Jump to content


- - - - -

Thread - a new networking protocol to unite the IoT initiated by Nest, ARM, and Samsung and others.

Breaking News Appliances Embedded Automation Energy Management Sensors

Things seem to be heating up in the battle to bring the Internet of Things to fruition.  The latest such entry into this market may just be the game changer we've been looking for to start bringing all these different systems together.  We've seen plenty of other attempts at standardizing communication in our homes - but they've all gone their own way trying to create the "best" - and instead it seems we've ended up with too many choices.  the Thread Group hopes to change this.
Things seem to be heating up in the battle to bring the Internet of Things to fruition.  The latest such entry into this market may just be the game changer we've been looking for to start bringing all these different systems together.  We've seen plenty of other attempts at standardizing communication in our homes - but they've all gone their own way trying to create the "best" - and instead it seems we've ended up with too many choices.  The Thread Group hopes to change this.  Cue the old xkcd.com comic:

Posted Image
 
 
What makes this different is the list of backers this time around.  The founding group consists of ARM, Big Ass Fans, Freescale Semiconductor, Nest Labs, Inc, Samsung, Silicon Labs and Yale Security.  New members should getting accepted Q3 2014).  This is a not-for-profit group that's centered around working with developers and consumers and creating a certification program for their new mesh networking technology.
 
I'm not 100% up to speed yet, but the idea is that a new protocol called Thread is being developed as a new mesh networking protocol based on 802.15.4.  This means that, in theory, existing devices which use ZigBee / 6LoWPAN and possibly others will be able to migrate to Thread with just a firmware update.  In fact, Nest thermostats are now shipping with a version of Thread running on them.
 
This new protocol shows promise, as it's designed from the ground up to be a self-healing network that's secure, scalable, and still easy on batteries making it work for portable devices.  It's intended to bring together appliances, access control, climate control, energy management, lighting, safety, and security - and it's designed specifically for the home.  Up to 250 devices are supported on a single network and can handle multiple hops (may be too few devices for some of us).  Setup looks like it'll eliminate the need for a lot of technical knowledge, as it seem like you'll be able to enroll devices with just your smartphone by entering in or scanning a product code on the packaging.  This will help open the doors to households everywhere, leaving the biggest remaining hurdle likely to be getting the electrical components switched out (switches, outlets) and possibly wiring up the new thermostat; everything else seems like it'll enroll in seconds.
 
I'm certainly intrigued; I think this will be the biggest jump-start the IoT needs to finally start bringing the masses together so we can finally start seeing the world where all of our appliances and fixtures in the home communicate and work together to keep us informed and help reduce our energy costs by allowing everything in the home to work together.  Of course this likely will bring with it yet another dependance on "The cloud" - but being a single standard, hopefully there'll be options for the edge routers to keep things closed within our own homes as well.
 
Thoughts?  Comment below!
 
 
 
 


15 Comments

It's interesting seeing companies join multiple competing standards group.  Guess they're all hedging their bets.
 
Samsung - Zigbee, Thread
LG - Zigbee, AllJoyn
Arm - Zigbee, Thread
Freescale - Zigbee, Thead
Silicon Labs - Zigbee, Thread
Cisco - Zigbee, AllJoyn
Dlink - Zigbee, AllJoyn
CableLabs - Zigbee, AllJoyn
iControl - Zigbee, AllJoyn

Apple is the elephant in the room.  By this time next year, 800 million (plus) iOS users will be running iOS 8 and thus Apple's HomeKit.  There has been no information as to the 'native' protocol for HomeKit; just that "bridges" will be available to make devices running other protocols appear as native HomeKit devices.  We should know more in the fall when iOS 8 is released.

 

With Androids fragmentation, it will be years before they can achieve any similar potential user base.  The others on the list are bit players.

 

Craig

The phone OS doesn't seem that important to me; there'll be an app that talks to the bridge in the house via wifi. 

 

Automate - interestingly all those players you listed that support Zigbee and Thread - well the two are close enough that via firmware alone they'll be able to move to Thread; no hardware changes required, in theory.

The phone OS doesn't seem that important to me; there'll be an app that talks to the bridge in the house via wifi.  ...

 

Ahh, but it is.  HomeKit is tightly integrated with Siri so you can do things like:

 

"Close the garage"

"Lock the back door"

"Set the thermostat back 2 degrees"

 

And includes 'Action Sets' so that telling Siri "Night Night" causes all the above to happen.  Plus secure remote access from _outside_ the home.  

 

Much of the really interesting information is still under NDA via the "Made for iOS" program.

 

Craig

Ahh, but it is.  HomeKit is tightly integrated with Siri so you can do things like:
 
"Close the garage"
"Lock the back door"
"Set the thermostat back 2 degrees"
 
And includes 'Action Sets' so that telling Siri "Night Night" causes all the above to happen.  Plus secure remote access from _outside_ the home.  
 
Much of the really interesting information is still under NDA via the "Made for iOS" program.
 
Craig

All that is going to do is pander to the Apple fanbase. Anyone can write an app that has those features. Sure those products will move units, if that's what you're saying.

People's needs are so diverse that the real winner is going to be a fully open cross-platform protocol with no NDAs, membership fees, and forced certification. Companies are too worried about poorly implemented products compromising the overall user experience. Let the free market take care of that.

All that is going to do is pander to the Apple fanbase. Anyone can write an app that has those features. Sure those products will move units, if that's what you're saying.

People's needs are so diverse that the real winner is going to be a fully open cross-platform protocol with no NDAs, membership fees, and forced certification. Companies are too worried about poorly implemented products compromising the overall user experience. Let the free market take care of that.

 

You may not like Apple but they are in the best position to bring home automation to a mass market.  Look at the Thread Group.  Only Samsung has significant reach but they don't control the Android platform.  They've got to depend on Google for OS-level support and that relationship is deteriorating.  Plus with Android's fragmentation, it will be YEARS before a majority of devices are running a version that includes (hypothetical) support for a new protocol.

 

For iOS, 90% of devices are now running iOS 7 that was released last September.  So by this time next year, 90% of iOS devices will be running a version of iOS that includes HomeKit.  90% of 800 million is a very large potential market for device makers.  That's an immediate market of people that have shown they're willing to spend money.  It is Apple's partners that are making the devices and I think they've got an opportunity to participate in a profitable mass market right out of the gate.  I presume Apple will make money by selling more iOS devices and taking a 30% cut on sales through the App Store.

 

Oh, and Apple has credibility when it comes to security and privacy.  Mobile malware is virtually all on Android.  Google has violated their motto so much now that "Don't be evil" is now "Don't be Google"!!  Google's bottom line depends on extracting as much personal info from you as they possibly can.  Apple depends on selling you premium products.  I know which relationship I'd rather be in.

 

AFAICT, "Thread" is just a protocol; far from being products that everyday people will want to buy for their home.  

 

Craig

They are in the position to bring something of their choosing to a large user base. Maybe this is what some people are looking for, but really it's just more of the same. I'm not saying Thread will be any better, but at least may strengthen the cause of 6lowpan meshes via interoperability at the network layer.

Honestly there is not much hope for something as universal as TCPIP at the application level, but it should at least be open API.

I guess I have a different perspective.  "Thread" is about getting the manufacturers together to agree on a way to make all the smart devices talk to each other.  From there I envision options as far as what gateway to use - whether it's purely cloud-based or hopefully with options that are 100% offline talking to a device in our homes.  From there, I expect the gateway of our choosing to have associated apps that we'll use on our mobile devices regardless of which OS they run.

 

I sure hope that just using a Thread enabled device doesn't mean that it's reporting my every action to the cloud; and I hope that whatever Apple comes out with will be able to talk to Thread-enabled devices.

 

I use Apple products; I'm no "fanboy" but I prefer them over others because they're simple, intuitive, and reliable - and they have a nice ecosystem where they can work with cars automatically; they have Airplay which blows any concept for Bluetooth streaming out of the water, and they get the best apps first.  That said, I'm not investing in products (lightswitches, appliances, etc) that ONLY work with Apple - ever - given the choice. 

 

Maybe I don't get it - but I don't get how everyone is comparing HomeKit to Thread.  I see them as totally different products and honestly expect that the day will come when they're forced to play together - otherwise this will be another failed attempt to bring the IoT to mass market because we'll continue to have a divided customer base which will keep the appliance manufacturers from embracing either.

On first reading this posting, I thought the Thread Grouop must really be scraping the bottom of the barrel to announce that Big Ass Fans was one of the founding members.  However, after reading http://www.cepro.com...est_on_thread/,

I can see that it makes sense.  More specifically, if a ceiling fan could learn to turn itself on-and-off by learning my family's preferences (ala Nest), I can see how that would be a good thing.

NeverDie:  Thanks for that link, it has some good information.  The takeaways, for me, are that "Thread" is really Nest's "Weave" protocol with enough changes to bring in (a handful of) partners.  It is Samsung's chip division that is involved--not their consumer electronics folks.  So, at this point, the only potential consumer products are from Nest, Yale and Big Ass Fans.  Not exactly overwhelming.

 

Work2Play, I'm not comparing Thread to HomeKit...more saying that Thread is going to get buried.  Home Automation is currently a tiny niche market.  While the Thread protocol might be very nice, it is not going to change that status.  Apple, however, is going enable all those iOS devices to be a fun, sexy, home automation controller in the next few months.  If an iOS user wants to dip their toe in automation, they can buy a thermostat, or a door lock, or a lighting product through (I presume) Apple's physical or online stores.   

 

Under that scenario, Apple has the bits in place to move HA from niche to mass market.  Whatever protocol they use (and it hasn't been announced at this point) will become the de facto standard.  To reach back a long way, Postscript became the de facto standard for high end page description languages (PDL) after Apple included it in the LaserWriter printer.  It was proprietary and there were lots of other PDLs before, but it became an important standard.  Thread is likely to be part of the old niche market.

 

There are lots of ways Apple's Home Kit can fail (lousy, faulty or expensive products, major security breach, etc) but I think the ingredients are there to create a smart home market that doesn't now exist.  BTW, there is no indication that Apple is making any of the smart home accessories.   They'll sell iOS devices and take a cut on devices and software sold through Apple's retail channels. Whether or not the devices are solely locked-in to Apple's protocol is up to the manufacturer, I would think.

 

Craig

Apple, however, is going enable all those iOS devices to be a fun, sexy, home automation an overhyped remote controller in the next few months.

Fixed.

Whatever protocol they use (and it hasn't been announced at this point) will become the de facto standard.

They won't endorse anything beyond WiFi and Bluetooth. So any bridges or hubs can bake in Homekit as well as Thread and whatever else but development restrictions will only hurt each ones cause.

Personally this is a great start - iOT=cOT

 

Waiting on Intel and AMD (BIG data) and the choice of free internet  ...                    everywhere. 

 

 

 

 

350px-Internet_of_Things.jpg

 Anyone can write an app that has those features. Sure those products will move units, if that's what you're saying.

It's actually pretty difficult for someone to write an app that can be controlled by Siri.  iOS doesn't have a generic API to get Siri command in an app.  A deve could do his own voice control but it would only work if you had the app open.  (Devs have done some hack solutions where you ask Siri to send a text message to a server which will spawn a command to a home automation controller)

Apart from communication the implementation has to be improved. Distributed data and logic requires a new programming paradigm.

I guess Siemens/Bosch is still fighting for their HomeConnect. Here is some literature about the needed standardisation, it is a shame that we still  don't have control of our appliances via standardized interfaces, this should be mandatory in every ecolabeling standard.

http://www.internet-...Access_2013.pdf

I personally think that networked embedded wireless internet servers will constitute the new standard:

http://www.aliexpress.com/item/Factory-directly-wholesale-New-generation-WiFi-Socket-WiFi-Power-Socket-WiFi-Outlet-Plug-App-Remote-Control/1994739217.html