Yeah, that '50bn IoT devices by 2020' claim is a load of dog toffee The Internet of Things is mostly a hype bubble, with real-world spending and deployments being just a fraction of their predicted level, according to a report by analysts IDTechEx. While spending on IoT runs to "billions yearly", the cost of buying and installing actual IoT networks "is much more modest, contrary to heroic …
Keep Calm and Firewall them All.
One thing we can expect is a massive pressure to move either towards fine grained CPE management or to CPE virtualization or both in residential over the next 5 years. If Joe Average Luser does not firewall them all, the providers will firewall them themselves. That is the only way they can tackle the gigantic outgoing attack volumes.
It is quite funny - In one of my previous lives I was a tech lead on a vCPE/next gen CPE management project in a Tier 1 networking vendor. I kept trying to budget and allocate for IoT use cases. The management above me kept telling me - not interesting, no use case, no need, you are not allowed to reserve two heads for that, put them on something "useful".
I wish I was still working in that company to take that "not interesting, no use case, no need" and shovel it... Somewhere... Where the VP, DE and Directors heads were till recently. Usually referred to in colloquial English as "where sun does not shine".
Anon. For obvious reasons.
> Who's buying all this IoT stuff?
Everyone in any manufacturing related industry.
This isn't consumer stuff. This is a sensor in your local waste treatment facility telling some tech when the filters need to be changed. This is a pump at your local brewery phoning out for a tech to come onsite when a pump is going to fail.
Consumer IoT is out there as a concept to help pay for the cost, but most IoT stuff isn't aimed at the consumer.
50Bn isn't unreasonable.
Consider Smart meters (electric/ water) traffic management systems, etc, etc.
It's not about how many consumer devices are connected to the internet (e.g. phones, chromecasts, fridges, etc.), it's about anything that needs an IP address... that doesn't limit it to physical devices or objects either!
All that I manged to retain from this article were exactly those two words.
They held my interest for far longer than the "surprise shock horror " that someone made a prediction that IT continues to evolve ....
Dog Toffee , I wonder how much the market for it will be in 2027 ?
Perhaps delaying IPv6 might delay the inevitable DDOS maelstrom that takes down the internet.
Frankly any IoT device that isn't properly secured should be banned from ever being sold let alone connected to a system that has connection to the internet.
I'd even go further and jail the CEO of any companies found to be peddling this crap in the first place.
mind you it might be hard given that most of the builders are in China and possibly working for the Chinese version of GCHQ.
If IoT is ever going to be big more than likely in the next few years they are going have to make the internet free for the public at least for the IoT devices. People are too lazy to put much more than their phone, computers, and printers (and maybe TVs) on their home internet. Sure they will probably get taxpayer money to pay for it while they reap the benefits monetizing our privacy.
"The report predicts the IoT as a whole (including software and system integration efforts) will be worth $322bn by 2027, exploding from $1bn next year to $73bn by 2018. The number of devices will, they say, barely break one billion by 2020."
So the market will be worth $73bn the year after next (and presumably similar if not more for the two years following that), yet there'll only be 1bn devices by 2020? 73 * 3 = 210, in other words $210 per device... isn't this obvious bollocks? Or does this headline figure count the entire cost of the device, including the non-IoT parts of it (this would be quite a lot on the average connected car, for example)... in which case it's still bollocks.
The truth is, no-one knows, but the marketeers and analysts are having a field day regardless. Wake me up when any of this crap does something useful without being a security nightmare...
> So the market will be worth $73bn the year after next (and presumably similar if not more for the two years following that), yet there'll only be 1bn devices by 2020? 73 * 3 = 210, in other words $210 per device... isn't this obvious bollocks?
You think they're just putting a bunch of random devices out there and not doing anything with the data or software?
"You think they're just putting a bunch of random devices out there and not doing anything with the data or software?"
Actually, pretty much. They want to "monetise" it, but from the results, they could do better with a random number generator.
Ex: My local grocery has Loyalty cards that help them identify me to what I purchase, and I benefit from a slight discount.
Every purchase generates scads of coupons allegedly tailored to my shopping habits.
Out of the years of shopping, the mounds of coupons have never been for anything I buy.
End result, mild privacy invasion to no benefit to them, and a slight benefit to me.
Random chance would be a better guide.
What? Someone would actually pay for a palm-sized computer...
Really, really doesn't need that. Assuming an enabled 'fridge, all you need is RF tags on the goods, which is already being done in some cases. Add RF readers to a range of other things - if the 'fridge, why not your kitchen cupboards? Why not your wardrobe? Why not the bath? Why not your wheelie bin? Many of these things already have power nearby (cupboard lighting for example) so it's do-able without much inconvenience to the "customer".
M.
Putting sensors on trackors is a good idea, there is some sense, some logic and quite a large number of farmers.
Why i need my toaster/kettel/fridge connected to the internet still illudes me. I had a smart 3D tv for a while (generous landlord) but it just wasn't worth plugging it into the internet. We plugged HMDI cables from laptops to watch torrented content, but that does not need a smart tv but provided said content in 1080p HD with no lag. The internet connected to stupid stuff doesn't make that stuff smart, it just makes it connected to the internet. Again why does my fridge / kettle/ toaster / light bulbs / door locks etc need to go online?
I honestly expect to see people with a cat 5 (or 6 now) cable up their butt anytime soon.
TBH as they dont see able to connect my remote control to my sat box and my sat box has bog all in the way of buttons because the remote control is supposed to do it I suspect they will crap out big time in trying to do anything else.
Demos of using a mobile phone to control your lights leave me cold and frankly the software I see spewing out these days is ghastly beyond belief, iphone, android, microsoft, and yes everyones beloved linux is reams of crap that 99% of the time fails at the simplest of tasks because the designer/architect/engineer are out competing each other to be complex and use buzz words rather than to be simple, accurate and correct.
And make a combine leave the cornfield through a fence and drive down the middle of a highway, forcing oncoming cars take the ditch to avoid getting shredded?
Because there's no way they don't have cellular modems on more recent models, and like cars wouldn't separate that stuff from the CAN bus (or whatever the farm machinery equivalent is) Those things can already pretty much drive themselves as it is, so I don't doubt this is possible.
And make a combine leave the cornfield through a fence and drive down the middle of a highway, forcing oncoming cars take the ditch to avoid getting shredded?
Because there's no way they don't have cellular modems on more recent models ...
That's OK then. No chance of getting a mobile signal in the middle of the countryside.
Of course IoT security is a joke, that's partly because nobody has sat down and said "Of course IoT security is a joke". There are as many protocols racing ahead as there are devices. If they opted into a standard, any sensible standard would contain a pathway to update firmware and a basic security model so that my kettle isn't a peer of my laptop.
Standards would also mean being able to buy best of breed, so a Philips light bulb and a LG telly and a Samsung fridge and a GoPro/whoever CCTV and..., and... And that would mean a market for third party control interfaces and apps instead of the current buggers muddle. That's the dream all this IoT speculation relies upon. But as with every good standards war, everybody loses because consumers don't trust that they'll get long term support, so don't invest in case they back the wrong dog. And without consumer investment, manufacturers see no market.
There's nothing inherently difficult about IoT and making it good, just this isn't the place to start from. Instead, form a IoT Alliance, get some basic standards and patent pool out there and get an "IoT Ready" logo out there, and promote it across the board. Make it IP6 only, make it able to sign up to any IoT-nominated (partitioned) route to the internet securely, manage the virtual networks centrally so my light bulb isn't watching my online banking traffic...
I can't be the only one who thinks that doing a professional job of this is worthwhile? It's worked out ok for WiFi standardisation, or Bluetooth etc.
Wake up people real IoT does not have a IPv4 or IPv6 stack.
LoRaWAN IoT devices don't have USB or any way to deliver large data packets.
WAKE UP FOOLS!
Read LoRaWAN spec's they say 1-255 bytes per 100 seconds.
Only fools build devices that can attack us all.
Only fools think IoT can attack us.
What your talking about is called IoT personnel, it's the peer to peer IoT solution. i.e NFC to Phone
Real IoT uses long range modems like LoRa from Semtech. This is peer to gateway. STUPID!
So many fools in charge, to many idiots make comments!
We are all F**ked, with fools at the table.
VP's & CEO are so ignorant they don't care it's all about THE $ not public security.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
