Automation turns us from actors into observers

November 2, 2013 7:50 PM

Nicholas Carr's latest article for The Atlantic posits that automation presents risks, specifically of losing skill and talent. "The lack of awareness and the degradation of know-how raise the odds that when something goes wrong, the operator will react ineptly. The assumption that the human will be the weakest link in the system becomes self-fulfilling."
posted by Athanassiel (92 comments total)

“If men learn this, it will implant forgetfulness in their souls; they will cease to exercise memory because they rely on that which is written, calling things to remembrance no longer from within themselves, but by means of external marks. What you have discovered is a recipe not for memory, but for reminder. And it is no true wisdom that you offer your disciples, but only its semblance, for by telling them of many things without teaching them you will make them seem to know much, while for the most part they know nothing, and as men filled, not with wisdom, but with the conceit of wisdom, they will be a burden to their fellows.”

Plato

Technology has always done this. I'll take the odds of autopilot, thank you.
posted by zabuni at 8:00 PM on November 2, 2013 [23 favorites]

“If it weren't for the people, the god-damn people' said Finnerty, 'always getting tangled up in the machinery. If it weren't for them, the world would be an engineer's paradise.”
posted by ZenMasterThis at 8:10 PM on November 2, 2013 [6 favorites]

I thought this was kind of a weak article when I read it last week or whenever it came out. It seems built on a false premise that it's actually aware of, and further in its litany of just-so stories does not really investigate what a solution would be, in favour of presupposing that having people learn all this shit that will be redundant 99.999% of the time will solve the "problem".

Carr acts like a) several one-off anecdotes are representative (eg the problem afflicting some pilots is a problem for aviation everywhere, is a problem for many automated fields, can be resolved in the same way) with very little evidence, and b) basically ignores how industires are currently addressing it, and what a solution outside of "stick with training methods that don't suit the current environment". Fact is, even with the odd failure, machines are better than humans.

I dunno, I thought it was very Gladwellian - lots of generalisations, conjuring a problem that may not actually exist with apocalyptic alluded slippery slope what-ifs, contrary suggestion etc etc.
posted by smoke at 8:24 PM on November 2, 2013 [8 favorites]

I would not totally dismiss Plato, or this article, more relevant today. A 21-year old, my daughter's friend, was disturbed by the fact that whenever a word or concept was called into question, someone with a smartphone would settle the issue.

In the good old days (like, three years ago), small groups would engage in conversation about the issue or word in question. Sometimes, the "answer" would never be resolved, that night. Maybe later, with Google. But that social and intellectual dynamic is being lost, I think. I just noticed this last summer with a brother-in-law with a know-it-all phone.

OK, it's a new world; nobody can change it. But I liked the old way better. Human beings, in groups, solving problems together. Smiling and hugging and shooting each other. The good old days. (Yeah, tonight, after sitting through a nice production of Chekov's The Three Sisters,, my wife, a lawyer, recounted an incident in which a couple of Russian guys got so heated in an argument about the tablua rasa theory versus the a priori theory of knowledge that gunplay ensued.)

Anyway, fuck instant computer knowledge. That's my P.O.V. I'm obviously too old to even be allowed on Metafilter, but this is my gut instinct about our reliance on computers as an extension of our meat computer: it decreases the value of the intuitive and rational capabilities of the brain that has been evolving for a hundred thousand years or so.
posted by kozad at 8:24 PM on November 2, 2013 [12 favorites]

The article cites Brynjolffson & McAfee's Race Against The Machine; there is a TEDx talk by McAfee on the book's arguments.
posted by Going To Maine at 8:26 PM on November 2, 2013

What a load of luddite shit.

Has there ever been an automated process that failed more often than humans doing it by hand? I'll take autopilot over human attention spans, exhaustion, and all the other problems that plague healthy, well-trained operators every time... not counting all the addicts, people contemplating divorce or suicide, or simply those who partied too long last night.

Statistics clearly prove, beyond a shadow of a doubt: ATMs make fewer mistaks by far than bank tellers. Automated mass transit is safer than any transport we've devised before. Pharmacists made more mistakes, with horrible consequences, before computerized supervision of their orders.
posted by IAmBroom at 8:29 PM on November 2, 2013 [10 favorites]

Technology has always done this. I'll take the odds of autopilot, thank you.

This shouldn't be treated as a zero-sum game. Skilled pilots and capable autopilots should both be available. It's a matter of honestly addressing the limitations that technology is introducing to our skills and training folks to compensate.
posted by Going To Maine at 8:29 PM on November 2, 2013 [14 favorites]

In the summer of 2012 I worked in a lab at NASA Ames that was looking at reducing the commercial cockpit to one pilot and, although it was very hush hush (so nobody got too scared), eventually zero. The idea is that it's more cost efficient and potentially safer (that's what the research was determining) to have a few highly trained people on the ground that are on-hand to deal with emergencies while computers do most of the actual work in the plane.

Unless the research shows that it's horribly dangerous and the FAA forbids it, I can guarantee you commercial cockpits will have only one pilot very soon (not hard to believe when you realize they used to have 4 or 5). I wouldn't be surprised to see it reduced to zero in only a couple decades, especially if automated cars are successful/widespread and people get comfortable with the idea of a computer controlling their fate in such a way.
posted by Defenestrator at 8:33 PM on November 2, 2013 [4 favorites]

Also, for anybody that is more interested in these kinds of human-meets-technology issues, especially when it comes to big systems like planes, you can check out the field of human factors. Also human-computer interaction. There is a lot of research on this kind of stuff, as the article hints at.
posted by Defenestrator at 8:37 PM on November 2, 2013 [6 favorites]

Human beings, in groups, solving problems together. Smiling and hugging and shooting each other. The good old days.

I'd say that that still happens, we're just out there smiling and hugging and solving different problems. Conversing about the definition of a word isn't interesting. Discussing the applications of the concept, instead? Far more interesting. IMHO, of course...
posted by CrystalDave at 8:39 PM on November 2, 2013 [5 favorites]

Technology has always done this. I'll take the odds of autopilot, thank you.

I dunno. I see a lot of college freshmen, and I think "typing the first thing I think of into Google and skim the first result" is partly to blame for a general incuriousness and passivity. These supposed "digital natives" are really only adept at doing things that companies want them to do, and that's a bit sobering. I seriously doubt whether technological solutions will ever give superior results compared to careful training, and the fact that the former is cheaper in the short run is not a great motive.
posted by GenjiandProust at 8:47 PM on November 2, 2013 [16 favorites]

Clive Thompson's book Smarter Than You Think is the thing to read if you want to see the "on balance, it's a huge net gain to have tons of physical and mental tasks automated and networked" case presented by somebody who's not a knee-jerk techno-utopian, a salesperson, or both.
posted by escabeche at 8:54 PM on November 2, 2013 [2 favorites]

Has there ever been an automated process that failed more often than humans doing it by hand?

Customer service?
Making bespoke clothing?
Finding a life partner?
Playing StarCraft?
Internet forum moderation (including right here on Metafilter)?
posted by FJT at 8:56 PM on November 2, 2013 [8 favorites]

GenjiandProust: "I seriously doubt whether technological solutions will ever give superior results compared to careful training, and the fact that the former is cheaper in the short run is not a great motive."

It's true that in some domains technology may not yield superior results compared to the job of an expert that has been carefully trained, but there is already plenty of evidence that it can and does often exceed the capability of the majority, who are usually less than carefully trained. Automation can still be a vastly safer or more efficient choice, overall, without reaching the capabilities of the best.
posted by Defenestrator at 8:57 PM on November 2, 2013 [2 favorites]

Rather like throwing sabots into the looms, isn't it? It's already happened, here in these parts where the future is better distributed. I'm interested to see where it leads. The young'uns are going to have all that space in their brains that's not taken up by lists of kings and the times tables and cursive writing - what on earth are they going to do with it all?
posted by Mary Ellen Carter at 9:04 PM on November 2, 2013 [2 favorites]

I think the article was a bit more thoughtful than "automation bad, human brain good". Of course automation has helped with things like operator error due to fatigue, etc. But he makes the point that with increased automation, "People end up functioning as mere monitors, passive watchers of screens. That’s a job that humans, with our notoriously wandering minds, are especially bad at." Automation isn't a bad thing on its own, since it could free us up for more interesting and creative tasks. But those tasks are more time-consuming when done by humans in a way that engages and interests them, and that undermines the almighty profit which is the goal of most industry.

I don't think it's mandatory to come up with solutions. What solution would you come up with? More automation means fewer jobs for humans at a time when populations are exploding, third world countries are working hard for a higher standard of living (and why shouldn't they?) while resources are beginning to run out. Maybe these are the big problems that we will be able to use our enormous brains to figure out now that automation has freed us up from all the mundane, boring tasks.
posted by Athanassiel at 9:15 PM on November 2, 2013 [6 favorites]

Sorry, I meant I don't think it's mandatory to come up with solutions at the same time as pointing out a problem. Whoops. Damn distractions. Human error triumphs again!
posted by Athanassiel at 9:16 PM on November 2, 2013 [1 favorite]

The ideas in this article were very well introduced back in 1983 by sociologist Lisanne Bainbridge.

Ironies of Automation. Good 4-page read, and widely used in grad courses in Human Factors.

My read of the consensus is that we know a fair amount about how to design highly automated systems that work well as a team with human supervisors, and keep them actively engaged and effective. It's just that the cost might be slightly higher than a crappy human-adapts design, so unless the consequences are dire (space travel) they don't always get built right.
posted by anthill at 9:21 PM on November 2, 2013 [9 favorites]

I've sold a lot of automation, designed some automation, led engineering teams that automated skilled labor processes successfully, and used a lot of automation I barely understood. The core problem, if there is such a thing, with automation, is that people don't think like machines easily, nor vice versa. When the automation breaks, or is presented situations out of its design range, people quickly stop working with the automation. And when people don't respond as the designers of automation expect they will when shown certain indications, the automation may do things its operators find counter-intuitive, and do them faster than its operators can understand or further control.

But there is a lot to be said, as a human being, for basic machines that have minimal automation. I drive a little truck with a manual clutch and a 5 speed, and less engine management than some more modern lawnmowers, because, well, it's fun, it's unambiguous, and it's reliable. And when it needs repairs or service, every guy I give it to grins a little, and dumps the clutch in low gear, just to squeak the tires a little, and I grin back...

That little truck doesn't do much, but what it doesn't do, it doesn't do extremely well, and all the time. But you know, for other people, I frequently recommend traction control and vehicle stability systems, and lots o' airbags. Safety is pretty cheap, these days. And who really want to shift a transmission in traffic, any more?
posted by paulsc at 9:42 PM on November 2, 2013 [9 favorites]

I'll take "looked it up on Google and accepted the first plausible result they found" over "didn't take the trouble to ever find out and continued in ignorance."
posted by Foosnark at 9:46 PM on November 2, 2013 [12 favorites]

Managing a stick in traffic doesn't have to be too bad. My diesel golf is a breeze with its ample torque and tall gears. I had an almost identical GTI with a DSG for a while, and I found myself getting tangled up with that transmission in the city, long after I ought to have mastered it.
posted by wotsac at 9:57 PM on November 2, 2013

"People end up functioning as mere monitors, passive watchers of screens. That’s a job that humans, with our notoriously wandering minds, are especially bad at."

That's OK, I'm designing an automated system to watch screens for me.
posted by happyroach at 9:58 PM on November 2, 2013 [2 favorites]

I hope that Carr will take the time to further flesh this argument out — especially around the question of why and when systems alone aren't better than systems and humans working together. He's not the knee- jerk Luddite that people seem to think he is, but it feels like he really glossed over the key question here.

He's definitely correct that automation harms human performance in a huge variety of tasks. But there needs to be a cost-benefit analysis to know if that's a worthwhile cost, like defenestrator's example with NASA.
posted by graphnerd at 10:04 PM on November 2, 2013 [1 favorite]

In the good old days (like, three years ago), small groups would engage in conversation about the issue or word in question. Sometimes, the "answer" would never be resolved, that night. Maybe later, with Google. But that social and intellectual dynamic is being lost, I think. I just noticed this last summer with a brother-in-law with a know-it-all phone.

I don't know, I think there can also be too much of the opposite. Smartphones and constant internet access has also revealed how pointless so much of that low-level uninformed pub debate is. One person is right. One person is not right. This is a question with an answer. We don't need to "teach the controversy" on what movie won Best Picture in 1977.
posted by Sara C. at 10:26 PM on November 2, 2013 [36 favorites]

Also, it's my experience that people stop doing that with smartphones after the first few months. Unless they're seriously insufferable or it's a real stumper of a question.
posted by Sara C. at 10:36 PM on November 2, 2013 [3 favorites]

The young'uns are going to have all that space in their brains that's not taken up by lists of kings and the times tables and cursive writing - what on earth are they going to do with it all?

We'll, if the young'uns don't know their multiplication tables, then they certainly aren't going to fill up that extra space with anything having anything to do with math. I know from experience that a calculator is no substitute for mental arithmetic, when doing anything much more advanced than arithmetic.

I think automation is great for repetitive tasks, especially ones that require some inhuman capacity (constant vigilance, precise timing, etc) or that are done too rarely if done manually. It's also fine if it can be done totally, to the point where it can completely replace a skilled human and give all people access to those skills. However, I think cognitive automation is unwise, and people should be taught to do things without computer assistance. Personal knowledge of things is how those things are expand and improved.

This story is relevant.
posted by cosmic.osmo at 10:39 PM on November 2, 2013 [10 favorites]

Teams of humans and computers beat the best humans or computers in chess.
posted by the man of twists and turns at 10:51 PM on November 2, 2013

I'll take "looked it up on Google and accepted the first plausible result they found" over "didn't take the trouble to ever find out and continued in ignorance."

Ah, yes, plausible result. I'd be ok with that too, but there are so many who have no idea what's plausible and merely take the first result. And then only half-read it. I think some people have selective literacy - they can read when it's something to do with entertainment or social media, but otherwise seem remarkably impervious to the informative properties of words.
posted by Athanassiel at 11:34 PM on November 2, 2013 [5 favorites]

From The Problem of Automation: Inappropriate Feedback and Interaction, not Over-Automation (PDF) by Donald Norman, 1990:

Consider the task of the crew on a modern commercial airplane. Most of the flight activity is routine. Large, modern aircraft are relatively easy to fly: the airplane is stable, responsive, and maneuverable. The automatic equipment monitors all operations and helps ease the workload of the crew. Indeed, whereas the commercial airplane of a few years ago required a crew of three, the newer planes need only two people to fly them, and most of the time, only one is really necessary. Most of this is good, and the accident rate with modern aircraft has been decreasing over the years, the decrease highly correlated with (and usually thought to be a result of) the introduction of high-technology controls and automation.

There are problems, however. For one, the sheer size of the plane means that the crew cannot know all that is happening. They are physically isolated from the passengers and any difficulties that may be occurring within the passenger section of the plane. They are isolated from most of the physical structures of the aircraft.

Even more important than physical isolation is the mental isolation caused by the nature of the controls. The automation tends to isolate the crew from the operations of the aircraft because the automatic equipment monitors and controls the aircraft, providing little or no trace of its operations to the crew, isolating them from the moment-to-moment activities of the aircraft and of the controls.

On the one hand, this combination of relative physical and mental isolation from the details of flying helps contribute to the safety by reducing workload and reliance on possible human variability or failure. On the other hand, when the automatic equipment fails, the crew's relative isolation can dramatically increase the difficulties and the magnitude of the problem faced in diagnosing the situation and determining the appropriate course of action.

Eyeballs on a screen and fingers on a keyboard are one thing, but an ass in the seat (that wants to get safely back on the ground) is quite another. Operators need enough hands-on experience to know when to use automation and when to turn it off.
posted by cenoxo at 11:44 PM on November 2, 2013 [8 favorites]

What a load of luddite shit.

Amen. His first two examples (which I couldn't get beyond) are of situations where the automation did exactly the right thing, yet the pilots chose to ignore the warnings. The problem here is not the automation, but the training of the pilots and their interactions with these systems.

Incidentally, Atul Gawande's Checklist Manifesto is an interesting book that deals with how humans interact with complex systems. If I recall correctly, it deals explicitly with Air France 447.
posted by kdar at 1:30 AM on November 3, 2013 [2 favorites]

I'll take "looked it up on Google and accepted the first plausible result they found" over "didn't take the trouble to ever find out and continued in ignorance."

But the problem is often a) they are searching in effectively, b) doing no evaluation, c) failing to understand what they find, and d) not integrating what they do find in any meaningful way, instead grabbing a "key phrase" or two whether or not they understand it.

Now, if the question is "Who was the King of Norway in 1437," that's probably sufficient. But, if the question is something more complex or significant, the same process is used, and that's not good.

It's not that I think automation is bad; it's often quite good. But the main motive for automating is more often commercial advantage than anything at the individual level. The two may intersect, but that's not the driving force. Google, after all, doesn't want to provide information, it wants to sell ads.
posted by GenjiandProust at 2:59 AM on November 3, 2013 [4 favorites]

I work in a highly automated library. There are no cash registers (this is to prevent embezzlement and robberies and wasting staff time on counting out all the coins and filling out the cash register forms every night; this was pretty time-consuming when I worked in a library that had cash registers). Instead, if you want to print out a paper or pay a fine, you go to an automatic kiosk and add money to your card.

The problem is, when it stops working--which is frequently--what can we say? "Well, you can pay your fines online if you have a credit card." (This is not helpful if you're an 11-or-12-year-old latchkey kid.) If you lose money in the machine, we can't give you a refund right there; we have to fill out a form and send it up to administration. I think it's very frustrating for people to know that a thing that should be simple has been made into an automated process where if things go wrong, there's nothing the staff can do to fix it without calling up IT and waiting who knows how long for them to fix it.

Obviously the solutions are better technology or better-trained staff, and there's not really the money for either. The solution to bad automation isn't no automation, it's better automation. But it does feel to me like a little parable about how shifting a low-tech system to a high-tech system can result in people being less empowered when things go wrong.
posted by Jeanne at 3:22 AM on November 3, 2013 [14 favorites]

His first two examples (which I couldn't get beyond) are of situations where the automation did exactly the right thing, yet the pilots chose to ignore the warnings. The problem here is not the automation, but the training of the pilots and their interactions with these systems.

Wow, that's exactly his point too! You are, of course, under no obligation to read the article but to excoriate the author for making the exact same point you are making seems a little odd.
posted by Athanassiel at 3:53 AM on November 3, 2013 [2 favorites]

I'll replace long boring diatribe about human-in-the-loop spaceflight with a link to the book Digital Apollo, which I haven't read yet.

But automation based on physical principles is easily reasoned about. It's the automation of trust itself -- e.g. computer security, banking, electioneering, mass communications -- that bothers me. Computers can propagate B.S. faster than a group of humans can detect and block it. It seems like we're trading a lot of localized failure modes (guy down the street doesn't pay you back) for a handful of massive ones (collapse of global banking system).

Also, when you are driving a stick shift it's much less likely you'll be texting.
posted by RobotVoodooPower at 4:32 AM on November 3, 2013 [4 favorites]

Jeanne's anecdote is not so dissimilar to the Air France disaster when you get right down to it. The issue is systems designed without a sufficiently holistic view of the environment in which those systems will be deployed. I'll trust fully automated aircraft when we fully understand, and can fully image in real-time, atmospheric fluid dynamics.
posted by Z. Aurelius Fraught at 4:39 AM on November 3, 2013 [1 favorite]

"Fully image in real-time, atmospheric fluid dynamics"? What does that even mean?
posted by indubitable at 5:46 AM on November 3, 2013

I came here to post Bainbridge's Ironies of Automation, but anthill beat me to it. So I'll post another great article (pdf): Ten Challenges for Making Automation a “Team Player” in Joint Human-Agent Activity

Carr mentions complacency as a problem with automation, but this is a common misunderstanding. If you have a source of support that has demonstrated sufficient reliability, and you have demands which make doing the task manually too costly, then it makes sense to rely on your source of support. We do this all the time - it is only when something goes wrong is there an accusation of over-reliance or complacency.

This is not to say that automation is always correct and should always be followed. The problem is that automated systems typically do not help the operators to understand when the systems have entered a state where their reliability has not been established. The autopilot does not tell the pilot "I've not been tested for this combination of weather factors - you need to keep an eye on me." The diagnostic support system does not tell the physician "My knowledge-base does not include the new bacterial infection that has been spreading in the hospital". Google doesn't say "I don't really understand your query, and could only find a few sites that remotely to what I suspect you are getting at."

It is not a question of abandoning this sort of technology, nor of dismissing the legitimate problems that critics bring up. It is about how to design these systems to support interaction better - including for situations when the automation can't do it alone.
posted by neutralmojo at 6:00 AM on November 3, 2013 [3 favorites]

We're All Luddites Now.

With regards to their own livelihoods working on a loom, the Luddites were right on the money. With a data-based economy working at a screen, are we prepared for a similar future that can change at the flick of a chip?
posted by cenoxo at 6:09 AM on November 3, 2013 [4 favorites]

Collaboration with computers is the name of the game. I'm literally mouthing platitudes when I say this, as others have pointed out, but the task isn't to automate humans out of the picture, it's to enhance human agency by finding ways for automation and computing to work with humans.

The flip side is that people who are willing and able to collaborate with computers will outperform people who don't. Most people on Metafilter already know this: if you're smart and generally well-informed, Google can transform you into a formidable proponent on matters you've just now discovered merely by giving you the facts and the theories to apply to your particular case.

Of course, as we also know well here on Metafilter, that's no substitute for expertise, but expertise is now a matter of doing that same computer-assisted investigation over and over again until you've hammered out most of the Dunning-Krueger problems.

Millennia ago we formed similar symbiotic relationships with dogs: training each other to be more effective as a unit than we were separately. Google reminds me that the term for this is co-evolution. :-)
posted by anotherpanacea at 6:09 AM on November 3, 2013 [5 favorites]

"Fully image in real-time, atmospheric fluid dynamics"? What does that even mean?

Trying to discern which way the wind is blowing.
posted by cenoxo at 6:13 AM on November 3, 2013 [1 favorite]

Trying to discern which way the wind is blowing.

Commercial airliners already do that, though.
posted by indubitable at 6:28 AM on November 3, 2013

I'll trust fully automated aircraft when we fully understand, and can fully image in real-time, atmospheric fluid dynamics.

Can human operators do this?

(That's an honest question. I'm not convinced that there are any situations where I'd want a human pilot over an autopilot, but I don't know enough to say for sure).
posted by graphnerd at 6:44 AM on November 3, 2013

But it does feel to me like a little parable about how shifting a low-tech system to a high-tech system can result in people being less empowered when things go wrong.

Isn't this actually a story about how whoever manages your library should've kept a cash register or two operational, or empowered staff to issue refunds as they deem appropriate? Having a hard time pinning this one on the computers.
posted by downing street memo at 6:53 AM on November 3, 2013 [1 favorite]

Pilot Executes 'Textbook' Crash Landing After Aircraft Loses Power On Student's First Training Flight

Heathrow plane crash pilot 'lost all power'

Black box confirms Hudson plane lost power in both engines

List of airline flights that required gliding

...just to name a few instances where human experience and presence was irreplaceable.
posted by cenoxo at 6:59 AM on November 3, 2013 [2 favorites]

These supposed "digital natives" are really only adept at doing things that companies want them to do, and that's a bit sobering.

As a working school IT technician too old to be a digital native but fortunate enough to grow up alongside the technology (the Apple II happened when I was 15), I see far fewer young people interested in messing with what's under the hood than I saw when I was their age. Partly this is because the tech is now so complex as to have become, to its users, indistinguishable from magic per Arthur C Clarke. Partly it's because the most ubiquitous tech is deliberately rendered tamper-resistant by its manufacturers. But quite a lot of it does seem to go hand in hand with a general dulling of curiosity about the natural world - it's widely known amongst the young that Google can instantly provide all "the answers", so there's less incentive to sit and think about things.

As an interesting exercise, offer your nearest 15 year old a $100 monetary reward for giving you a correct and easily grasped physical explanation, using only a lamp bulb and a basketball for props, for why Los Angeles should be expected to experience a wider range of seasonal temperatures than Panama.

The young'uns are going to have all that space in their brains that's not taken up by lists of kings and the times tables and cursive writing - what on earth are they going to do with it all?

Take selfies at funerals between visits to the unemployment benefits office?
posted by flabdablet at 7:33 AM on November 3, 2013 [10 favorites]

The flip side is that people who are willing and able to collaborate with computers will outperform people who don't. Most people on Metafilter already know this: if you're smart and generally well-informed, Google can transform you into a formidable proponent on matters you've just now discovered merely by giving you the facts and the theories to apply to your particular case.

Or, conversely, gives you the illusion that you are a formidable proponent, allowing you to neatly ignore inconvenient information from actual experts. Additionally, chances are you are considerably less adept at searching, evaluating, and comprehending information than you think, so you are already behind, unable to improve the skills you desperately need. And this "you" covers a very broad range -- STEM people are especially prone to imagining that their specific skill set will work as a sort of broad competence, as are MBAs.

Of course, as we also know well here on Metafilter, that's no substitute for expertise, but expertise is now a matter of doing that same computer-assisted investigation over and over again until you've hammered out most of the Dunning-Krueger problems.

Except, there is no "over and over again." Most people run one search, look at a result or two, and call it a day. And, if I understand Dunning-Krueger correctly, it predicts this -- you are unlikely to recognize your own incomprehension, so, rather than seeking further information, you stop at one or two sources. This is no way to master a complex skill or concept, although it is an adequate tactic for finding what movies are showing at the mall or making an online purchase.
posted by GenjiandProust at 7:40 AM on November 3, 2013 [6 favorites]

Yeah. But on balance, how do those incidents compare to incidents of straight-up human error?
posted by graphnerd at 7:46 AM on November 3, 2013 [1 favorite]

Defenestrator - was that at the FDDRL? Some fascinating documents on the web site.

The thing about going from two-pilot to single-pilot operations is that you have to deal with incapacitation. Another human in the cockpit can spot this - which includes a broad range of events, from experiencing mental problems that produce bad decisions to keel-over-dead - and, of course, deal with it.

In a single-pilot flightdeck, that can't happen. So you have to have the automation (and support systems) able to cope with detecting and dealing... and some of those situations will be a zero-pilot flightdeck.

That has to be part of the single-pilot automation. So, why not go straight to zero? From a purely technical point of view, I think that's a strong argument; from a real-world point of view, I think it'll be too much, too soon.

But single-pilot is interesting enough. I love the concepts of being able to rope in pilots on other aircraft to help out with problems, and the pool of very skilled people on the ground also available to become virtual co-pilots when required. In a lot of situations that would already have helped; the establishing of a protocol for the aircraft to sound an alarm back at base when it thinks that something's gone wrong and for someone to then call in and say "What's up?" while having a full view of flight data - well, there are a number of recent accidents due to pilot error where the aircraft really did 'know' that things were going wrong, but the pilots didn't respond to the local situation appropriately.
posted by Devonian at 7:47 AM on November 3, 2013

Back when I was an engineer, and before I wound up on too many DoD contracts for me to be comfortable with what I did, I designed fuzzy control systems for the eventuality of automated slaughterhouses. Think about it: here is a thankless first generation imigrant labor job, which had an injury rate of slightly less than 25%, where the conditions are barely bearable.

Error in these cases means someone loses a finger, accidentally slices a major artery in their arm or leg (the conditions are cold enough and the knives are sharp enough that meat feels like meat). I don't know if the meat production client was more concerned about the cost of the employe, our are the loss of the line time and (potentially) a whole side of an animal in the event of cross-contamination... Regardless, there was downtime, a loss of product, and a legal cost of associated with an injury. So here I am, writing code, effectively to control the cost of meat by replacing a worker. Part of doing this was being instructed, by a master butcher, on how to navigate around bone, what the difference between a tendon and muscle felt like, what cartlidge felt like, when we were supposed to cut through certain parts, how to maximize yield, and how to minimize botching up a piece of meat.

I would take the knowledge, distil it to measureable parameters, and watch as the robot would fail to live up to the mastery I had witnessed. Like Kasparov though, both the master butcher, and I as an onlooker knew that it was only a matter of time before someone was replaced by a future version of this machine. When that happened, the knowledge he had would be removed from the general consciousness, and would fall keenly onto artisans and specialists. It makes us 'dumber' in the sense that it becomes a skill fewer people have, a seemingly simple low-skill job. Erode enough of those low skill jobs, and the opportunity to succeed in any fashion as an unskilled laborer moves more and more to service, equally diluting some other profession as well. We push folks with less skill, ability, aptitude, and/or drive into fewer and fewer jobs. The quality of life for the poor becomes worse.

Moreover, the same things apply to middle class jobs like high end tech support, accountancy, automated IT roles, minute clinics, and so on and so on...

As a people, yes, we no longer have to think about some task, but the consequence is real as the division from those.with an unautomated (yet) skill are still fairly compensated, while those who aren't are put in an increasingly difficult position. Yes, society improves with technological advancement, but some of is doesn't. And right now, we collectively need to be figuring out how to solve that problem before it becomes too big... Because it will... That is progress..
posted by Nanukthedog at 8:10 AM on November 3, 2013 [10 favorites]

"Fully image in real-time, atmospheric fluid dynamics"? What does that even mean?
posted by indubitable at 8:46 AM on November 3
[+] [!]

I presume it's a reference to Direct Numerical Simulation.
posted by oceanjesse at 8:15 AM on November 3, 2013

Anyway, to try to bring my comments back to the article, I think there is a push to use technology to solve problems that are (at least for the moment) more effectively solved by better training of humans. So we make internet searching appear simpler, and, since most people only need "good enough" information, it works pretty well. Except, when it doesn't, we don't know how to fix the problem. We automate cars or planes, and it will likely work well enough most of the time, except, when it doesn't, there won't be someone with the skill to fix the problem. And, even if the incidence is low, that's cold comfort when it's your problem that's currently unfixable.
posted by GenjiandProust at 8:16 AM on November 3, 2013 [4 favorites]

Why are we trying to generalize about this? Is it too expensive to judge particular technologies by their merits?
posted by LogicalDash at 8:17 AM on November 3, 2013 [2 favorites]

It’s amazing how successful the Capitalist story has been, and for how many people "the most efficient way in the short term is the best way" is never even questioned, some don’t even seem to understand that there is anything else to consider. What is the goal?

We automate cars or planes, and it will likely work well enough most of the time, except, when it doesn't, there won't be someone with the skill to fix the problem.

The Machine Stops, Dancers at the End of Time, Pump Six, and many more. Yes, people have been saying this for a long time. It’s weird to me that some use that as some sort of indication that the idea can’t be true.
posted by bongo_x at 8:28 AM on November 3, 2013 [1 favorite]

When the hell did every aspect of human life come to be about "getting the best results" and doing things the most efficiently for some nonhuman system's definition of the goals? I seriously feel we seem to have too little regard for the quality of life experience and the value of living. It's like, if we could automate the need for human beings out of the world completely, we'd do it in an instant. What are we supposed to do with our lives once there's nothibg left for us to do? Eat, fuck and poop? Maybe play at some activities we all know don't really matter in any practical sense? Who are we designing the future for? And why do we seem so intent on making ourselves completely irrelevant?

As far as autopilot systems go, fine, they're probably a good thing. But how long before it's not just autotuned voices singing all our songs but human voice simulating algorithms? How long before it just makes economic sense to replace all our actors with computer simulations? Then hell, why stop there? Surely it would be more efficient if the audiences were bots too? Life is not about efficiency. Living systems are inherently inefficient. If our only love is efficiency, we will only come to hate life with all its magnificently pointless art more and more over time.
posted by saulgoodman at 8:54 AM on November 3, 2013 [12 favorites]

Glad we've established, once again, that this is all the fault of us millennials for being incurious, inferior louts. Because, y'know. Anecdotes.
posted by Solon and Thanks at 9:13 AM on November 3, 2013 [3 favorites]

We automate cars or planes, and it will likely work well enough most of the time, except, when it doesn't, there won't be someone with the skill to fix the problem

This to me seems like a trivial objection given that every piece of evidence we have about increased automation in transportation systems - autopilot, automatic train control, the self-driving car - indicates that these technologies are vastly safer than human-only operation. Like, orders of magnitude safer.

I mean, if you're worried about people feeling bad because they've lost the illusion of control, then fine - but I'll take dramatically-reduced highway deaths any day.
posted by downing street memo at 9:18 AM on November 3, 2013 [1 favorite]

But how long before it's not just autotuned voices singing all our songs but human voice simulating algorithms?

Who cares? A good song is a good song.

How long before it just makes economic sense to replace all our actors with computer simulations?

Have you ever watched an animated film? There is functionally no difference between that and what you're describing here.

Then hell, why stop there? Surely it would be more efficient if the audiences were bots too?

Where will the bots get money to pay for the show? At the end of the day, an artistic production has to have some mechanism of paying for itself, so human beings will always be part of the feedback loop.

Life is not about efficiency. Living systems are inherently inefficient. If our only love is efficiency, we will only come to hate life with all its magnificently pointless art more and more over time.

This feels like a first-world concern if I ever heard one. Large parts of the world would lead dramatically-better lives with better, more efficient transportation systems (to name one thing that could be better with improved efficiency)
posted by downing street memo at 9:24 AM on November 3, 2013 [2 favorites]

Interesting article, it's a subject I've thought about often lately, though not in the same snap-decision/life-or-death scenarios like the autopilot anecdotes which draws the eye but buries the lede. Many tasks, as noted here, are fine for automation.

Others, that traditionally require intensive training, such as -- to pick an autopilot-level example -- surveillance, are now being entrusted to the blunt instrument of mass data collection in the belief that whatever information is being sought will be caught by "the system" while the majority of its human operators require less and less education. As well-noted above, both skilled humans and data accessibility are required for many sophisticated tasks, but we're becoming unbalanced toward the latter. The long-range implications for us are frightening.
posted by ipe at 9:32 AM on November 3, 2013 [1 favorite]

Surely, if machine-made music displaces the artist in thousands of instances, the incentive for any individual to improve his talent--so necessary in all art--is minimized...
posted by The Underpants Monster at 10:31 AM on November 3, 2013 [2 favorites]

I see far fewer young people interested in messing with what's under the hood than I saw when I was their age.

I actually don't think this is true.

I think what you're seeing is that, while in the early 80s, the ONLY people who used computers had to be interested in poking around under the hood, today, basically everyone needs to be able to use computers, but virtually none of those people need to know how to build a computer or need to be interested in the technical aspects of their computers.

I'm a digital native. I'm reasonably techy in the sense that I know how to operate a lot of kinds of computer-based technology, and I can learn new systems easily, and even troubleshoot minor problems. I did some low-level programming in school and could probably learn that if I needed to. Back in the olden times when my family had an IBM running Windows 3.1 which we bought at CompUSA, I was reasonably skilled at opening up the computer and doing things to the hardware. I'm also quite good at hardware issues in printers, copiers, and other common office machines that tend to fail a lot.

But I have NO FUCKING DESIRE to build a computer for fun, or to take the computer I have apart and see how it works, or really do anything that even vaguely resembles an engineering skillset. I'm a words and ideas and people person. I don't actually enjoy dealing with hardware. I love the fact that, in 2013, unless there's a paper jam in the copier, I don't ever have to deal with hardware. Everything just works. I buy a laptop or a phone or a tablet. I turn it on. It just works. It's unlikely to ever have a hardware failure, and if it does, there are people who tangle with that stuff for a living. And all of the above is very affordable, accessible, and easier than taking my car to the mechanic or unclogging a toilet.

I get that the hardware focused engineer people feel like their little club got hijacked by grandma facebooking pictures of the grandkids or whatever, but I'd appreciate not being deemed some kind of sheeple just because I enjoy owning objects that do what they're supposed to do right out of the box without having to be dissected on a weekly basis.
posted by Sara C. at 11:53 AM on November 3, 2013 [2 favorites]

The quality of life for the poor becomes worse.

But it seems to me that this is because of raw capitalist evil, not because of technology.

On the face of it, replacing a human slaughterhouse worker with a machine eliminates HUGE amounts of potential human injury, disability, and death from the picture. Suddenly that poor immigrant worker has a 100% chance of keeping all his fingers. Which is obviously a good thing for everyone involved.

The problem comes when the poor finger-retaining immigrant looks for her next job. While we're pushing unskilled labor into the service industry, we're not paying service industry workers a living wage.

A world where you keep all your fingers, do something vaguely not terrible for a living like the proverbial burger-flipping, AND get paid a living wage for it would be a win for humanity across the board.

But we won't. And it's not for any practical reason like it would make the burgers too expensive, thus destroying the economy the poor immigrant depends on. It's to put a few pennies in a CEO's pocket.

If we had a reasonable minimum wage pegged to actual living expenses in the US, all the problems of mechanization would be solved. Get the human beings with their delicate fingers out of slaughterhouses and get them into burger joints or t-shirt emporiums or nail salons, but pay them the same wage, and suddenly this problem vanishes.
posted by Sara C. at 12:05 PM on November 3, 2013 [4 favorites]

(That said, I've always wondered about the occupational hazards of nail salons. Constant exposure to all those chemicals can't be great.)
posted by Sara C. at 12:06 PM on November 3, 2013

John Henry was a steel drivin man.
John Henry had a hammer in his hand.
posted by charlie don't surf at 12:17 PM on November 3, 2013 [1 favorite]

But how long before it's not just autotuned voices singing all our songs but human voice simulating algorithms?

oh, about 10 years ago
posted by pyramid termite at 12:52 PM on November 3, 2013

What a load of luddite shit.

That's not what's the Luddites were about:

Despite their modern reputation, the original Luddites were neither opposed to technology nor inept at using it. Many were highly skilled machine operators in the textile industry. Nor was the technology they attacked particularly new. Moreover, the idea of smashing machines as a form of industrial protest did not begin or end with them.

posted by sneebler at 2:10 PM on November 3, 2013

I'm not a luddite. I'm an electronic musician and computer programmer. Simplifications are silly.
posted by saulgoodman at 2:40 PM on November 3, 2013 [1 favorite]

One of the problem with automation in high-profile situations is the tendency to blame the machine for any failure that occurs when it's in charge, when we generally are much less harsh when people are in charge. It's acceptable to put down a failure when a person is at the controls to "human error," but if a failure happens when under automated control, the machine must necessarily be defective.

This is sort of a backhanded compliment to mechanization, because it implies that the automated system can always be improved, always made more perfect, while people are inherently fallible and at some point are just as good as they're ever going to get and we just have to learn to live with that. And we do live with it, tolerating failure far more often than we probably should.

E.g., people are really shitty at controlling automobiles. Like tremendously bad, to the point where I fully expect that future generations will be astounded that we ever allowed people to manually pilot huge hunks of steel down the public roads at many tens of miles per hour. And yet, rarely after a car crash does a team of engineers do a root-cause analysis in order to ensure that whatever caused the crash is remediated. Most of the time, we just shrug, fill out some paperwork, hose the blood off the pavement, tow the wrecked machines to the scrapyard, and continue onwards. Only if there is a huge pattern of identical "human error" failures is there even an attempt to fix the issue. (Think of how long we've had the ability to put impending-collision alarms into the rear bumpers of cars, and how many people have probably be run over in that time, and yet such relatively trivial devices are still not mandatory.) But when the first generation of fully-automated cars get on the road, the first time one of them gets into a serious accident, there will probably be an NTSB investigation. Engineering studies. Lawsuits and finger-pointing galore. And, eventually, there will be new code written or sensors added, and the next generation of cars won't be subject to that particular failure case.

And so the machines will get better with each design iteration, in a way that people don't, and the lessons learned from one failure can be used to prevent similar failures much more effectively. In many situations where an instant response is required, that almost certainly outweighs the sole human advantage of creative thought and problem-solving. Coming up with a creative, out-of-the-box solution is great, when the time exists to do so. People are good at that, and it seems unlikely (the realm of hard AI) that machines will encroach on it in the foreseeable future. But when it comes to thinking quickly and making snap decisions, machines will eventually always win.
posted by Kadin2048 at 2:49 PM on November 3, 2013 [4 favorites]

a seemingly simple low-skill job

Having had a crack at quite a few of these, and having seen so many of them so thoroughly botched by people who just don't give a shit about what they're doing, I am no longer convinced that any such job exists. Everything that people get paid to do, and most of the things we don't, involves improvable skills. Everything.

It's like, if we could automate the need for human beings out of the world completely, we'd do it in an instant. What are we supposed to do with our lives once there's nothing left for us to do?

In theory we're supposed to live as philosopher kings, with lives of leisure and personal fulfillment and an occasional jaunt over Eden in our flying cars. Those of us who ate the red pill understand that we're supposed to sign ourselves into bonded servitude to our local 1%er in order to secure a deemed-adequate supply of kibble for ourselves and our children.

Who are we designing the future for?

Most of us are not designing the future. Most of us are working for some CEO or other who is busy desiging it for him (it usually is a him) and others like him that he wants to impress.

And why do we seem so intent on making ourselves completely irrelevant?

Again, most of us are intent on no such thing; we're merely allowing our employers to make us increasingly disposable because we have been convinced by a relentless propaganda machine that Efficiency In Business automatically leads to Lower Prices For The Consumer, an absolute good.
posted by flabdablet at 4:12 PM on November 3, 2013 [6 favorites]

Me: What a load of luddite shit.

sneebler: That's not what's the Luddites were about:

That's not what the adjective "luddite" means.
posted by IAmBroom at 4:28 PM on November 3, 2013

No, and that's unfortunate. Yet again a word that should label a quite nuanced idea has been rendered useless by deliberate misuse as propaganda. See also 'green', 'genetically modified', 'clean', 'renewable', 'economic', 'theory', 'climate', 'fascist'.
posted by flabdablet at 6:16 PM on November 3, 2013

I get that the hardware focused engineer people feel like their little club got hijacked by grandma facebooking pictures of the grandkids or whatever

It's not about preserving the club thing for me. I do occasionally get accused of having some kind of high-priest attitude toward technology because I do object to stuff getting dumbed down to the point where it becomes manifestly less useful (Windows 8, I'm looking at you!) but as one who "tangles with that stuff for a living", I look at the rising generation of those who will be expected to do my job once I'm too old and tired to do it myself, and feel sad for the people I'm currently supporting.

Sure, it's nice when equipment "just works" and of course I prefer that my own equipment does exactly that. But there's a non-obvious downside to the pervasiveness of equipment that "just works", which is that it might as well work by magic.

Once all that's readily available is inscrutable tech that "just works" for some well-defined range of designer-envisioned use cases, the average person's chance of finding workable ways to engage with their own potentially tech-addressable problems gets lower and lower each year. And if that isn't a recipe for top-down control and social stagnation I don't know what is.

And the problem doesn't stop with "average people". Tech is already so complex that even the putative high priesthood is not capable of acquiring sufficient real clue about it any more. More and more technology problems in the 21st century get tackled by the following method: hire a huge team, throw money at it, and if they can't quite make it work well enough, release it anyway and wait for better hardware. And what we end up is insanely convoluted, fragile and costly ways to do many of the things that were already being done adequately n 1980 - but they look prettier, their corners are all nicely rounded, and Google can index and appropriate every! fucking! aspect! of what they do.

I'm 51 years old now, so obviously there's some element of nostalgia operating here. But I truly don't think that's all that's going on. Young people today - digital natives all - really truly are less curious about the world they live in than I was, and my peers were, at their age; they're also much less likely to question the necessity of the wonderful toys that huge corporations provide for them to play with.

At 15, I was very fond of my scientific calculator and my Kodak Instamatic camera and my portable cassette player. But the idea of going somewhere without any of these devices caused me no anxiety at all. Ask a 15 year old today to go somewhere without their phone and you might as well be asking them to go without their hands or eyes.

And of course Facebook is where you keep your life now. Why would you not? All your friends do. The fact that Facebook is a huge corporation whose primary purpose is mining your life for nuggets to sell to advertisers in order to increase the efficiency with which they can make you dissatisfied with what you already have? Totally off the radar.

The other day I told one of my kids the story of the kid who fell in a well and had to wait several days for rescue because nobody responded to her plaintive Facebook status updates, and she didn't see the point. Dialling 112 for emergency response would not have occurred to her either. She's a digital native, and she's just totally locked into what's normal. Facebook is normal. 112 is not.
posted by flabdablet at 6:58 PM on November 3, 2013 [5 favorites]

...shifting a low-tech system to a high-tech system can result in people being less empowered when things go wrong.

This. Given the problems we're having with just the climate, fer instance, what happens when things go wrong? Major storms aren't just an inconvenient now, they are life-threatening above and beyond the natural effects. The national power grid is especially vulnerable. It's bad enough when the dang computers shut down when you're trying to get gas.

I'm all for automation. But sometimes it's good to be able to fly by the seat of your pants.
posted by BlueHorse at 9:08 PM on November 3, 2013 [2 favorites]

"A good song is a good song."

Why?
posted by carping demon at 11:13 PM on November 3, 2013 [1 favorite]

Glad we've established, once again, that this is all the fault of us millennials for being incurious, inferior louts. Because, y'know. Anecdotes.

I never said, and do not believe, that being a millennial means being an incurious, inferior lout.

What I do believe is that if social conditions add ubiquitous distraction to the lives of a given class of people, then the aggregate amount of curiosity expressed by those people, and the amount of time those people will end up willing to devote to dealing with difficult problems, has to decline.

This is not an inferiority thing or a lout thing. This is a human thing, and a straight time allocation thing; an inevitable consequence of a technology-mediated shift toward structured, scripted, planned, supervised and monitored activity at the expense of unstructured discovery and play and contemplation. It's exactly the same effect as has caused a decrease in the likelihood of encountering a young person who habitually chooses to read the paperback rather than wait for the movie to come out.

And of course if you're a millennial and you're reading this kind of stuff from a 51-year-old, you're going to be thinking that he's just another clueless old hippie and where does this useless arrogant retard who doesn't even own a smart phone or have a Facebook account think he gets off telling us how we're all doing it wrong? I bet his car and his clothes and where he lives and the way he wears his hair are all just a total and complete embarrassment and he doesn't even know it, the poor dopey senile fuckwit gimp.

But here's the thing: I grew up in a home without a television, and I never, ever, ever get bored. In my half century on this planet I have been bored infrequently enough that I now have only the vaguest memory of what being bored even feels like. Many of my friends are the same way. But I have never, not even once, met a digital native who could honestly make such a claim.

That, it seems to me, means that something important has indeed been traded away.
posted by flabdablet at 11:55 PM on November 3, 2013

Flabdablet, I think, if you were to survey people of your same age, you might actually find people who did get bored! Its very easy to look at oneself and wonder why kids aren't like that, but we are all exceptional, and doing this is usually a mistake.

I agree that it is a shame that some tools are closed boxes that just work, but the problem with that is the new generation of hack kids won't be hacking those devices. They're out there, almost ce4rtainly in higher quantites than before, but probably a lower proportion. The majority of the population will never be technically adept in the way you wish them to.

The reason automation is so good is because computers are so much more reliable than humans for most tasks one can think of. I have no doubt that when it is perfected, self driving cars will be orders of magnitude more safe than current cars. We will look back to the thousands killed every year, almost always by driver error, with utter horror. And yes, those who can drive will curse those who don't really know what they're doing, and there will be the occasional story where a machine shut off and the driver crashed into a wall. And there will be horror, and the occasional voice of sanity pointing out its one death compared to thousands...

I am one of the people who will just look up the answer to questions on my phone. The internet is full of information. It has this wonderful website for instance! An interesting argument in a pub is which film produced in 1997 was best. An interesting argument is not whether a particular film was produced in 1997 or not. The latter is a question of fact, and one person is correct and the other is incorrect. There is nothing to be gained by people repeatedly insisting that their memories are better than the other.
posted by Cannon Fodder at 1:11 AM on November 4, 2013

Flabdablet, I think, if you were to survey people of your same age, you might actually find people who did get bored!

Sure, they still exist, and when I was 15 boredom was certainly more prevalent in my peer group than it is at 51. But it was certainly nowhere near as common then as now, and depriving us of our customary technologies didn't cause the almost allergy-grade boredom seen almost universally in digital natives similarly deprived.

Having "nothing to do" is now akin to an acute medical condition requiring instant application of soothing entertainment or, failing that, Facebook. Suggest to a bored 15 year old today that they go find a novel or a newspaper or just go out for an aimless walk with the dog or maybe try a little meditation or amuse themselves with spooky mirror faces in dim light and they look at you like you're 51.
posted by flabdablet at 1:35 AM on November 4, 2013

The majority of the population will never be technically adept in the way you wish them to.

I would never expect the majority of any population to become technically adept. What I mourn in regard to technical adeptness is the loss of both opportunity and motivation to develop it, the first due to modern technology's complexity and inscrutability, and the second due to the perception of the existing technology's enormous competence. Building your own toy Siri from pieces whose functions you comprehend is just too high of a bar to contemplate clearing as a beginner, and not worth doing anyway when Siri is already ubiquitous.

But my wider point is not about technical aptitude or adeptness; it's about the gradual dulling of human curiosity and adaptability in almost every field. I can definitely see this happening, I think the ready availability of pseudo-prosthetic technology has a great deal to do with it, and it makes me sad.

I am not surprised at all to find that digital natives are less saddened by this than I am; these are people who have spent their whole lives getting taught not to pay attention to anything that isn't put right in front of them, and it's hard to miss what you never had.
posted by flabdablet at 1:53 AM on November 4, 2013

(This is a derail, but the thread's mostly over.)

Do we have any statistics on what people do when they're on the facebook/checking email/fiddling with their phone? I spend a fairly high proportion of my facebook and email time exchanging and discussing articles. I probably get more of my subway reading off email and the facebook than Metafilter. And that wouldn't be possible unless there were a bunch of other people sending me links and commenting on my links. I suspect there's not as great a difference between the facebook and a newspaper as you imply.

Of course, that's not to deny that facebook notifications are exactly a variable-interval reinforcement schedule or that this is almost certainly a deliberate design decision to make people check the facebook more frequently, in a way newspapers or novels were never designed to make people buy another multiple times a day.

Nor is it to deny that young folk nowadays have this infuriating habit of checking their telephones while engaged in small group or even one-on-one conversations (?!). That, I think, is a real change, substantively different from spacing out without external aids. But I wonder how much that is a change in scheduling rather than overall time allocation. Are people really spending different proportions of their days reading and discussing things they read, or are they just scheduling the same amount of time in more, shorter chunks interspersed through other conversations?
posted by d. z. wang at 4:51 AM on November 4, 2013 [1 favorite]

We try our best to avoid it, but boredom has its benefits. Today, it's a lost art form. (April 2004).

Embrace the boredom, be the boredom. Turn off the infernal machines once in a while, give the overworked NSA a little less to do, and you may find there's never enough time to do all the nothing you want.

Oh, and everything flabdablet has said here.
posted by cenoxo at 4:53 AM on November 4, 2013 [2 favorites]

And of course if you're a millennial and you're reading this kind of stuff from a 51-year-old, you're going to be thinking that he's just another clueless old hippie and where does this useless arrogant retard who doesn't even own a smart phone or have a Facebook account think he gets off telling us how we're all doing it wrong? I bet his car and his clothes and where he lives and the way he wears his hair are all just a total and complete embarrassment and he doesn't even know it, the poor dopey senile fuckwit gimp.

But here's the thing: I grew up in a home without a television, and I never, ever, ever get bored. In my half century on this planet I have been bored infrequently enough that I now have only the vaguest memory of what being bored even feels like. Many of my friends are the same way. But I have never, not even once, met a digital native who could honestly make such a claim.

Those are extraordinarily offensive words to put in someone's mouth. I get that we're all sarcastic here - I was in the previous comment - but I think that's beyond the pale. I would NEVER use the word "retarded" in a derogatory sense, or use it as a slur, and seriously resent the implication that I'm someone who would. Or that I have something against 50-somethings (I have good friends in their 50s), or unfashionable people (I'm not super fashionable), or people who go off facebook (it's admirable.) It seems like you have some very specific negative images of what all young people are like, because you are painting with a very broad brush.

Anyway, as long as we're comparing credentials, I just spent two years in a rural village in the Peace Corps where me and many other 20-somethings lived without running water, televisions, and/or electricity in general. We found plenty of ways to entertain ourselves. So, hey, nice to meet you. I can't claim to have never been bored in my entire life. I've sat through some pretty tedious staff meetings. But I genuinely think that's an awfully unreasonable line to draw in the sand and people from MOST generations could not claim to never have known boredom.
posted by Solon and Thanks at 6:36 AM on November 4, 2013 [2 favorites]

Nice to meet you too.

Here, let me whip out the broad brush and offend you a little more.
posted by flabdablet at 7:09 AM on November 4, 2013

It's also possible to believe both that the benefits of boredom are overstated and that current consumer electronics technology has advanced to the point that it discourages the kind of tinkering that teaches problem-solving among people who don't already have that "I want to learn about computers!" goal in mind. Similarly, the increased complexity of automobiles has made learning about how to repair them more difficult and less accessible.

Boredom sucks. The best thing about miniaturization of electronics is that it means I can carry enough books for weeks in less space than I used to require for one of the books I carried for a single day's reading.
posted by asperity at 9:16 AM on November 4, 2013 [1 favorite]

it's about the gradual dulling of human curiosity and adaptability in almost every field.

Oh, man I so could not disagree with you more.

I think the internet is absolutely the best thing to come along, in my lifetime at least, if not in the last century, to inspire curiosity and spread information that previously was not available to people.

I remember all the things I used to wonder about as a kid and youngish teenager (pre ubiquitous internet), that were 100% fact-based things which would be easily googled in thirty seconds nowadays.

But back then, I had to get an adult to drive me to a library, and I had to talk to a reference librarian (if they were interested in engaging with a twelve year old who was inexplicably obsessed with the Tudor dynasty), and then if I was lucky maybe, after an afternoon of research, I could find something beyond the Encyclopedia Brittanica article I'd already read 20 times. And, again, this is all assuming every adult gatekeeper of this knowledge was 100% devoted to helping me find out what I wanted to know. Which was rare, even in libraries, in rural Louisiana in the 80s/90s.

So, I don't know, it's amazing to me that, in 2013, I can listen to a podcast about the causes of the First World War (just to mention something I actually did this weekend), and wonder about the current consensus on Gavrilo Princip and whether war was inevitable, or whether the assassination of Franz Ferdinand really changed the course of the 20th Century. And then I just google some keywords about that and find out what I want to know. And maybe I order a book on Amazon which can be electronically delivered to my Kindle, if five minutes with google doesn't answer my question. And now I'm a more informed person about a subject that interests me. Without leaving my house.

The usefulness of the internet as a sort of Curiosity Clearinghouse is amazing on a level that would have been inconceivable to people before the last few decades.

The insistence that tech has created a generation of passive consumers is especially silly when you consider how much better the internet is about this as compared to TV, radio, film, or even libraries built around collections of physical paper books.

I can know anything, find anything, research, fact-check, track down, etc. anything, using a slim rectangular object that fits in my pocket (OK, some things are easier done on my laptop, but even that is fucking UNREAL amazing compared to anything in 1992).

If my car is broken, I can look up how to fix the problem myself rather than bringing it to my mechanic.

If I want to learn how to do a new craft, and I don't already know someone who can physically teach me, I can find a youtube video with insanely detailed instructions, and which I can watch as many times as I need until I figure out the thing I'm stuck on without destroying a tape or wearing out a video player or even needing to return the video somewhere.

If I'm curious about some esoteric bit of knowledge that isn't currently popular in the wider media (for example my Gavrilo Princip question), I can spend as long as I want researching it, without ever running into a roadblock like a book being unavailable or a librarian not knowing of any additional resources. I have literally every single piece of scholarship ever done about this question at my fingertips.

If I'm tired of the music I've been listening to lately, I can endlessly dig into new stuff I'm not familiar with, completely unconnected with what's on the radio, MTV, or currently stocked by my local record store. I can find music that is completely uncommercial and/or unlike anything that I've ever heard before through traditional channels. If I see "King Sunny Ade" or "Oum Kaltoum" or "Souad Massi" referenced somewhere, I can be listening to their music within seconds despite the fact that these people live on different continents and play music in traditions I'm unfamiliar with.

But, yeah, sure, people don't break open their computers as often anymore and dissect them, so I guess the world is going to hell in a handbasket.
posted by Sara C. at 10:52 AM on November 4, 2013 [5 favorites]

what movie won Best Picture in 1977.

It was Rocky, btw.
posted by FJT at 11:54 AM on November 4, 2013 [1 favorite]

If my car is broken, I can look up how to fix the problem myself rather than bringing it to my mechanic.

Really? Even though your mechanic now requires a special computer diagnostic system to diagnose problems with most newer vehicles? I assume you've actually done this, like so many other people?
posted by saulgoodman at 1:29 PM on November 5, 2013

This is actually a really good example of the kind of thing some of us oldsters are fretting about: my grandfather was a natural-born mechanic who despite never having learned to read could practically rebuild a car engine blindfolded. And yet, when as a young adult, I started driving newer cars, anytime I had car trouble, he and others among the mechanically inclined in my family complained about how hard it was to work on newer cars. Why? Because many are designed now to be serviceable only with special, proprietary diagnostic tools you have to pay for access to, in many cases virtually guaranteeing you'll need to see a mechanic to get the job done. Even mechanics are now often dependent on those proprietary diagnostic tools and couldn't even do the diagnostic work without them as many of the subsystems in modern vehicles are software based. Were these industrial developments motivated solely by a desire to improve the quality of products and services, or were they motivated as much by a desire to create new revenue streams (in the form of diagnostic software licensing)? Well, a little from column A and a little from column B, maybe, but in the sense that consumers have less power over their economic choices as a result, it's probably not a net good for consumers in the final analysis as most of the performance gains from these newer systems are incremental at best. And the trade-off in terms of reduced user serviceability is significant.
posted by saulgoodman at 1:52 PM on November 5, 2013

Oops. Got my facts wrong. Since 1995, apparently, automobile diagnostic tools aren't allowed to be proprietary. Still, the tools are effectively a black box, so users don't really have to understand the problems being diagnosed, which is a big change from the way we've had to solve problems in the past. So I still think it's fair to question the potential long-term effects of reliance on these newer technology-enabled systems.
posted by saulgoodman at 2:00 PM on November 5, 2013 [1 favorite]

saulgoodman, you're generalizing quite badly.

I can fix my brakes without a computer system. I can fix my exhaust system. I can fix my steering...

I can't fix my fuel-air mix, engine timing, or a few other system-level features without a computer. But that's a far cry from "You're helpless without hi-tech gadgets!!!!"
posted by IAmBroom at 3:05 PM on November 5, 2013 [1 favorite]

Calvin sez: "I'm not dumb..."
posted by cenoxo at 4:31 PM on November 5, 2013

Really? Even though your mechanic now requires a special computer diagnostic system to diagnose problems with most newer vehicles? I assume you've actually done this, like so many other people?

I drive a '97, so, yeah, it is definitely possible to change my own wiper blades or check my transmission fluid.

I use google and youtube all the time to get a ballpark idea of what's up with my car, whether I need to take it to the mechanic, or whether it's something I can fix myself. Because I'm not very handy, I especially love the ability to find a video of someone else doing it (often on my exact make and model of car), which I can watch over and over until I figure out how it's supposed to work.
posted by Sara C. at 5:53 PM on November 5, 2013

I can't fix my fuel-air mix, engine timing, or a few other system-level features without a computer. But that's a far cry from "You're helpless without hi-tech gadgets!!!!"

Well, I'm not saying you're helpless to fix anything, really (even most tech designed not to be user serviceable can be user-serviced in limited ways, especially if you're willing to void the warranty). But consumer products come with many more features users can't be expected to fix or service themselves than in the past, and that's a trend I think hard data would support across a wider variety of consumer products. We may not be anywhere near Idiocracy levels of incomprehension about how our own tech works, but it's getting easier to imagine how we might end up there, and there are definitely trade-offs with the trend--particularly for poorer, rural people.
posted by saulgoodman at 5:59 AM on November 6, 2013

saulgoodman, that is a far more nuanced - and limited - complaint than your original one.

Still, I'll take increased fuel efficiency and lower failure rates over the freedom to spend my personal time fixing my car. YMMV (literally), and if so I suppose older used cars are the solution. After all, Cuba has 50yo cars driving around.
posted by IAmBroom at 10:36 AM on November 8, 2013 [1 favorite]