incentives are rather attractive ones.

I will agree that there are going to be IoT whizbangs that are solutions in search of a problem, but there are bound to be a handful of actual problems that get addressed.

eximius• 12 years ago

Why is there no information for individual developers?

ausjke• 12 years ago

as a developer who dealt with zigbee, it sucks. z-wave is great, but not open enough and not IP-friendly. 6lowpan is a logical choice for IT companies, which is what Thread about here. I hope it becomes more popular over time.

Thread is a mesh network to connect products around the home

Related Stories