When NOT to take a stand



  • This is happening as I write this...

    When I first started working here, the main project had about 15-ish sub-projects. The "build" system was a bunch of javac/jar commands that people kept in a text file and cut-n-pasted into a command window in the proper sequence. And it was considered "good".

    I created a hierarchy of ant build scripts to automate and simplify things. I took all the copies of 3rd party jars and put them in a centralized lib project, and simply linked to them via an environment variable passed into ant. All you needed to do was to check out the projects in the same directory structure as in svn. It was considered "better". Everything worked.

    Until the New Guy came along.

    We gave him the instruction sheet on what to check out and how to lay it out, and how to get to the point of building and launching the application.

    He creates a directory called: C:\Users\xxx\MyStuff and puts everything (build files, projects, sub projects, the lib directory, the deploy staging area, unrelated *.docx and *.xlsx files - everything) in there. He managed to configure eclipse to point to the right places so that everything built. And he was coding away for a couple of weeks.

    However, when it came time to test his deployment, he just couldn't get the ant scripts to work. So he started moving directories (eg: lib, parent projects, etc) UP to be in the correct location relative to the project he was attempting to build. That is, instead of moving "common" from . to code/java/common and then referencing ../../lib/xxx.jar, he moved lib to C:\lib, but that broke other builds, and he just couldn't figure it out.

    Instead of just going back to the handy instructions, he sent out a flame mail that the build didn't work. 

    Thirteen people responded - to all - that the build did work, that it's more likely he put stuff in the wrong place, and he should refer to the setup instructions. He replied to each one - to all - that they were wrong and that it didn't work.

    Laughter abounded throughout the office. My boss (a pretty good natured guy) told us all to keep the conversation going and not to go over and show him what he did wrong - just to mess with him.

    Fun



  • Sounds like someone in a company I used to work for. Our central project we maintained was a web application and on his third day this guy insisted the whole project must've been screwed up all because he didn't have it configured properly.

    The punchline is the administrator actually set up his computer and IDE so it would run the project without any additional configuration on his part. He managed to completely tear down that setup, downgrade his to PHP4, install a bunch of packages he didn't need, and completely fubar his computer. When the administrator asked him why he did what he did the guy just replied, "I was trying to set my workspace up so that it reflected what I was used to in my previous job." He was really upset that he had to learn PHP5, as nobody had told him that during his interview. A month later, his desk was vacant. Management wouldn't tell us the circumstances, and given his demeanor, it was hard to tell if he quit in a huff or his supervisor had enough of his antics.



  • Problem Solving 101

    A process used by dozens has functioned flawlessly for years.  The new guy insists it doesn't work.  What is the cause of the problem and its solution?



  • This is just begging for someone to hire an impersonator to hand him his dismissal notice with one of the lines "The tribe has spoken" or "You are the weakest link. Goodbye."

    Bonus points if you can get the actual host.

     

     

     



  • @snoofle said:

    He creates a directory called: C:\Users\xxx\MyStuff and puts everything (build files, projects, sub projects, the lib directory, the deploy staging area, unrelated *.docx and *.xlsx files - everything) in there. He managed to configure eclipse to point to the right places so that everything built. And he was coding away for a couple of weeks.

    I'm confused as to how he could have possibly version controlled all of this unless he was very VERY VERRRRY careful not to add new files into the repo.



  • @Paddles said:

    This is just begging for someone to hire an impersonator to hand him his dismissal notice with one of the lines "The tribe has spoken" or "You are the weakest link. Goodbye."

    It is? Because I disagree.



  • @Soviut said:

    I disagree.

    That's a nice animation. Still like this one better though.



  • Awww, I love a nice feelgood story. 

    Especially when it comes from Snoof.



  •  @Soviut said:

    I'm confused as to how he could have possibly version controlled all of this unless he was very VERY VERRRRY careful not to add new files into the repo.

    Subversion lets you take any subproject and check it out anywhere you want. You can then change and commit it from that location and svn will put it back in the right place. So if you have ROOT, ROOT/A and ROOT/B in the repository, you can check them out to C:\ROOT, C:\ROOT\A and C:\ROOT\B, or C:\ROOT, C:\A and C:\B and you can still commit correctly.

    The problem was that the ant build scripts assumed a certain directory structure, so if you check out in the latter form, nothing is where the build scripts think it is, so nothing resolves.



  • @snoofle said:

    The problem was that the ant build scripts assumed a certain directory structure
     

    This describes my sordid experience with ant.

    But before you get all irritated with me: the assumed directory structure (inherited from a third party) assumed a certain directory structure involving network mapped drives like I: and P: which were specific to that supplier's network, not the end-customer's network.  What's worse is, due to certain regulations, even the build scripts had to match a particular hash so we could prove that the same environment was always used to generate the code.  The hard-coded paths made this impossible to do on any but the original network topology, resulting in all kinds of extra work (re-baselining of the hashes and build scripts) to hand the codebase off to a different team.

    Yeah, all kinds of  WTF arise from hard-coded directory structures....

    (/rant)



  • @too_many_usernames said:

    @snoofle said:

    The problem was that the ant build scripts assumed a certain directory structure
     

    This describes my sordid experience with ant.

    But before you get all irritated with me: the assumed directory structure (inherited from a third party) assumed a certain directory structure involving network mapped drives like I: and P: which were specific to that supplier's network, not the end-customer's network.  What's worse is, due to certain regulations, even the build scripts had to match a particular hash so we could prove that the same environment was always used to generate the code.  The hard-coded paths made this impossible to do on any but the original network topology, resulting in all kinds of extra work (re-baselining of the hashes and build scripts) to hand the codebase off to a different team.

    Yeah, all kinds of  WTF arise from hard-coded directory structures....

    (/rant)

    TRWTF is being held hostage to a VENDOR's topology. WTH? (Not at you, but at the vendor you described)



  • @too_many_usernames said:

    assumed a certain directory structure involving network mapped drives like I: and P: ... impossible to do on any but the original network topology
    So none of you knew about the SUBST command then? It's only been around since probably MS-DOS 2.0.



  • @derula said:

    @Soviut said:
    I disagree.

    That's a nice animation. Still like this one better though.

    Ugh. Over the years I've developed an intense hatred of Carmina Burana. Which is a shame because I liked it before every 2-bit crap movie decided to use it when then needed something "epic-sounding".

    Also Flash movies without pause or rewind or volume controls need to die, in a fire.



  • @blakeyrat said:

    Ugh. Over the years I've developed an intense hatred of Carmina Burana.

    Yeah, that's a bit sad. I still like the animation though, and the original voicetrack, which can be heard without o fortuna and in a pause- and seekable state here. The story behind this is that the guy who made that voicetrack is the same one who made the voices in the RPG parody game for which axeman13 wrote this review because he didn't get the joke.



  • @derula said:

    I still like the animation though, and the original voicetrack, which can be heard without o fortuna and in a pause- and seekable state here.

    ... why is there a smoking penis with a pistol on the header of that site? Am I tripping balls?



  • @blakeyrat said:

    ... why is there a smoking penis with a pistol on the header of that site? Am I tripping balls?

    It's not a penis, it's a popular meme on that site.



  • @Sir Twist said:

    So none of you knew about the SUBST command then
     

    This doesn't work very well when your internal network topology already uses those drive letters for something else...



  • @too_many_usernames said:

    @Sir Twist said:

    So none of you knew about the SUBST command then
     

    This doesn't work very well when your internal network topology already uses those drive letters for something else...

    Virtual machines? (That's what we ended up doing when we needed to supply an escrow service with our code+build 'instructions' - we simply supplied the code, the build tools and the OS in one image.)



    Granted that in our case no extra money was needed for any licences, and this may be a problem with Windows....



  • @PJH said:

    Virtual machines?
     

    It was a thought, but our customer (who needed to duplicate the build environment) has a hard enough time getting physical machines that setting up a virtual machine is...not a pleasant thought.  Trying to get them to purchase a virtual machine wasn't really in the books.

    In general, I think a virtual machine is a fairly overkill solution to poor file structure management.  (In the same vein I think that vehicle stability control is an overkill solution for people not willing or able to learn how to control their vehicles in handling maneuvers.)



  • And I think safety features on planes are overkill when both the manufacturers of the bare minimum components and the pilots should be incapable of making mistakes.



    I mean, seriously, you want ESC removed from cars? You troll, yea?



  • @nexekho said:

    I mean, seriously, you want ESC removed from cars? You troll, yea?

    ESC was designed to handle situations where you went "oh sht" and yanked the wheel over hard to the side to avoid obstacles, in vehicles whose body type was never designed to tolerate that, in the same vein as ABS was designed to handle situations where you went "oh sht" and slammed your foot on the brakes, when normally that'd lock the wheels. These technologies were designed to convert emergency, panicked reactions into something the car could handle to save the driver's life.

    However, then you get people who go "That them thur car don't stay flat on the ground with a pianer tied to the roof. I need them ESC thing there. Yup."

    ESC was never designed to handle stupid, and frankly [i]shouldn't[/i] be; the right tool for the piano job is a truck. And, according to too_many_usernames, the same is true for virtual machines: They have their place, but Silly Filesystem Shenanigans isn't it.



  • @TwelveBaud said:

    @nexekho said:
    I mean, seriously, you want ESC removed from cars? You troll, yea?

    ESC was designed to handle situations where you went "oh sht" and yanked the wheel over hard to the side to avoid obstacles, in vehicles whose body type was never designed to tolerate that, in the same vein as ABS was designed to handle situations where you went "oh sht" and slammed your foot on the brakes, when normally that'd lock the wheels. These technologies were designed to convert emergency, panicked reactions into something the car could handle to save the driver's life.

    However, then you get people who go "That them thur car don't stay flat on the ground with a pianer tied to the roof. I need them ESC thing there. Yup."

    ESC was never designed to handle stupid, and frankly shouldn't be; the right tool for the piano job is a truck. And, according to too_many_usernames, the same is true for virtual machines: They have their place, but Silly Filesystem Shenanigans isn't it.

     

    I agree with your analysis, except that isn't what he said:

    @too_many_usernames said:

    In the same vein I think that vehicle stability control is an overkill
    solution for people not willing or able to learn how to control their
    vehicles in handling maneuvers.

    I don't know what "handling maneuvers" is supposed to mean exactly, but it sounds a bit like the situations you describe (panic-mode attempts at evasion).  And from what you're saying, ESC and ABS are excellent things to have, because it means you can survive these situations without being a trained stunt driver.  Nobody said anything about moving pianos, and besides, the correct solution is to hire professional piano movers, thus abstracting away the problem.



  • @Justice said:

    I don't know what "handling maneuvers" is supposed to mean exactly, but it sounds a bit like the situations you describe (panic-mode attempts at evasion).  And from what you're saying, ESC and ABS are excellent things to have, because it means you can survive these situations without being a trained stunt driver.  Nobody said anything about moving pianos, and besides, the correct solution is to hire professional piano movers, thus abstracting away the problem.

     

    Two or three years ago, a deer lept nearly in front of my car around dusk coming over a hill.  I jammed on the brakes (though not so hard as to lock them up - which would have been quite difficult since I had ABS anyway) and swerved halfway off the road, still slowing but unable to "avoid" much more without careening down a rather lengthy, slippery grade and into some trees.  I almost got completely around him and only tagged him with the driver's side mirror, which promptly "folded" back in so hard it smacked the driver's side window which shattered.  Otherwise no damage. (I think the worst part was driving the remaining 20-25 minutes home at highway speeds with an open window in, IIRC, about 10 degrees Fahrenheit (that's -12 C?))  Two coworkers were following me (let's call them "Alice" and "Bob"), they stopped to check if there was anything they could do, but there really wasn't so they continued on.

    Anyway, the next day, me, Alice and Bob were talking about it and Alice said "Yeah, what were you even doing, trying to swerve and stuff?"  Saving me from having to say the obvious, Bob said "What the hell else are you going to do, just be stupid and slam it at full speed head on?"Alice replied "No, but you are just supposed to hit the brakes and not swerve.  Swerving is how most people end up hitting trees and dying."  Bob replied "Well, sure, but the key is to swerve WITHOUT hitting the trees.  Duh.  Rich looked like he did a fine job of that, and probably shaved a week off the time his car will need to be in the shop and saved his insurance company money too."

    Now, while I'm certainly not claiming to be some expert driver*, really, folks, this stuff isn't that hard.  

    My point is that Electronic Stability Control, anti-lock brakes - that whole mess of "driver's aids" - they're all great helps most of the time, but they really should never replace actually knowing how the hell to drive.  If you can't control a vehicle in a panic situation, then you truly shouldn't be driving either with or without ESC and whatnot. You need to get out in some kind farmer's harvested bean field (with permission) or on some slick empty parking lot and start learning how to handle a car sliding, spinning or fishtailing.  Preferably a car without ABS and without ESC.  

    *I've lurked around here enough to know that not only would you believe that I'm not an expert driver, you'll probably insist I couldn't possibly be smart enough to start the car in the first place.  I'm also sure that choice parts of my ancestry will be called into question at some point as well, either in this thread or another. 

     



  • @mahlerrd said:

    You need to get out in some kind farmer's harvested bean field (with permission) or on some slick empty parking lot and start learning how to handle a car sliding, spinning or fishtailing.  Preferably a car without ABS and without ESC.

    Actually, I believe it would be better to do this in a car with the same features as you'll be driving.  Learning how to pump breaks manually, for example, has a tendency to get one to do that in an emergency, even when you have ABS.  Even average ABS systems do a better job than a human being can possibly do.  Pumping the breaks with such a system merely provides relatively massive gaps when there's no breaking at all.  The difference is even worse in some air brake systems, where pumping the breaks can actually defeat the ABS system, causing the breaks to lock.

    I haven't seen the stats on ESC.  I'm not aware of ways that training without the feature will result in worse handling with it.  However, I'm probably SOL regarding any such issues as may exist, because my panic driving situations practice was in a car with ABS and no ESC.  Fortunately, it hasn't seemed a problem yet - I've been driving a car with ESC in the US North-east for over five years now, and the only accident I got into was one where I didn't even see the other car until impact.  (I was judged to have been less than 50% at fault in that accident, for what it's worth.)

    The best argument I've heard for learning to practice panic situations without ABS is that one could end up driving a car without the feature, and have a panic situation.  However, the chances of that happening are getting to be more remote all the time, as cars without it are being replaced with cars that have it, as most, if not all, new cars made today have it.  Also, simple awareness can help out there: every time I drove a car without ABS, I always made certain to keep about double the space between me and the cars in front of me as normal, to minimize the chance of a panic situation.  I couldn't maintain that long-term, but I've never had to drive such a car for multiple days in a row.  I've also never had to drive such a car at night in deer country.



  • @tgape said:

    Actually, I believe it would be better to do this in a car with the same features as you'll be driving.

    First, all ESC systems are not created equal.  Some are vaguely helpful, some are fantastic.  (I am assuming, probably incorrectly, that any that actively detracted from controlling the vehicle will not be offered.)  The common denominator is no help at all and if you can manage that situation then anything beyond that is just icing on the cake.  Second, there are times when ABS actually doesn't help much, if at all.   Ice, slippery pavement, when you are already at the limits of traction, etc.  (By ice, I mean "The road is completely iced over," not "There's a patch of ice there.").  In those cases, sometimes ABS can be completely and utterly useless - perhaps it may help slightly, but it also is quite likely to not help at all.  All systems (at least a few years ago) depend on relative rotation, not absolute.  If you can lock 'em all up, ABS just thinks you are stopped.  So, not having any experience at braking at the threshold of traction can never help you do it. Knowing how may or may not help, but won't hurt.

    By the way, "pumping your brakes" was ALSO a "panic situation" response.  The best solution in all cases is to modulate your braking pressure to available traction.  Yes, it can be done in a panic situation as well, but it takes practice.  

    And, also note, I'm not in any way saying ABS isn't a godsend!  It's good stuff, f'sure.

    @tgape said:

    I haven't seen the stats on ESC.  I'm not aware of ways that training without the feature will result in worse handling with it.

    IMO, If it becomes ubiquitous and reasonably consistently performant it will have graduated to ABS status (usually helpful, still useful to know how to brake effectively without it).  Until then, I think one should still know how the hell to drive a car without it.  Otherwise, I see your point and don't necessarily disagree with it in theory at least.  I just lump ABS in with it.  It all comes down to something I think is simple.  The less you *have* to rely on these aids, the better off you are.  

    @tgape said:

    I always made certain to keep about double the space between me and the cars in front of me as normal, to minimize the chance of a panic situation.  I couldn't maintain that long-term, but I've never had to drive such a car for multiple days in a row.  I've also never had to drive such a car at night in deer country.

     Ah, a dangerous pet peeve of mine.  Bastards who simply can't leave safe distance.  Where I live there's very little traffic and nice, wide open two lane roads.  So WHY THE FUCK do people in huge, noisy, expensive pickup trucks hauling 10,000 pound boats behing them insist on leaving 3 feet between them and the person ahead of them?  And another thing - it is my belief disc brakes ruined driving, making people brake late at stops, follow too closely, etc. By extension, ALL these "safety" features have changed the driving landscape significantly.

    A simple thought experiment: Let's assume two cars.  One car is a deathtrap.  It is rigged up so that any bump - the lightest of impact with anything - will cause it to explode.  The other is filled with features not yet invented yet and will absolutly guarantee that even crashing headlong into a tree at 120 mph will not even get your hair out of place - completely safe and secure.  Now, would you drive those two cars the same way? Cars all fall at different points between those two extremes, and each person has a different opinion as to where on that continuum particular cars will fall.  As such, all cars are driven differently, and IMO, the more safety features you pack into a car, the less carefully it will be driven.

     



  • @mahlerrd said:

    Second, there are times when ABS actually doesn't help much, if at all.   Ice, slippery pavement, when you are already at the limits of traction, etc.  (By ice, I mean "The road is completely iced over," not "There's a patch of ice there.").  In those cases, sometimes ABS can be completely and utterly useless - perhaps it may help slightly, but it also is quite likely to not help at all.  All systems (at least a few years ago) depend on relative rotation, not absolute.  If you can lock 'em all up, ABS just thinks you are stopped.  So, not having any experience at braking at the threshold of traction can never help you do it. Knowing how may or may not help, but won't hurt.

    So long as the tires are getting enough traction while skidding that they can resume turning during the period of time when an ABS system has disengaged the brake, the ABS system will improve braking, even on glaze ice.  That having been said, you're right, sometimes that help isn't much.  But in those situations, no help can possibly be much, unless you can drop an anchor out the back.  When all four tires have locked up at the same time, despite having an ABS system, is guaranteed to be one of those times.

    That having been said, having the ABS system engage can be as disconcerting as realizing you're in a skid, when you haven't experienced it before.  The first time my ABS engaged while my wife was in the car, she freaked out.  Fortunately, she wasn't driving.

    Also, I feel the need to point out, even though I bought my car at a point in time where every ABS system commercially available would let the brakes stay locked if all four tires simultaneously were locked, not all systems at that time were equal.  My car's ABS system cycles each wheel "independently", so the chances of all four tires locking simultaneously is reduced.  (As I understand it, it's a bit more complicated than that, as it also has some tricks I can't explain to minimize the swerve that results from differential traction.  I may not understand everything about how it works, but I have seen that it *does* work, as I've had the ABS engage more times than I would have liked, during a single snowstorm a few years ago, including at least one instance of all four wheels doing the ABS thing.)  Other systems cycle the brakes for all wheels at the same time, which maximizes the chances all wheels will skid at the same time.  The fact my car has a good ABS system was *not* an accident.  Do your research before you buy your car.

    @mahlerrd said:

    By the way, "pumping your brakes" was ALSO a "panic situation" response.  The best solution in all cases is to modulate your braking pressure to available traction.  Yes, it can be done in a panic situation as well, but it takes practice.

    Yes.  Nothing I said contested that.  In fact, I believe I specifically brought up "pumping your brakes" as a panic situation response that was not generally appropriate on a car with ABS, but one you would be trained to do if you trained for the situation with a car that doesn't have ABS.  This point was brought up very forcefully to me by my mother, once, many years ago, as she screamed "Pump the brakes! Pump the brakes!" as the ABS in the family van brought us to a stop in less than half the distance I could've managed by pumping the brakes.

    But to know how to modulate your brake pressure to available traction, it's also important to know *when* your ABS system will engage.  Many people cannot tell when a car without ABS has only a couple of wheels start skidding, so long as that does not cause the car to turn.  But an ABS system would engage that soon - frequently even before the second wheel starts to skid.  Also, the only way to do it in a panic situation is to train yourself enough, as we're both proposing, so that you're not actually panicking.  Accomplishing that is, IMHO, much of the point of the training.

    @mahlerrd said:

    So WHY THE FUCK do morons in huge, noisy, expensive pickup trucks hauling 10,000 pound boats behing them insist on leaving 3 feet between them and the person ahead of them?

    FTFY.

    @mahlerrd said:

    As such, all cars are driven differently, and IMO, the more safety features you pack into a car, the less carefully it will be driven.

    First, this is only true to the extent that the driver is aware of safety features, or lack thereof.  The one accident I've had since moving to the US NE was caused by someone in a much less safe car than mine, who did exactly what her car manufacturer's commercials recommended she do.  ("Punch it!")  Of course, rapid acceleration does not necessarily work as well in dense city traffic as it does when trying to get on a highway after you've decided to stop on the onramp, rather than merging like a decent driver.  This principle is why sports convertible drivers so frequently drive like maniacs - they don't know, and don't care.  (Not to say all convertible drivers don't know and don't care.  Just most of the ones driving like maniacs.)

    Second, there are always exceptions.  I try to keep the Massachusetts statutory 2 seconds distance between me and the driver in front of me (rather than the Boston standard 2 yards), even though I drive a relatively safe car.

    But I do agree with the basic premise, which is why most Volvo drivers drive like maniacs (and, to a lesser extent, BMW and Audi drivers).  But the perception point is also important - it's been my impression that Smart Car drivers tend to be a bit more cautious than most about tail gating, and I suspect that's because they're sitting so close to the windshield, and there's no front crumple zone of the car to speak of, so they have the perception of being very exposed.  The fact that their safety rating for front impacts is the same as my car doesn't apparently get most of them to drive like most people drive the model of car I drive.



  • @mahlerrd said:

    A simple thought experiment: Let's assume two cars.  One car is a deathtrap.  It is rigged up so that any bump - the lightest of impact with anything - will cause it to explode.  The other is filled with features not yet invented yet and will absolutly guarantee that even crashing headlong into a tree at 120 mph will not even get your hair out of place - completely safe and secure.  Now, would you drive those two cars the same way? Cars all fall at different points between those two extremes, and each person has a different opinion as to where on that continuum particular cars will fall.  As such, all cars are driven differently, and IMO, the more safety features you pack into a car, the less carefully it will be driven.

    In the first car, I would reckon millions of people would die of traffic accidents each year whereas in the second car zero traffic-related deaths would occur each year. Statistically speaking, regardless of driving skills, the second car is beneficial.

    To be honest, I've driven with people who have not kept safe driving distance to the car ahead of them, and, being the asshole backseat driver I am, would point it out to them. None of them have ever said, "Oh, well, I just got awesome new brake pads, I'll be fine." or "I'm in a Lexus, don't worry, have you seen their safety commercials?" Usually they say, "Oh, now that you mention it you're right. I wasn't paying attention" or "Fuck you, this is my car, I drive it the way I want."

    Therefore, I would argue that the rationale behind unsafe driving habits is more about simply not being aware of your environment (think highway hypnosis) or simply they're being an asshole. I live in a state that has notoriously bad drivers who are overly-aggressive, road-hogging jackasses who run red lights, speed regularly at 80+mph on urban highways, cross 6 lanes of traffic to exit in a quarter-mile, and never let anyone enter a busy street from a driveway or parking lot. I think the reason for that has absolutely nothing to do with the faith they have in the safety of their car, but is rather due to the relatively outdated infrastructure in many regions that invites this kind of behavior. Onramps that enter onto the left lane followed by right-lane offramps half a mile away, unsensored traffic lights that don't let enough cars to pass the intersection resulting in major delays, and just plain overcrowded arteries make people feel like they need to be aggressive just to get from point A to point B. Everyone is in a hurry, and they'll risk injury or death just to get where they're going.


  • Discourse touched me in a no-no place

    @mahlerrd said:

    First, all ESC systems are not created equal.  Some are vaguely helpful, some are fantastic.  (I am assuming, probably incorrectly, that any that actively detracted from controlling the vehicle will not be offered.) 
    From a racing driver's perspective, ALL ESC systems detract from controlling the vehicle. Every last one of them. It's a crutch to let drivers drive beyond their ability - if your ability already exceeds that which the computers are willing to allow (but which are well within the laws of physics) it's slowing you down. This is why very high end cars will have multi-mode ESC systems that change the allowed parameters. "Race" mode will typically be either all the way off or very heavily restricted to make up for engineered-in handling flaws that the manufacturer believes even seasoned drivers can't handle. Two drivers of equally high skill in equal cars. The only difference is if ESC is turned on or not. If the drivers are above a certain skill threshold, the one with the system turned off will ALWAYS win.

     

    The cheap systems that are in everything now because the government mandates it are among the worst of the worst. To get a decent one, you need to look at sports cars or, oddly enough, heavy-duty trucks. If you need stability control in a compact family car or a minivan, you're driving like an idiot and deserve to crash.

     

     

    Incidentally, a car manufacturer is developing a system that changes its ESC program based on input regarding your surroundings - where other cars and obstacles are, etc. This is just a few short software steps from letting the computer decide how best to let you hit something - which is a nasty rabbit hole with such decision trees as "Which is better: Hit this car and risk killing many people, or hit this static obstacle and 100% for sure kill my driver?"

    Call me a luddite if necessary, but I am absolutely not prepared to let computers make those decisions and I am absolutely not prepared to let automotive engineers write the code that does it.



  • @Weng said:

    Call me a luddite if necessary,

    Luddite.



  • @RHuckster said:

    @mahlerrd said:

    A simple thought experiment: Let's assume two cars.  One car is a deathtrap.  It is rigged up so that any bump - the lightest of impact with anything - will cause it to explode.  The other is filled with features not yet invented yet and will absolutly guarantee that even crashing headlong into a tree at 120 mph will not even get your hair out of place - completely safe and secure.  Now, would you drive those two cars the same way? Cars all fall at different points between those two extremes, and each person has a different opinion as to where on that continuum particular cars will fall.  As such, all cars are driven differently, and IMO, the more safety features you pack into a car, the less carefully it will be driven.

    In the first car, I would reckon millions of people would die of traffic accidents each year whereas in the second car zero traffic-related deaths would occur each year. Statistically speaking, regardless of driving skills, the second car is beneficial.

    To be honest, I've driven with people who have not kept safe driving distance to the car ahead of them, and, being the asshole backseat driver I am, would point it out to them. None of them have ever said, "Oh, well, I just got awesome new brake pads, I'll be fine." or "I'm in a Lexus, don't worry, have you seen their safety commercials?" Usually they say, "Oh, now that you mention it you're right. I wasn't paying attention" or "Fuck you, this is my car, I drive it the way I want."

    Therefore, I would argue that the rationale behind unsafe driving habits is more about simply not being aware of your environment (think highway hypnosis) or simply they're being an asshole.

     

    Agreed.  The "deathtrap car" thought experiment fails to consider the issue of who's driving the car.  I'd be interested to see statistics on traffic collisions broken down by safety features, though.  From what I've read, the crappy old deathtrap is more likely to be driven by someone in a higher-risk category (say, men under 25), who is more likely to have a collision regardless of the sort of vehicle they drive.  I'd be very surprised if any empirical evidence showed that advanced safety features led to more collisions, regardless of fatality rates.  The type of person who is interested in safety features is probably a risk-averse driver as it is.



  • @Justice said:

    a higher-risk category (say, men under 25), who is more likely to have a collision

    Who I would argue is a higher risk category because he can't afford a decent car or to maintain it properly.



  • @nexekho said:

    @Justice said:
    a higher-risk category (say, men under 25), who is more likely to have a collision

    Who I would argue is a higher risk category because he can't afford a decent car or to maintain it properly.

    Why would you argue something like that? I mean, I get why you would argue it here (because we argue everything here). But would you argue that somewhere else, too?



  • @boomzilla said:

    Why would you argue something like that?

    I'd argue that through a combination of elderly people being more likely to have a newer, ergo safer car, and that people retire when they're elderly thus reducing the need to commute, it likely puts a huge skew in the risk data. I don't have that data so I can't say but I find it hard to believe that experience is the only factor involved in the lower accident rate of the elderly especially given the difficulties of being elderly. I don't think it causes quite so many accidents as inexperience, rather a combination of cars with lacking safety features and inexperience.



  • @nexekho said:

    @boomzilla said:
    Why would you argue something like that?

    I'd argue that through a combination of elderly people being more likely to have a newer, ergo safer car, and that people retire when they're elderly thus reducing the need to commute, it likely puts a huge skew in the risk data. I don't have that data so I can't say but I find it hard to believe that experience is the only factor involved in the lower accident rate of the elderly especially given the difficulties of being elderly. I don't think it causes quite so many accidents as inexperience, rather a combination of cars with lacking safety features and inexperience.

    Oh, you were serious.

    I'd bet that the elderly drive less. If nothing else, retired people don't commute to work. Also, younger people have less experience, plus young males like to drive fast and stuff, and they're less experienced. I'd be shocked if most accidents had much to do with safety features and technology relative to driver error.

    Also, rental cars are much more difficult to come by if you're an unmarried male under 25, and everybody drives the same cars there. Granted, a lot of that probably comes from the same actuarial tables as the insurance companies, but there are some companies who will rent to you, and I'm sure if there was enough of a difference due to the quality and maintenance of the car, it wouldn't be nearly so difficult to find someone to rent a car to young men.



  • @boomzilla said:

    Also, rental cars are much more difficult to come by if you're an unmarried male under 25, and everybody drives the same cars there.

    Not true.  Even if I have a car reservation with a particular rental company, if they don't have a safe car for me to drive, I'll go to another rental place until I find one that does.  No, I'm not familiar with every car make and model, but I *do* know how to google for car safety ratings on my phone.

    I'm less than half way to retirement, yet I purchased my car with safety as my top priority.  Admittedly, I also don't drive recklessly.

    Now, I've seen quite a few winter accidents that were very clearly relating to both safety features and driver error.  Specifically, AWD does not mean your car can break faster, corner faster, or do anything significantly better on ice.  (Note: Yes, double the acceleration is normally considered significant.  But double 5% of normal max acceleration is still pathetic.)  But I've seen many drivers who made the error of thinking it did.  Also, as previously mentioned, ABS may help on ice, but not enough to matter much.  It may be enough to cover your miscalculation if you're trying to drive safely.  It probably won't cover your miscalculation if you're trying to push the limit.

    @boomzilla said:

    Granted, a lotall of that probably comes from the same actuarial tables as the insurance companies.

    FTFY.  The premiums they need to pay to rent to an unmarried male under 25 are insane.  Yes, they're supposed to be able to count on the driver's insurance.  However, in practice, they can't.  But that's actually a lot more about the ability of the unmarried male under 25 to pay the accident damage than it is about their driving ability.  The 85 year-old legally blind bat who rents a car and rams it into the car in front of it - at the rental place - is pretty much guaranteed to have insurance, and the accident was at low speed, so the cost isn't so high.  The 18 year-old punk who wrecks his rental car is much more likely to total it, and much more likely to not have effective insurance.  (Sometimes, he may think he's covered under his dad's policy - but is only covered when driving the dad's vehicles.  Other times, he may not have any coverage, and not be concerned by it, waiving the rental company insurance anyway because it's too expensive.)

    As I understand it, companies that do rent cars to unmarried males under 25 are generally either self-insured, or they require unmarried males under 25 to purchase the normally optional insurance policy - which is as expensive for the unmarried male under 25 as you'd expect, given the actuarial tables previously mentioned.  But some unmarried males under 25 take offense at that, and some rental car companies don't want to deal with that offense.

    Disclaimer: I am not a lawyer, and my exposure to unmarried male under 25 rental policies is over a decade out of date, since I'm no longer an unmarried male under 25.  But I'm certain the basic premise that the rental companies don't want to rent to them because of insurance costs is still true.



  • @tgape said:

    @boomzilla said:
    Granted, a lotall of that probably comes from the same actuarial tables as the insurance companies.

    FTFY.  The premiums they need to pay to rent to an unmarried male under 25 are insane.  Yes, they're supposed to be able to count on the driver's insurance.  However, in practice, they can't.  But that's actually a lot more about the ability of the unmarried male under 25 to pay the accident damage than it is about their driving ability.  The 85 year-old legally blind bat who rents a car and rams it into the car in front of it - at the rental place - is pretty much guaranteed to have insurance, and the accident was at low speed, so the cost isn't so high.  The 18 year-old punk who wrecks his rental car is much more likely to total it, and much more likely to not have effective insurance.  (Sometimes, he may think he's covered under his dad's policy - but is only covered when driving the dad's vehicles.  Other times, he may not have any coverage, and not be concerned by it, waiving the rental company insurance anyway because it's too expensive.)

    As I understand it, companies that do rent cars to unmarried males under 25 are generally either self-insured, or they require unmarried males under 25 to purchase the normally optional insurance policy - which is as expensive for the unmarried male under 25 as you'd expect, given the actuarial tables previously mentioned.  But some unmarried males under 25 take offense at that, and some rental car companies don't want to deal with that offense.

    Disclaimer: I am not a lawyer, and my exposure to unmarried male under 25 rental policies is over a decade out of date, since I'm no longer an unmarried male under 25.  But I'm certain the basic premise that the rental companies don't want to rent to them because of insurance costs is still true.

     

    In my recent experience (I'm 27), there's usually just a huge premium charge.  I believe Enterprise will rent to someone under 25 as long as the money keeps coming; I had a rental from them some years ago while my car was in the body shop, but that was all paid by the other driver's insurance, so I don't know what the exact premium was.  The few times I did look into renting a car in my younger days, it usually just cost twice as much, regardless of gender or marital status.

    Regarding the elderly, insurance, and lower collision rates: driving less helps tremendously, yes.  I saved a bundle on my insurance because I walk to work, so my car gets classified as a pleasure vehicle.  I don't know what the collision rates look like in the high age brackets, but from what I've heard, age 40-60 is about the lowest.

    I'd have to ask an actuary about this, but I think the high rates for young men are partly based on high-risk lifestyle factors (PARTY HARD BRO), so little things like "being employed" can help with the premiums.  Correlation is not causation and all that jazz, but tell that to the insurance companies.



  • @Justice said:

    Correlation is not causation
     

    But it does imply it. ;)



  • @Justice said:

    Correlation is not causation and all that jazz, but tell that to the insurance companies.

    It is when you make money based on actuarial calculations.


Log in to reply
 

Looks like your connection to What the Daily WTF? was lost, please wait while we try to reconnect.