Google Nest, ARM, Samsung pull out Thread to strangle ZigBee Nest Labs, the connected thermostat upstart acquired by Google for $3.2bn, is leading the charge of tech companies launching a new wireless protocol for Internet of Things devices. The Google-owned Internet of Things biz has, in effect, turned up the heat on a standards war. The major difference between the new low-power …
How much of a "have to" is that these have to be either mains or battery powered? Is this something to do with electrical safety standards, building regulations or just the fact that manufacturers don't offer mainstream alternatives?
I ask simply as I'd thought of wiring the house with a secondary 12v dc supply in future to effectively UPS my alarm and a few other things (backup lighting etc) and it would be nice if I could power other bits like home automation from this kind of supply.
"You don’t want to find out that the fire alarm has lost its connection by having it fail to signal the fire brigade."
I am reminded of a story (from 30 years ago) of a factory that had a fire alarm system installed that would automatically phone the local fire station. However the station became a part time one and so one night when there was a fire, it called up the station and the system there replied with an answer phone message that the station was closed and to call the "Emergency Services". This exchange then repeated until the factory was raised to the ground.
Your story actually flags up the article's example as slightly bogus. If the "Thing" is important enough that I worry about it falling off the net even temporarily, then I'm going to insist on it phoning home to the router every so often. OTOH, if I'm more concerned about saving power, I don't mind if it falls off for a while and then sorts out its new address when it finally returns.
There are plenty of existing protocols for automation and fuck Google in the ass (with broken glass) for trying to add another. What we need is to decrease the number of protocols and standardize, not increase them. This "news" (or it's content) was also posted 4 hrs ago at PC World.
I have been in Automation for 24 years so my opinion has some validity.
"Thread" will not catch on for the reason that it is not a wireless mesh type STANDARD like Zigbee is. IE if one device can't communicate directly, it will patch through another that still can. There are hundreds of existing building automation products that use Zigbee RIGHT NOW.
All it will take is to try to use "Thread" in a older building with diamond mesh plaster lath and with no repeaters or mesh arrangement they will fail to work at ALL. Wifi does not travel well room to room in such an environment and neither will Thread. And for the record only the main interface to the internet will need an IP address. Everything else either has a MAC or a subnet. Nothing more is needed.
There are not any current products beside Googles that offer "Thread" so none of the many home automation companies out there will touch it. A complete offering of products such as Pulse Input, Digital Input, Analog Input and corresponding Output signal types is required. There are complete lighting, thermostat, intrusion alarm, fire alarm and BMS systems available right now. And that is only some of the certified stuff.
For Andrew Jones who says there are no home products that use Zigbee, try here. He's WRONG.
http://www.zigbee.org/Products/ByStandard/ZigBeeHomeAutomation.aspx
Thread is just a one trick pony created over a bunch of drinks by people who don't know anything about building automation systems and can't stand any competition so the morons created another protocol.
There are several standardized protocols that do a much better job at wireless mesh network communication, reliability, redundancy, battery operation, than ZigBee.
Protocols that are approved for deployment in Industrial Internet of Things, including sites such as oil rigs, chemical plants, nuclear plants, process monitoring control and automation, etc., must be good enough for home and building management systems ;-)
Just search for ISA100 Wireless which is 6LoWPAN/IPv6 enabled, or WirelessHART, or IPSO Alliance, or check this readily available products:
http://centerotech.com/products/smartobject (a complete IPv6 / 6LoWPAN stack for #IoT)
http://www.nivis.com/products
All it takes is for those big companies to pick one of the proven technologies and help taking it mainstream!
The subject is about home or building automation (IE inexpensive) , not Industrial automation.
The companies you link to have protocols and routers and IP Stacks but not home thermostats. They may be good enough but are they cheap enough? HART anything has expensive licensing costs. It's NOT cheap. I sold that technology for 24 years. AND anything designed for industrial use is unnecessarily complicated and unnecessarily feature laden compared to home use. Too expensive!
Zigbee and Zigbee Pro are good enough, inexpensive enough, OPEN enough (Try tro join the HART Foundation) and there are a huge number of supported HOME automation products. Big companies have already chosen Zigbee (ever hear of Schneider Electric?)
FFS, Who needs IPv6 for a smoke alarm at home? Nobody! One point of connection to the Internet is enough, one main IP is enough. A simple inexpensive home router solves many connectivity issues at reasonable cost. Less than a hundred dollars versus thousands for your Wireless Hart devices.
Every device in a system does not need it's own internet connection.
Yes,
I meant in the UK.
Example - a Google search for "Zigbee home automation uk" returns one shop on the first page which is Vesternet, who while they sell Zigbee stuff - make their ZWave stuff far more prominent. Now that I have actually found some Zigbee stuff I see that it is at least twice as expensive as ZWave and some products are as much as 3 times more expensive. I have a ZWave USB stick on my Domoticz server it works wonderfully but more importantly I can pick ZWave stuff in loads of places - including Amazon.
Zigbee seems to be more of a product that was designed for the professional world and is now trying to penetrate the domestic market, and at the prices I am seeing stuff - I doubt it's going to a foothold
This article should have been entitled 'Too late Google - Thread is Dead (for the UK at least)'
Some of you may have missed this but the £11bn roll out of Smart Meters may be a little bit relevant here as they will all connect to the 'Home Hub' using a Zigbee protocol. (Feel free to read the several thousand pages of technical specifications available on the DECC website like I had to over the last 5 years.)
This makes all other protocols largely irrelevant in the UK in the long term as all (well most as it will not include people with tin foil 'Radiation' defending hats) houses will have some type of Zigbee based home automation network. Manufacturers of goods that want to piggyback on this network and measure energy efficiency will have to use this protocol.
This monopolisation of the energy home networking standard has made DECC very nervous about how it promotes and talks about this element of the smart metering standards (as it doesn't want to get sued by proponents of other standards) but if you look hard enough you will find it.
So why Zigbee? It was felt by the majority involved in the smart metering project that it was the only standard that was mature enough to be reasonably secure (don't flame with security flaws - I know), could meet all the technical specifications of the smart metering project and had commercially available hardware at a reasonable price. (Lots of promises from other standards but 'no beef' on actual proven delivery.)
...interesting times!
1) Google
2) Connecting ANY Gadget to be accessed via Public Internet. You need at least a 2 layer model. A hardened secured SINGLE "firewall like device" is Internet connected. Then ONLY one device needs security updates and can have all kinds of intrusion detection. It's daft to try and include this in every light bulb and plug top. Possibly the Gadgets should in fact use a non-routeable network protocol.
Really I can't trust any gadget maker for Security or Privacy, especially not Google. I'm sure some suitable plugin SW even HW can be used with OpenWRT and similar projects.
"In Europe, however, where thermostats either have to be mains-powered – which means a more challenging 240v install – or battery-powered, it’s more of a problem. In this respect, wireless is less life-threatening"
Sure the wireless thermostat can talk to the internet, but how is it going to talk to the heating system? Your going to have to connect something to the heating system and your right back to 240v.
The real advantage of wireless is being able to put the thermostat in a good spot instead of were it was easy to run the wires.
Quote
The real advantage of wireless is being able to put the thermostat in a good spot instead of were it was easy to run the wires.
And then came the drive-by hackers that get their kicks out of messing up yours (and many others) heating system.
And then came Google with street view mk 2 and got a detailed list of your IOT Kit and started flogging it to just about ad agency on the planet.
What about WiFi channel contention?
WiFi is not the panacea that you might think it is.
Tom 35,
The benefit of remote connection to a 'Stat is not having comm wiring, you still should have it wired back to the Boiler, AC or Furnace for control. But all THAT wiring is low voltage, 24 VAC (or Less) usually, not 240 VAC. This low voltage is used every day on thermostats.
The 'Stat internet connection only allows remote adjustability from your phone or device.
There are existing Zigbee wireless temperature sensors that require NO connections, they provide signals to controllers with wireless communications that do the actual work of controlling the equipment.
Let's not even try to complicate this further unless you know what you're talking about.
There are some good reasons for a new protocol for home automation, as most of what we have is trying to adapt the needs of devices to it, rather than vice versa. Whether Thread goes far enough remains to be seen, but at least the folk at Nest come from an embedded device background which should give them a better perspective.
I do think it will kill ZigBee (http://www.nickhunn.com/is-google-and-nests-thread-a-zigbee-killer/), which will have some other ramifications, not least that we'll soon be deploying 50 million smart meters with an obsolete wireless standard that connects to nothing.
Every word of your comment is completely biased as you have an investment in your customers, blog and books. You are wrong about zigbee, wrong about any need for a "new" protocol for home automation, wrong about embedded devices (2 versus thousands) and wrong about smart meters connecting to an "obsolete" protocol.
There are no issues with Zigbee and it is used NOW by too many products to mention. No need to wait for Google who would screw up a wet dream not to mention create the largest data mine and security hole in history.
We need standardization not more new protocols.
The smart meters have Zigbee for compatability with hanheld equipment for reading and temp sensors so that the systems can be shut off for load shedding.
I live in a house with no hallway so several rooms have more than one doorway. It's old but funnily enough there are light switches next to each doorway that were put in when it was last rewired (years ago). The exception is the bedrooms which have a IR remote near the bed (no installation required).
I can't see light switches being the must-have feature that drives this.
There are several standardized protocols that do a much better job at wireless mesh network communication, reliability, redundancy, battery operation, than ZigBee.
Protocols that are approved for deployment in Industrial Internet of Things, including sites such as oil rigs, chemical plants, nuclear plants, process monitoring control and automation, etc., must be good enough for home and building management systems ;-)
Just search for ISA100 Wireless which is 6LoWPAN/IPv6 enabled, or WirelessHART, or IPSO Alliance, or check this readily available products:
http://centerotech.com/products/smartobject (a complete IPv6 / 6LoWPAN stack for #IoT)
http://www.nivis.com/products
All it takes is for those big companies to pick one of the proven technologies and help taking it mainstream!
I really don't get the IoT Home Automation thing - this isn't just a "it doesn't suit my lifestyle so it's rubbish" - I've gone through as many scenarios I can think of and I can't find a good argument for using it.
Unlike the small electric cars thing, where it is currently clearly unsuitable for long distance or rural drivers, but there's a good case for having one in a city, I just cannot think of a situation where this home automation is useful. There may possibly be some limited application in mobility restricted markets but other than that I'm struggling.
I don't see how controlling lights/heating/door locks(!?) remotely via an app offers anything significantly better than the current methods of control? With PIR light switches you can be pretty efficient, with a zoned heating system and a decent controller you can make your heating pretty good, and what is wrong with a physical key (or pin pad) as an entry system?
Honest question - have I missed something?
>I really don't get the IoT Home Automation thing
I have a certain bias in this area (covered by NDA) but even I can see a lot of it is just junk.
But there are some good things out there...
You don't want or need every switch in your house to be internet controlled of course but adding some intelligence to certain systems in your home is a good thing. If you have something in your home that relies on the weather or can be made more efficient by monitoring weather patterns then that makes a good candidate for being made into an internet thing. Those devices wouldn't need to send any data out and could be purely data consumers too.
Maybe you have something in your home that would turn into an "oh shit" moment if it went wrong and you didn't notice while you were at work. Maybe you have a sensitive tropical fish tank that the contents of which would die if the pump system stopped working for too long. With some simple IoT tech you could have those systems tell you when "oh shit" is about to happen when you're not around.
Again, I don't think anyone except people that have to have everything shiny will need every light switch in the house hosting it's own embedded system but I think most people have one or two systems in their home that could be made more efficient or safer with some intelligence built into it.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
