“Self-driving car” and “disaster” is no oxymoron

We probably need go no further than the opening paragraphs to get to the nub of the matter:

IT’S BEEN NEARLY a year and a half since Joshua Brown became the first person to die in a car driving itself.

In May 2016, Brown was on a Florida highway in his Tesla Model S using Autopilot, the semi-autonomous driver assist feature that handles steering and speed during highway driving.

Tesla has always warned drivers that Autopilot isn’t perfect. According to car’s driving manual and the disclaimer drivers accept before they can engage it, the system should only have been used on highways with clear lane markings, strict medians, and exit and entrance ramps.

So when a tractor trailer turning left crossed into the Model S’s lane, the system did not recognize it—and the car crashed into its side, killing Brown instantly.

That system, written and constructed by humans enthused by the concept and dedicated to Musk, plus those humans tasked, like the old camel and committee, with foreseeing EVERY SINGLE POSSIBLE contingency, combined, encompassing ALL ramifications, each dovetailing with the input of every other human possibly encountered on the road – that is the task of Mr. Musk’s engineers.

Not unlike Network Rail dovetailing with each and every train company and each and every set of tracks and points through history, throughout the country, even taking into account the weather on the day and/or the folly of some drunk, some child or some jihadi.

Let alone, as Ivan once mentioned, the teething period while one “all or nothing” system is replaced by another, or as I put it – the human replaced by the robot, something Musk’s backers are right into in a big way in their arrogance.

Some other unforeseen aspects:

Semi-autonomous driver assist feature

Think that one through. It’s not “selfdrive” at all, it still requires human input but not only does it require constant human oversight and instant mental/physical reaction to stimuli of the human kind when the human does not obey the rules as imagined by Musk … but it also now requires a new set of human skills on top of that … and needs to give the human instant permission to override.

Permission to override

This is the crux of the matter, for we have a philosophical, Steppenwolf type dilemma here between Man and Machine [or between Man and AI].

And for someone like me, trusting that humans WILL act stupidly and planning around that every day, comes this enforced override by the machine, within a split second.

What I’m up against in fact, before even mentioning the foibles of tech itself and my own distrust of the humans behind it, therefore of it, causing me to automatically reach for manual override, while it assures: “Trust me, I know everything …” is that I am here pitted against a mind or minds who think they have it all sussed, that they’ve thought of everything, when I don’t believe they have done that.

And clearly, in Mr. Brown’s case – they hadn’t.

Read this blog, read OoL and Julia’s case after case and one sees human folly writ large. What, must every human who plans to drive one of these Frankensteins sit an aptitude test before taking delivery?

Has no one heard of cascade effect? The compounding of error? Murphy’s Law? Am I being over-emotional here, hysterical? Just like half of humanity? Do you resent my implications? Is your annoyance with my attitude and behaviour conducive to the smooth running of an auto-drive system design?

Oh, self-drive compensates for my human folly, does it?  Precisely which aspects does it compensate for and which does it demand the now laziness-induced, sloppy, 20 year old hooligan driver who religiously believes in his new tech to the point of blind faith – which bits are left to his “discretion”?

And on whose say-so? Some person who came to work one day and made that life and death decision.

No thank you.

Practical example

My mate had a Volvo, which had many of the auto-decision features which constantly override human experience – and he’s very experienced.

Now, think about the Swedish mind for a moment, the same mind which invites in the Muslim disaster and “loves it” into failed integration.

All of this – the sum total of all human foolishness combined – came together one day, just as he made the decision to overtake and then saw the car’s computer take the matter out of his hands, chiding him: “Naughty boy, decision rescinded, acceleration denied …” whilst an oncoming car barrelled down on him at a rate of knots.

Interesting isn’t it, the combined speed at which two cars meet, each doing 60 mph.  Survival chances?

He sold the Volvo days later and bought an older model, a basic car with few bells and whistles, no overrides.

The cognitive dissonance of a techie’s mind

My mate’s a techie, he believes in tech, loves it, eats it, breathes it, is as happy as a pig in s*** when wallowing in it for hours.  He can design systems which do it all for you and delights in taking into account all contingencies and ramifications.  He’s paid big bikkies for this.

On the other hand, like most techies, he couldn’t stand what Jobs had done, the way Apple allows no discretion – it’s either their way or the highway and now it’s £1000 for an iPhone.  He has multiple parts of his own system, all manually combined, written in by hand and one can mix and match, he orders in the bits he needs.

He is a “manual” type person who likes to do things hands-on, so he knows it’s done right.

Me?  I had a Mac and still do have with my iPad.  Starting this post, I was in bed with the iPad but it has limited use for a blogpost so I got up and switched to my laptop which has Ubuntu, which allows me to put on it whatever I damn well like.  It was a reaction to Windows 10 which goes in for this “cradle to grave” nannying, leaving you at the mercy of mischievous and hopeless script kiddies.

Despite this post, I appreciate tech, love to use what it offers … as long as I can manually overrride.  Example – I can alter PHP in the Editor but don’t because I don’t know what the guy who wrote it had designed in.  However, where he makes it clear I can alter a user-interface message, then I might do, providing I’d saved the file first and then after.

My filing system’s not bad.

All of the above was to underline that we live by trusted systems, tempered by our own experience and ability to alter.  Take the discretion away and give it to some kiddy in Sweden or even Musk and I for one start to worry, not unlike my mate on that road on that day.

Obdurate tech arrogance especially threatens the innocent

It’s not just the blind believer in the omnipotence and infallibility of tech who puts himself at risk – he puts at risk every other person in the vicinity.  Perfect example is Skynet. How convinced was Arnie’s bird’s father that he had it all at his fingertips?

It takes me a long time to trust and it can be lost in an instant.  In 2006, I wrote a post on Maglev.  I’m just coming around to it now with a view to going on a Maglev train.  Almost.  Still not ready for Boeing or Airbus.

And one thing that not only will I not use but do bitterly resent is the enforced self-drive garbage being prematurely pushed onto people by pollies and script kiddies – it’s fine, trust us.

No thanks.

7 comments for ““Self-driving car” and “disaster” is no oxymoron

  1. Sackerson
    September 14, 2017 at 06:51
  2. September 14, 2017 at 08:33

    ” I can’t do that Dave”.

    Daisy, Daisy, give me your answer do…..

  3. Richard
    September 14, 2017 at 08:55

    And if you call in sick then program a journey to take the kids to the beach it’ll take you to work instead, or if a council worker sees you chuck an apple core it’ll be “pinged” and stop to allow you to be fined, and if you tweet “hate speech” it’ll drop you off at the courthouse.
    Not to mention a reduction in non-essential journeys to appease Gaia when the bastard won’t go.
    There’s an agenda here, sure as fuck. Someone will be at the wheel and it won’t be you.

  4. September 14, 2017 at 09:39

    Sackers, Amfortas and Richard – yes.

    “I’m half crazy, All for the love of you!” Indeed.

  5. microdave
    September 14, 2017 at 10:25

    A self driving 1989 Fiat Panda 4×4? – I don’t think so. With a manual gearbox, NO powered anything (except the brake servo), no sensors or cameras – and it’s manual choke, so I’d LOVE to see a computer try and juggle that on a cold morning!

    Just makes me more determined to keep the cantankerous old ‘bus running as long as possible…

  6. Letmebe
    September 14, 2017 at 10:56

    Most mornings I like to sit at the side of our High Street with a coffee before going on to work. Yesterday, third time since last Summer, another goon backed into me. “Sorry, the parking sensor doesn’t seem to be working”. Third time. All men. I’m going to have to start drinking coffee with the stupid plastic lid on. My secretary thinks I’m incontinent.

  7. Ken Craggs
    September 15, 2017 at 11:49

    Self-driving police patrol cars are being introduced in Dubai. It’s called the O-R3

    The government has announced that partially self-driving lorries will be tried out on major British roads by the end of next year. The Americans might refer to similar vehicles as being ‘Autonomous Juggernauts’.

    In this video from 2013, DARPA Program Manager Kathleen Fisher talks about the ability to hack into a car computer system and take control of the vehicle.

    In 2013, Rolling Stone’s reporter Mike Hastings was investigating CIA Director John Brennan before Hastings was killed in a suspicious car crash. Some people claim that his car was hacked, taken control of remotely, and then crashed into a tree at about 100 mph.

Leave a Reply

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