I have to admit that I don’t get this internet of things

David Cameron has announced that the world is on the brink of “a new industrial revolution,” where the internet will allow everyday objects like fridges to “talk to each other.”

Speaking at a trade fair in Hanover, Germany, the Prime Minister said that “the internet of things” would transform the world, as he announced a package of measures to promote Britain’s position in the global technological race.

“I see the Internet of Things as a huge transformative development – a way of boosting productivity, of keeping us healthier, making transport more efficient, reducing energy needs, tackling climate change,” said Mr Cameron, adding that the world was now “on fast-forward”.

Yes, I get the point of connectivity. Smart metering of electricity, this sounds like a pretty good idea (although who controls it is another issue). Telling the washing machine only to go on when the wind’s blowing equally. But why does everyone talk about fridges communicating with each other?

What, exactly, does anyone think a fridge has to say to another one? “Turned out chilly again” is going to old pretty quickly isn’t it?

And I will guarantee that the first thing that will happen to fridges that are on the internet is that they will get hacked. And everyone’s food is going to be either hot and rotten or frozen as the kiddies giggle in glee.

104 thoughts on “I have to admit that I don’t get this internet of things”

  1. So Much For Subtlety

    And I will guarantee that the first thing that will happen to fridges that are on the internet is that they will get hacked. And everyone’s food is going to be either hot and rotten or frozen as the kiddies giggle in glee.

    Cameron is a fool. He knows nothing and is merely repeating what every other idiot is saying – and what his officials are writing for him. But listen closely to what he says:

    “I see the Internet of Things as a huge transformative development – a way of boosting productivity, of keeping us healthier…”

    Healthier? The first thing your fridge is going to do is report you to the Government for not eating properly. Well before anyone gets around to hacking it, it is going to be lecturing you on the need to buy less beer and more greens. That is the point.

    And he knows it. He is not telling us openly, but he knows it.

    Liberal Britain, where the State stopped at your front gate, is long dead. If John Stuart Mill came back from the dead he would be speaking a language as dead as Sanskrit.

  2. Is it not more likely that your fridge will talk to one of the large online supermarkets and begin to create an order for you based on what is already in your fridge? RFID tags on food packaging would let this happen – fridge/cupboards keep track of what’s inside.

  3. Your fridge will talk to Ocado (or similar).

    Or update you in the supermarket as to whether you are low on creme fraiche.

  4. Working on it. Working on it
    Thanks to Raspberry Pi, I can now turn the lights on from anywhere in the world.
    Proof of principal, anyway.
    s long as the bloody Spanish internet hasn’t dropped out again.

  5. One application did come to mind.
    Motion detection could let me know if someone was sniffing around the place in the mountains.
    Let me check them out through the camera.
    And give them a round of buckshot if I didn’t like the look of them. Targeting solution performed in software.
    Cool.

  6. It’s not about fridges, it’s about everything. You can already buy network attached light bulbs which can be individually controlled, same goes for smoke detectors. It’s about lighting, cars, heating, security – everything you can think of and it’s happening fast.

  7. The Laughing Cavalier

    Yes, very useful that your fridge can order more beer or butter but I fail to see how that is going to have a significant positive impact on GDP.

  8. dc96,

    “Is it not more likely that your fridge will talk to one of the large online supermarkets and begin to create an order for you based on what is already in your fridge? RFID tags on food packaging would let this happen – fridge/cupboards keep track of what’s inside.”

    And how’s it going to know that the 6 pint carton of milk is full or nearly empty?

  9. It’s the usual utopian futurism, in which people think that certain awesome things will be awesome. See also, autogyro to work, personal nuclear reactors, meal in a pill, etc.

    In this case, it’s the erroneous belief that algorithmic manipulation of metadata can produce artificial intelligence. See also Tim Berners Lee and his “semantic web” pipe dream.

    Our own Tim is right. Your fridge has nothing useful to say.

  10. steve,

    “It’s about lighting, cars, heating, security – everything you can think of and it’s happening fast.”

    Does anyone want a lot of that? If I want a light turning on, it’s because I’m walking into a room. Heating? I have a thermostat. It does the job.

  11. I shall be kept busy ensuring all such devices are neatly lobotomised and unable to communicate anything to anybody.

    Also Tim, smart meters are a very BAD idea. They only need them so their green windwankery won’t collapse the national grid–and once you are dumb enough to accept them you have allowed them to install a control mechanism in your home. Now–if they want you to freeze in the dark because you won’t pay extra taxes/levys to fund the state/banks failure (or won’t submit to an id card or whatever) they can only cut off whole areas. With smart meters your house alone freezes in the dark. Piss on smart meters.

  12. I’ve been in IT for 35 years and I see the pace of change constantly increasing. This stuff is happening now and there are indeed enormous opportunities; why, do you imagine, Google just bought Nest?

  13. I dunno, why did Google launch Buzz, Wave, Plus, Chromeboxes and all their other shitty big ideas that nobody wants?

  14. So Much For Subtlety

    bloke in spain – “One application did come to mind. Motion detection could let me know if someone was sniffing around the place in the mountains.”

    You have a place in the mountains? The Japanese are a little ahead of you here. A while ago they built an electric kettle with a pager attached. Every time you turned it on, it sent a message. The idea was that now all the children are 40 and working in Tokyo, they can keep track of whether their Mother is dead or not by checking how often she makes herself a nice cup of tea. Of course it would depend on the intruder making himself a a hot drink. But I like to think they would, wouldn’t they?

    “Let me check them out through the camera. And give them a round of buckshot if I didn’t like the look of them. Targeting solution performed in software. Cool.”

    The Thais may be ahead of you here because someone did build what was essentially a security camera, some servos, and a .45 magnum. If they did not like the look of an intruder you could blow him away. As for targeting, TrackingPoint is building what you need – for “hunters”:

    http://www.gizmag.com/trackingpoint-precision-guided-firearms-scopes-digital/25264/

    Of course it will cause some jurisdictional problems. If you are sitting in Florida and shoot someone in Andalucia, does American or Spanish law apply?

  15. steve,

    “I’ve been in IT for 35 years and I see the pace of change constantly increasing. This stuff is happening now and there are indeed enormous opportunities; why, do you imagine, Google just bought Nest?”

    Money burning a hole in their pocket? Trying to grab the people who created it to bring them into Google?

    Because of all the things in my home, the one thing I really couldn’t care much about improving is the thermostat/boiler timer. The thermostat does its job – regulating temperature. The boiler timer switches on the heat for when it gets colder at night. If I think it’s going to be colder, I might manually hit the button. I don’t care to spend hundreds of pounds and to add complexity to my heating system to save me a couple of times having to get up off the sofa.

  16. Tim you’re absolute correct about security issues but security on the Internet is hardly an unsolvable problem.

    As I understand it, the idea behind the Internet of Things is not just that “things” connect to the Internet as is. Then you’d be right the fridge would have little to say. The point is that with fridges connecting to the ‘net they can then be further instrumented in ways that can bring big efficiency improvements. For example, internal sensors can monitor moving parts and provide early alerts that parts need replacing. Imagine having the three weeks’ notice that your fridge is going to need a repair or replacing rather than coming home one evening to find that it’s broken.

    Moreover, if instrumented then your fridge can make continual minor adjustments to the way it works to ensure it is operating as efficiently as possible. Sure it won’t save you a huge amount of money but over all fridges we could be reducing our wasted energy by a significant amount.

  17. I’d like my fridge and washing machine to communicate. The motors draw peak kW on start up, so I’d like them not to start at the same time, thus economising on the fixed price part of my elec bill.

  18. I’m very much with Steve on this.
    One gets so used to hypertexted annotation on webpages its getting slightly annoying the real world doesn’t supply the same facility. With the aid of something similar to Googleglass, embedded RFID tags & interconnection the whole thing would be so much more comprehensible

  19. Good point BiF.
    I’d imagine, like me, you’re paying for the amperage of your grid connection. Being able to smooth demand would enable getting by on a mush cheaper gridlink.

  20. How long after this technological breakthrough before the first fridge is arrested for browsing terrorist-related material under a pseudonym?

  21. bloke in spain
    March 10, 2014 at 8:52 am

    Working on it. Working on it
    Thanks to Raspberry Pi, I can now turn the lights on from anywhere in the world.
    =================================
    Which means that someone else can turn them off whenever they feel like it.

  22. Fridge to social services – you had better come round, quickly, they’re letting the kids eat too much ice cream!

    Smart meters as per Ecks above.

    The key to all this is “who controls it”.

    Security video to your smartphone, and similar stuff, you can already control 100% yourself.

    Smart meters are not compulsory. If they ever are made compulsory, I shall be investigating firewalls for them.

  23. I’d like my washing machine to remind me when it has finished. My cooker to remind me to take food out. A food hopper let me feed the dog remotely. To be stood in the supermarket and find out how much cheese I have in the fridge. To turn the lights on at home when I find I’m going to be home late

    Of course the ultimate internet-of-things is going to be driverless cars, both in terms of being where you want when you need it, and communicating with each other to find the best routes and maintain constant speeds.

  24. Huge security problem. We can’t even secure routers and surveillance cameras properly. And there’s almost zero incentive for a fridge manufacturer to bother to fix something you’ve already bought (see buggy old Blu-Ray players, for example).

    Anyway, home automation has been promised since the ’60s and never materialises just as video phone calls never do. Not because it can’t be done but because no-one wants it (we’ve had video capable 3G phones for 15 years and who ever bothers with that?).

  25. Having been into nerdy electronica since the early 80s (yes, I used to buy the Maplin catalogue in the good old days) I can never understand the constant fascination with automating the turning off and on of lights.

    I’ve worked in large commercial properties where this makes enormous sense. But they can already do that. Why everyone thinks that utopia will arrive when they can do the same with their home, I cannot understand. But they always do. Back in the days of ZX81s and VIC20s, you’d hear this same justification of, “and with an interface, it can turn lights on and off”.

    Seriously people, you do not need this and, when you get it, you will not use it. It’s like that juicer you get for christmas. One week of use, then, box in the attic.

    PS how does your fridge measure the quantity of cheese within it? Tiny robot walking around the shelves pacing out the volumes of the comestibles? What?

  26. The most interesting thing about reading those predictions from the 50s about how we’d be living in the year 2000 is how unbelievably wrong they got everything. Some things just don’t change – like the basic setup of a car, which has barely changed in the fundamentals since the thing was invented a hundred years ago (wing mirrors are wing mirrors). And who remembers the paperless office predictions? Our printers run red-hot all day, this in an era when we supposedly don’t need them any more.

    Most of this is a solution in search of a problem. I remember they struggled for years to try to find an application for the internet in cars, which nobody wanted or needed. Even now, in a brand new BMW with all the communications options, you find the connectivity limited to the GPS (which is good) and all this useless crap such as being able to Tweet from your car. Just because the internet exists and is good, doesn’t mean it needs to be plugged into everything.

  27. Bloke in Spain – you and steve are crazy.

    Have you SEEN that film ‘Demon Seed’, where a guy turns his house into an internet of things, and the first thing his computer does is rape Julie Christie?

    On a slightly less rapey but still creepy note, in the film ‘Electric Dreams’ a guy connects all his home appliances to a 1984 IBM PC and it uses them to try to kill him and steal his girlfriend. And it was probably planning to rape her later.

    My wife and I both agree we don’t want to be raped by computers. Even the sexy ones from ‘A.I.’

  28. Anthony said “The point is that with fridges connecting to the ‘net they can then be further instrumented in ways that can bring big efficiency improvements. For example, internal sensors can monitor moving parts and provide early alerts that parts need replacing. ”

    That fridge would not require internet connectivity. It could be done internally in a similar fashion to a car engine management computer.

    While I am sure there are benefits to home automation (noticing the fridge is drawing more power than usual could give a clue that it needs looking at for example) the benefits of putting all that on the internet is less clear to me. The most obvious benefit of smart appliances lays in the hands of the government and energy suppliers to remotely disconnect you, know precisely how much to charge you or to manage demand at times of shortages.

  29. Can you go to prison if your refrigerator sends out threatening emails?

    I have already heard of a case where mass email spam was traced to a refrigerator.

    Let’s hope the refrigerators aren’t running Windows XP, we’ll all die. Seriously, as we are seeing with cars, the addition of IT content will limit useful life. The oven in my house is dead, because a replacement ECU is no longer available. If your car is over 10 years old, it’s likely that the ECU is no longer made.

  30. “Some things just don’t change – like the basic setup of a car, which has barely changed in the fundamentals since the thing was invented a hundred years ago ”
    Uh? I’m thinking on my own wheels. Full media centre, fone & internet connection. 240AC power outlets. Coffee maker & microwave. I could live in it. Do on trips. 1000km a day, GPS navigated one finger driving to a location I’ve already eyeballed through Streetview. Model-T it ain’t

  31. As Gareth says, what is teh benefit of the internet telling you your fridge is going to fail? Red light on the front would do that.

    But then, there’s the cost of the monitoring gubbins to predict these occasional failures and the unreliability of doing so. It’s not like that thing in science fiction where the warp core announces it’s going to fail in precisely 37 minutes 3 seconds.

  32. @Gareth

    Technically you’re correct for that specific example. But then your fridge would need to include at least a set of indicator lights for all the moving parts and the amount of information you could get from that is small. If it is connected to the Internet then it can release much more information.

    @Bob Saget

    Of course security is a major issue. But you shouldn’t be relying on your fridge to handle security. Unless you plan on directly hooking up your fridge to your ISP you will probably handle security on your local router. Traffic from your fridge need not ever leave your house. People are confusing the Internet with the World Wide Web, it seems.

    The point is that in order to connect your devices with the outside world it is better for them to use the TCP/IP stack and connect at the Internet layer than use only Link Layer connections.

    If you prefer – think of this as the “Network of Things” with the option to connect your local NoT to the Internet.

  33. @Gamecock
    “If your car is over 10 years old, it’s likely that the ECU is no longer made.”
    I’m pondering this because my bus is getting on. But that clunky manufacturer’s ECU isn’t necessary. My garage pal can manage it all on a laptop, by-passes through the diagnosis port. The processing power required’s trivial. Twenty-five quid’s worth of Raspberry Pi’d blow an ECU away. It’s purely a software hack.

  34. Anthony=

    if there were really some worthwhile “due to fail” tech, a single “failing” LED or beeper would suffice. Hardly any fridge owners are fridge engineers, so you’d just have one “call an engineer” light.

    But there is no use to this at all. You wait until it fails, then call an engineer.

  35. @steve 12:21 pm

    I don’t recall ever saying that i agreed with any of this, merely that it is happening; and it is happening, if you want to believe otherwise that’s fine.

  36. Uh? I’m thinking on my own wheels. Full media centre, fone & internet connection. 240AC power outlets. Coffee maker & microwave. I could live in it. Do on trips. 1000km a day, GPS navigated one finger driving to a location I’ve already eyeballed through Streetview.

    This is all just dressing up, nice-to-haves. The important bits – engine block, crankshaft, pistons, valves, flywheel, differential gears, steering wheel, four wheels (one at each corner), etc. are fundamentally the same, i.e. there has been no step-change in how the whole thing works (like, say, the difference between a propeller and jet aircraft). There have been refinements galore, but other than the automatic gearbox (and maybe the move to direct fuel injection) I’m struggling to think of a fundamental design development in cars. The one which amazes me is the retention of the wing and main mirrors: 100 years of breathtaking technological advancement and a shiny piece of glass bolted to the side is still the optimum solution.

  37. What Ian B said.

    The touted advantages of this bollocks seem to me at best not things I would pay for or could be achieved by means less likely to lead to some cunt in the civil service switching things on and off for me.

    As Tim implies, hacking is endemic in everything connected to the web, and so as far as I am concerned the less of my stuff that is, the happier I shall be.

    At work nowadays one lives immersed in the internet. For me home is a refuge.

  38. Steve – yeah, right.

    How do I know you’re not IBM’s Watson? Right now you could be looking at pictures of Julie Christie with nefarious intent while stroking your USB dongle with a robotic 3D-printed hand. ‘Steve’ is obviously the sort of generic human name an artificially intelligent would-be sex offender would choose, because people know better than to get their kit off for someone called HAL 9000.

  39. @ Tim Newman,

    I’m not sure what you are expecting. The difference between prop and jet is one of speed and efficiency, cars have largely achieved this in other ways, refinements, but we can point to multiple different ways of powering cars, petrol, diesel, electric, LPG, hybrid. Furthermore, all those refinements amount to a similar change in performance.

    The difference between Wright brothers vs F35 Lightning being comparable to Benz’s first car and a 2014 F1 car (the F1 car is even limited by regulations, it is not the pinnacle of what could be achievable).

    Besides, the step change is right around the corner, perhaps the greatest inefficiency of a car is that it sits idle for 22 hours a day and then consumes two hours of the driver’s time. Driverless cars will allow us to get rid of half the country’s fleet of cars and let people do things other than driving. It’ll also unemploy ever cab driver in the country, making the country happier, safer and give cab drivers the opportunity to go out in to the world and finally put it to rights.

  40. My, what a lot of paranoia! Is this International Survivalists Day or something?
    For once, Cameron is right.
    And even if you don’t like it you’ll be able to “accidentally” disable it and join the six month waiting list for the Repair Man from the Ministry to come round to fix it.
    Oops! There it blows again!

  41. perhaps the greatest inefficiency of a car is that it sits idle for 22 hours a day and then consumes two hours of the driver’s time.

    That’ll be that danged “property ownership” thing, where something is idle and nobody else is using it because it belongs to you.

    My bed is idle around 16 hours a day. If you want the internet to alert two other people to come use the other two 8 hour timeslots, you can fuck off.

  42. @bloke in france
    “My, what a lot of paranoia! Is this International Survivalists Day or something?”
    Great deal of linear thinking.
    This is data FFS! What happens to data? It gets hacked. Any competent 14 y/o could convince the Man from the Ministry you don’t even own a fridge. let alone use any electricity.

  43. I’m not sure what you are expecting.

    I’m not expecting anything: but those who made the predictions in the 50s thought we’d be in hovercrafts powered by who knows what.

    The difference between prop and jet is one of speed and efficiency, cars have largely achieved this in other ways, refinements, but we can point to multiple different ways of powering cars, petrol, diesel, electric, LPG, hybrid. Furthermore, all those refinements amount to a similar change in performance.

    It’s not about performance, it’s about the fundamentals in how they work. The hybrid threatens to be a step-change, using a completely different power system, but the difference between a diesel and petrol engine is miniscule in the grand scheme of things. They both rely on reciprocating pistons under pressure turning a crankshaft, and then transmitting that rotary action to the wheels. The jet engine works in a fundamentally different manner from a piston engine, it’s a completely different animal.

    The difference between Wright brothers vs F35 Lightning being comparable to Benz’s first car and a 2014 F1 car

    I don’t think it is. Other than go faster, what can an F1 car do that a Model T could not? Whereas the development of the aircraft has seen a move from piston engines to turbo props and jets; from atmospheric to pressurised fuselages; and to fly-by-wire which allows control which would be impossible using only human inputs (and yes, this is fundamentally different to cars, where electronics greatly assist human inputs but are not essential for the general operation of the vehicle: take all the electronics out of a modern car and replace them with cables, and it’s still driveable; do that with a modern fighter and it simply wouldn’t fly). All of those are step changes which have opened up whole new possibilities and applications for aircraft, to a far greater extent than we’ve seen with cars.

    Besides, the step change is right around the corner

    It’s been right around the corner since the 1950s, along with the paperless office. Yet somehow, it stays pretty much the same, year after year.

  44. Tim Newman,

    “The most interesting thing about reading those predictions from the 50s about how we’d be living in the year 2000 is how unbelievably wrong they got everything.”

    That’s why I’m really suspicious of government ever talking about future technology. In the 1970s they thought the world was going supersonic, but what people really wanted was to fly cheaply. I remember Blair talking about virtual reality, just before everyone forgot about that when the internet exploded.

    My guess – the next 20 years will see more technological advances in medicine than almost anything else – replacing expensive surgeons with tablets and maybe robots.

  45. That’s why I’m really suspicious of government ever talking about future technology.

    Yes, and one other thing I forgot to add: not one of those 1950s predictions mentions the internet, or something similar. Silver space suits and chemicals instead of shaving yes, but never a worldwide web.

  46. Tim
    It depends on who’s doing the predictions.We did this a few months ago with flying jet-packs, didn’t we?. Nobody who does physics or chemistry predicted flying jetpacks. It’s an energy density thing. There’s no possible way enough energy can be packed into a small enough mass to make them anything but a novelty demonstrator. There can’t be the endurance to make them a viable everyday transport option. Not until you get some new physics because the limits of chemistry are known.
    There’s some serious people played with VR. They know the colossal processing power needed to create acceptable VR. One of them wasn’t Blair.

  47. @ Ian B, indeed, but in cars, the major benefit of ownership is convenience, beds have greater hygiene and privacy issues. If convenience is less of an issue because you can get a car almost anywhere then ownership is less of an issue. In fact driverless cars will be more convenient than ownership, no time spent parking or walking from parking space to destination. The major driver of the cause though is likely to be insurance companies, with premiums for self driven cars increasing.

  48. Tim Newman – the closest I can think of was Asimov’s Multivac, and that was a single gigantic computer.

  49. but in cars, the major benefit of ownership is convenience, beds have greater hygiene and privacy issues. If convenience is less of an issue because you can get a car almost anywhere then ownership is less of an issue.

    I think this is the reason why most of these predictions fail: they ignore obvious human factors such as aesthetics and emotional attachments. For whatever reason, a *lot* of people are *very* attached to their cars, and want a particular model, style, colour, and range of accessories which suits them. If this were not the case, we’d all be driving about in silver Toyota corollas, and nobody would buy a BMW with an M-Sport bodykit and interior. An awful lot of people *like* owning a car, their *own* car, and don’t want to hire one…even if for a lot of people it’s cheaper and easier to do so.

    And that’s why I don’t think the convenience of shared car ownership will be enough to fundamentally change the relationship between people and cars, there are other messy human factors to consider which inexplicably get ignored. The huge one that got missed was that when reviewing all but small amounts of information quickly and accurately, most human beings prefer a piece of paper in front of them rather than using a screen: hence printers companies are going great guns.

  50. @Tim newman

    It’s not about performance, it’s about the fundamentals in how they work.
    Why worry about changing the fundamentals for the sake of it if you can get the performance increase from incremental changes? Gains are still gains.

    The change in aircraft design allows the movement of people and things from one place to another faster and in greater numbers, how it does it is irrelevant. Exactly the same can be said for motor vehicles. it matters not one jot that it is because of step-change or incremental gains.

    The fundamental reason for fly-by wire computer controlled planes is to allow them to be built with added instability to improve performance. The only reason this hasn’t been done with F1 cars is because of the regulations. If there were a motorsport category that allowed for driverless cars then they could be built so unstable as to be undriveable by a human and be faster than anything currently in use, however as a sport it might not be very popular as it loses the human drama element. Such a car would also be useless on the public highway. With no market in sport or personal use there has been no cause to build it yet.

    However an element of this technology is present in modern cars, Electronic Stability Control takes over at the point where the driver cannot control the car any longer, it lies on the boundary of what is driveable and what is undriveable. ESC may be largely responsible for the difficult-to-attribute 30% reduction in road fatalities in the last 5 years after quite a long period of relative stability in the fatality rate.

  51. @ Tim Newman

    Re. aesthetics etc. I agree this is a difficult to quantify element but I think we are entering an age where car ownership is not as attractive as it once was. Younger people in particular are turning away from car ownership either because it is simply too expensive or, some suppose, because they are more attached to their smartphones which they cannot use while driving but can while on the bus/taxi.

    Numbers of drivers under 22 has fallen by something like 20% in the last few years, not limited to the UK, similar experience in the US

    http://news.nationalgeographic.com/news/energy/2013/12/131217-four-theories-why-teens-drive-less-today/

  52. bloke in spain,

    fair points. One of the problems of the past is that we are often looking through the lens of the media that gets stories fed to them by corporations looking to provide glamorous PR, rather than anyone actually doing a real analysis.

    That said, if you’d told me in the early 90s that you could put 4GB of storage onto a fingernail within 20 years I’d have been highly skeptical. You needed a box the size of a washing machine for that.

  53. cont.

    In addition I would say it will depend what the cool kids do. If self-driving becomes a quaint 20th century anachronism, as I believe it will, it’s over.

    Also known as… Why drive a car when you could be having sex in a car driving itself?

  54. Why worry about changing the fundamentals for the sake of it if you can get the performance increase from incremental changes? Gains are still gains.

    Yes, the gains are fine, but they’e not going to change the way we live and catapult us to a new era of human civilisation.

    The change in aircraft design allows the movement of people and things from one place to another faster and in greater numbers, how it does it is irrelevant. Exactly the same can be said for motor vehicles. it matters not one jot that it is because of step-change or incremental gains.

    Well, yes it does. It’s the difference between people getting to work more comfortably, cheaply and safely…and the entire world being within a day’s travel time. Fundamentally different things, only one of which changed the entire world as we know it.

    Such a car would also be useless on the public highway. With no market in sport or personal use there has been no cause to build it yet.

    Exactly: there has been no fundamental change because one has not been required. Yes, it would have been possible to fundamentally change the way cars work…but it would not have been to anybody’s benefit. With planes it was different.

    Back to this:

    Why worry about changing the fundamentals for the sake of it if you can get the performance increase from incremental changes?

    The best example I can think of is the pissing about with windmills. Yes, they have made gains since the early Dutch designs but there has been no fundamental shift in how they operate (or their limitations). This is why at least one major oil company has pulled out of their development, and instead put their money into solar power – where an enormous step-change is expected (or at least possible) whereby solar power genuinely can alter the world as we know it. The efficiencies (gains) are already on the way, once they’ve solved the storage problem (which would be the step change) we’ll barely look back.

  55. Magnusw – “Why drive a car when you could be having sex in a car driving itself?”

    Because KITT might get jealous and deliberately jostle you on speed bumps.

  56. – a way of boosting productivity, of keeping us healthier, making transport more efficient, reducing energy needs, tackling climate change…. warts, shingles, piles, the ague, diarrhoea, infertility…

    Dr Cameron is a snake-oil salesman. Don’t buy what he’s selling.

    And I would have thought it clear by now, Cameron does not understand that Internet thingy, or anything to do with science, or much at all really.

  57. So we have two politicians with crystal balls picking technological winners for the future with taxpayer money… what could go wrong?

  58. Exactly: there has been no fundamental change because one has not been required.

    Going back to your original point then, one can’t have expected to see any fundamental changes because of the limitations of the road system, not because of an inability to come up with changes, but because the changes that are required are changes to the system not changes to the vehicle as they were with planes.

    The system used for flight has never changed but the planes have. you go to an airport, you fly, you land, the planes just got faster.

    The fundamental change required for cars has not been one dependent upon car manufacturers, but on computer manufacturers, GPS which has only been available since 2000 and the ability of cars to communicate with one another instantly. Once those technologies were in place it was just a case of the computer boffins and the car manufacturers working together. It all exists, it all works, we are just waiting on society and governments to catch up.

  59. Going back to your original point then, one can’t have expected to see any fundamental changes because of the limitations of the road system, not because of an inability to come up with changes, but because the changes that are required are changes to the system not changes to the vehicle as they were with planes.

    My original point was that the motorcar has barely changed in the fundamentals when all the predictions from decades ago thought it would.

    The system used for flight has never changed but the planes have. you go to an airport, you fly, you land, the planes just got faster.

    Yes, but the step-changes in technology allowed planes to go from carrying a handful of people relatively short distances in extreme discomfort to flying hundreds of people long distances in comfort (insert jokes about economy class here). I’m not convinced it was merely because the system of the motorcar didn’t allow a commensurate change, more than it was that such a change just wasn’t required: they got it pretty much right first time off, which is pretty amazing when you think about it.

    It all exists, it all works, we are just waiting on society and governments to catch up.

    Perhaps, but it very much remains to be seen. Having not changed in the fundamentals for over a hundred years, and despite being written off on countless occasions, the motorcar as we know it is still alive and well. Personally, I think driverless cars will be a flop, possibly for the same reason pilotless planes won’t take off: the public won’t take to them.

  60. they got it pretty much right first time off

    60 million dead in the 20th century and over 1 million dead each year since would beg to differ. Driverless tech could save hundreds of thousands of lives per year because 90% of crashes are caused by drivers. Cars don’t kill, people do.

    the public won’t take to them

    We are purely speculating, bookmark this post and come back in 20 years to see who is correct!

    IMO as people’s priorities change, technology matures, we become ever more risk averse, ever more trusting in technology, insurance premiums spiral upwards, we value convenience more and more, society will quite readily accept driverless cars.

    Perhaps the difference here is that once a government has permitted the use of driverless cars, it doens’t matter much if you are in one or not, you are sharing the road with them and the associated risk or lack thereof. Once people see the benefits to the early adopters of convenience, cost and safety, everyone else will follow.

  61. I personally would love a driverless car (assuming the technology can be trusted obviously).
    I’ve never learnt to drive because I’ve never needed to – lived & worked in cities with good enough transport systems + taxis, plus owning a car with all the costs made no sense when it would be an occasional use item.
    So I suspect I may be more accepting as I wouldn’t be comparing it to my driving and I also accept that I now lack ~20 years of ingrained driving experience and wouldn’t fancy tackling motorways, etc now.
    But I can also see the pool of people who fit at least some of this isn’t small – if you’re young with less disposable income and owning a car isn’t a necessity, would you have one? – and once you’ve gotten used to not having to worry about one, would you be more inclined to view ownership as another burden?

  62. 60 million dead in the 20th century and over 1 million dead each year since would beg to differ. Driverless tech could save hundreds of thousands of lives per year because 90% of crashes are caused by drivers. Cars don’t kill, people do.

    The reason I’m not so sure is because such levels of safety and reliability come at an enormous cost*, and outside the airline, nuclear, and petrochemical/oil industry – which still needs human intervention when the shit hits the fan – thus far its not been seen as a price worth paying. Technology has existed to reduce road deaths for decades, but the public – to varying extents depending on the country – has accepted road deaths as an acceptable trade-off for the convenience of individual car travel. I’m not sure that sticking a honking great price tag onto reducing deaths which the public, by and large, found acceptable is going to fly.

    *My day job involves me in this. Our control systems have what’s called a SIL rating – Safety Integrity Level, which is basically a reliability measurement of any particular piece of equipment. The greater the reliability, the more expensive it is, so we only use the high-rated stuff where we have to. Coupled with that, you need redundancy and at least 2/3 voting for safety-critical systems. I just don’t see aircraft-levels of reliability being applied to cars at a reasonable cost, and the public willing to pay the premium over what we have already.

  63. and once you’ve gotten used to not having to worry about one, would you be more inclined to view ownership as another burden?

    Not just the financial burden; It’s something we don’t tend to hear much about at the moment but I’m sure it will be used to promote driverless cars, that is, the liability of driving a car. For the vast majority of people who are generally law abiding, driving may be the only reason they ever find themselves in the dock of a court. Why continue unnecessarily with the risk of killing someone and going to prison if you can if you can avoid it?

    As I say, not something people really think about at the moment when they jump in the car but I’m sure the issue will be used in future to promote driverless.

  64. As I say, not something people really think about at the moment when they jump in the car but I’m sure the issue will be used in future to promote driverless.

    Which will be undermined the minute somebody dies in a driverless car due to a system fuckup, and the company running it just shrugs its shoulders.

  65. Tim.

    I would argue that the difference is that for aeroplanes you are looking largely for mechanical reliability and if it goes wrong it kills hundreds of people.

    For cars, we are looking more at software reliability and if it goes wrong its only requirement is to bring the car to a controlled stop at the side of the road and even if that fails it’s probably only going to kill a couple of people. Of course software can also be expensive but it’s also relatively easy and quick to fix millions of systems in a short space of time, so the costs are widely distributed.

    I also would disagree that the public has not accepted higher costs for increased safety. Cars today are loaded with safety systems, crumple zones, multiple airbags, ABS, EBD, ESC, seat belts, head ‘rests’, tyre design, suspension systems, headlight design. You can still buy cheap cars without a lot of this gubbins, but hardly anyone would.

  66. It would be interesting to see how strict liability in motoring offences translates to driverless vehicles.

    Also, does this mean “have a drink, have a drive” will become a popular activity again? Will cars of the future tout a mini bar as an optional extra? Sozzled minds want to know.

  67. Which will be undermined the minute somebody dies in a driverless car due to a system fuckup, and the company running it just shrugs its shoulders.

    It’s certainly going to be interesting to see how countries deal with the different legal scenarios, where liability lies and whether it will be considered criminal. Another “wait and see”.

  68. Anthony said: “Technically you’re correct for that specific example. But then your fridge would need to include at least a set of indicator lights for all the moving parts and the amount of information you could get from that is small. If it is connected to the Internet then it can release much more information.”

    The internet still isn’t necessary. Just plug in a computer, tablet, smartphone or whatever like they already do with On Board Diagnostics code readers for cars or make them network enabled and accessible via a home computer.

    Even if appliances become able to churn out useful telemetry the need to be transmitting it to somewhere else via the internet has yet to be well made imo, with the exception of those who wish to remotely monitor their appliances themselves.

  69. Of course, we have a well developed body of case law (since Donoghue v Stevenson’s famous snail-tainted ginger beer) and more recent legislation concerning liability for harm caused by consumer products.

    We also have a number of strict liability offences in criminal law for naughty things people do when “in charge” of a vehicle, but who is “in charge” of a Johnnycab when it goes awry? The owner? The passenger who specifies his destination? Google? Navteq? ARM? BMW? The part of the US government that runs the Global Positioning System? Should be good business for lawyers and insurers.

  70. It will be good business for them for a few years while the justice system gets used to it, but with fewer possible scenarios and much greater evidence available (3D super slo-mo visualisations of accident scenes) it won’t be long before most scenarios don’t even reach court without a settlement.

    Maybe more liability shifts from driver to manufacturers and insurance premiums move with it, but with far fewer accidents it will overall be much cheaper assuming that eventually the consumer always pays anyway.

  71. @Steve – presumably out of all those we could eliminate the owner (unless it was shown to be related to maintainence negligence).

    I’d arguing it’s like a taxi ride – if it crashes, it may be the mechanic who maintained it, the cabbie who drove it or the owner who was responsible for maintaining it, but it’s generally not the passenger.

  72. Bloke in Costa Rica

    Prediction is very difficult, especially about the future. I think the most likely way the the Internet of Things will come about is in the same way that mobile phones evolved from portable voice communications devices to general purpose computers with a telephony function: incrementally, but radically. My previous phone was a Nokia 1616 dumbphone. It does its job well, but its functionality is very constrained. My current phone is an iPhone 5S, which has roughly the same performance as a mid-80s supercomputer or a mid-2000s desktop and has functionality that is extensible almost without limit. They are chalk and cheese. Yet the line connecting the two is continuous. If there’s one invariant in technology, it’s that every advance engenders new ways of doing things that were practically unforeseeable. There will be an Internet of Things, but the form it will take will not be apparent until after the fact.

  73. Driverless cars are inevitable, in the same way that cars were inevitable.

    It was the cart (or the wheel) which wasn’t inevitable.

    Once you had that, it was just a question of waiting to see how and when, and driverless-wise the how is mostly sorted.

  74. Interested: it was the axle that was the invention, not the wheel. (As you and any fule kno.)

    Anyway, channeling P J O’Rourke:
    At least when I’m transporting an underage girl across state lines with a kilo of cocaine in the boot I won’t get busted for drunk driving.

  75. @ Interested

    I’m not sure even TN is arguing they are not inevitable, rather questioning, will they be popular.

    Eg. Amazon’s octocopter was a pre-Christmas publicity stunt, however I’d bet one of the first driverless vehicles we’ll see on the streets is essentially an Amazon locker with an electric motor, that’s inevitable.

    Mainstream ownership/rental is less inevitable though I do think it will happen in part because of the above Amazon example. We’ll get used to having driverless cars around us, it’s not that one day driverless cars will be released and we’ll all change. First you’ll see an Amazon locker roaming the streets; then you’ll have a delivery made to you by one; then a friend will tell you they had a driverless taxi ride; then you’ll dip your toe in with the odd short driverless taxi ride when you’re pissed, etc. Any taxi/delivery company that can will be looking to unemploy as many people as possible, there’s a comparable number of cabbies (300,000) in the UK as there were miners in 1980 (230,000), imagine the trouble that’s coming.

    I think the way tech seems to have accelerated in the last 10-15 years this process will be much shorter than it might have been in the past.

  76. Magnus
    I agree it’s coming. Don’t agree it’s coming faster.
    30 years after Gutenberg there was a printing press in every major town in Europe.
    60 years after the invention of the micro-wave oven, they were still a niche machine for ships and canteens.

  77. Bloke in Costa Rica

    “60 years after the invention of the micro-wave oven, they were still a niche machine for ships and canteens.”

    What does this mean? Sixty years after the invention of the microwave was 2007. Most people I know had had one for 20 years by that point.

  78. Costa
    re microwaves
    They had worked out the physics before WW2. Like the laser, they didn’t know what to use it for.

  79. Tim Newman: Vannevar Bush suggested the “memex” machine in the 30s and wrote a lengthy essay in 1945 which effectively predicted hypertext and links. To be fair though he imagined it as a purely mechanical device with microfiche.

  80. bloke in spain
    March 10, 2014 at 12:47 pm

    @Gamecock
    “If your car is over 10 years old, it’s likely that the ECU is no longer made.”
    I’m pondering this because my bus is getting on. But that clunky manufacturer’s ECU isn’t necessary. My garage pal can manage it all on a laptop, by-passes through the diagnosis port. The processing power required’s trivial. Twenty-five quid’s worth of Raspberry Pi’d blow an ECU away. It’s purely a software hack.

    ============================

    If you ECU craps out, there’s no where to put a software hack. You’ll need another ECU. Which isn’t made any more.

  81. For cars, we are looking more at software reliability and if it goes wrong its only requirement is to bring the car to a controlled stop at the side of the road and even if that fails it’s probably only going to kill a couple of people.

    The problem is that risk is the product of consequence and likelihood; aeroplane safety is taken seriously because the consequences result in a high risk; whereas car safety is taken seriously because the likelihood results in a high risk. If something goes wrong with the software only once per million (which will be difficult to achieve cheaply if my knowledge of SIL rated equipment is anything to go by) we’re still looking at multiple failures every day. So it comes down to ensuring a software or communications failure doesn’t result in a fatality or injury, and I’m sure we’re going to have a lot of fun trying to achieve that state of affairs, especially when even bringing a car to a controlled stop by the side of the road can have pretty unpleasant consequences in certain conditions if the driver has no say in the matter.

    I think those who think driverless cars are inevitable are overlooking the one, irrational factor that humans have which makes them prefer cars to planes: the feel of being in control. I’m just not sure people want to surrender the control, not least because people seem to be willing to spend tens of thousands on driving for fun.

  82. Have a though for the poor old pensioner. What will he use for money to buy all these expensive devices that break and/or be come obsolescent with ease.

  83. @Gamecock
    If you haven’t heard of the Raspberry Pi, it’s an ARM processor on a motherboard’ll fit in a fag packet. About the same processing capability as a mid 2000’s desktop. It’s also possible to directly input & output data from & to sensors & actuators from the board. Cost’s £25 for the basic computer.
    The vehicle diagnosis kit my garage pal has runs one of the legacy O/S’s. Sun I think it is. There’s a port for Raspberry which otherwise usually ruins Linux. Pal’s diagnosis kit will manage engine parameters directly through the diagnosis port.
    It’s simply a case of sending all the car data to the Raspberry & letting it do the ECU’s work. The processor’s orders of magnitude more capable than an ECU chip. Set up right, you’d have the same sort of management quality as a F1 car.
    There’s a few people playing with code, currently. All very much state of the art.
    There’s Raspberry’s running small drone aircraft autopilots including GPS navigation. The code for that’s Open Source, downloadable off the net.
    There’s one sitting on the table by me, currently rendering a movie file into a TV, via thitse HDMI port & its integral graphics processor. It’s not even breathing hard. It’s being controlled from the laptop I’m typing on. I’m only doing that to save plugging in a keyboard & wifiing it to the internet to write this.

  84. bloke in spain
    March 11, 2014 at 12:28 am

    @Gamecock
    If you haven’t heard of the Raspberry Pi,

    ========================

    No, hadn’t heard of it. Sounds promising, but it’s not around here, yet. I blew the ECU on my ’97 Dodge pickup, and the shop had to scrounge a rebuilt one off da net. $475 US.

  85. Um, Raspberry Pis are everywhere. I have one in Costa Rica. Anyone with any programming chops can make it do things that would have required several tens of thousands of dollars’ worth of SCADA gear ten years ago.

  86. @magnusw: March 10, 2014 at 4:03 pm

    Also known as… Why drive a car when you could be having sex in a car driving itself?

    Assuming that this is in one of the “public/pool vehicles” one hopes that there would be sufficient AI built into the thing that it will nip off to get the back seat cleaned before putting itself back “on hire”! 🙂

  87. @Ian
    “who programs the Raspberry Pi to emulate the ECU?”
    Well, logically the user does.
    But it’s not quite as daunting as that.
    The Pi community tend to post hacks as Open Source downloads. So if you wanted, for instance, to build the drone aircraft with it’s GPS navigation, you don’t have to write all the code yourself. It’s all there available.
    Duplicating a vehicle ECU shouldn’t be all that hard, compared with what a Pi’s capable of. There isn’t that much going on in one. Pi could probably handle the car’s entertainment system in its spare time.
    And it’s not hard to see the incentive for the hack. New ECU for my wheels is listed at $1500

  88. Not necessarily, a driverless bus can still carry more people more efficiently than 80 driverless cars. Trains will still cover longer distances faster. Driverless will be more efficient than currently but there are still physical limits of space, and speed, congestion.

    As cities move towards greater restrictions on private cars entering their centres new methods of linking up public/private driven and driverless will emerge.

  89. I wonder whether we’ll ever move to pilotless airliners? Entirely possible now, but there is a bit of pushback.

    I must confess I’m not sure why – they would probably be safer than piloted aircraft. But I wouldn’t get on one, despite my claims of inevitability for driverless cars.

  90. magnusw
    March 11, 2014 at 3:03 pm

    Driverless will be more efficient than currently but there are still physical limits of space, and speed, congestion.

    =========================

    A friendly government agent will program your car. 40 mph seems fast enough.

  91. Interested
    March 11, 2014 at 3:35 pm

    I wonder whether we’ll ever move to pilotless airliners? Entirely possible now, but there is a bit of pushback.

    I must confess I’m not sure why – they would probably be safer than piloted aircraft. But I wouldn’t get on one, despite my claims of inevitability for driverless cars.

    ==========================

    Stories of epic failure are legend. 3 airbus crashes that could have been prevented had the pilots been able to override the computer.

    And the Kursk sinking was a programming error.

  92. ITBoy
    March 12, 2014 at 8:19 pm

    “Kursk sinking was a programming error” – citation?
    Never seen any mention of that

    =========================

    My bad. I was thinking of another nuke sub disaster, and not the Kursk.

    Browsing duh net for more info on the disaster, I couldn’t find more details. It occurs to me that what I heard was off the record, so I’m not going to name the sub. The story basically went that the sub was doing max depth testing, and the reactor had a fault and shut down. They quickly determined that there was no serious problem, and tried to restart the reactor. There was a 7 minute programmed delay on restart, and the sub sank to crush depth with a perfectly usable reactor that locked out by software.

    At least that’s the story I heard.

Leave a Reply

Your email address will not be published. Required fields are marked *