Thanks a lot asshole!!! I followed your advice and now I can't see anything.
Congratulations... you have more faith in human programmers than I do, part of which comes from my consistently having to fix system bugs that are introduced both by colleagues and occasionally errors that I missed in my own code. Also, as a former professional driver, I used to see the rolling box scenario quite often, sometimes several times in a day (note that I'm not saying I was in the box all those times, though it has happened a couple times when I wasn't paying attention to their pattern). The "victims" of such did not necessarily have to be right up on the tailgate of the truck to be slipstreaming either. 9 times out of 10, the truckers in these incidents weren't trying to perform the box intentionally as it was just how the convoy formation was moving, and often times a quick warning on the CB was all it took to make the guy on the left realize he needed to make an opening for the car. The 1 in ten case I would usually hear one of the members of the team call out "Make the box" or something similar over the CB before the formation started. This usually happened when the driver was following too close.
And actually, one can be in the slipstream even if you're 500-1000 feet back from the truck. Pro-Tip: If you feel your vehicle doing a slight side to side shimmy while you're behind the truck, and you know your front end is in good repair and proper alignment, you're in his stream and sucking his fuel. The Shimmy is even worse on the motorcycle. You could be cruising down the roadway and suddenly feel the bike wobbling side to side on a windless day. Look a mile down the road, and there's the Semi that's disturbing your corridor. Get within a half to three quarter mile of his tail, and the bike is inside his stream and getting mixed up like a bird in a hurricane. Close to less than one quarter mile away from him, and the bike is now stable in the stream.
That's the correct procedure to follow and answers the question of what does the car do in the situation, but you completely missed the mark of "How does it recognize that it's getting boxed in?" It's actually a special case problem that requires intuition and human reasoning. Not only that, but to maneuver out of the situation is going to generate a lot of conflicting sensor trips that human intuition and aggression would ignore to get out of the situation.
The question wasn't so much posed as "what should the car do?" but more as how does the car recognize and react? What sort of If-Then-Else logic does the car need to perform to override what some of its sensors are telling it? There are rare times when throwing caution to the wind will save a person, and if you find yourself in positions where you need to throw caution to the wind to get out of a bad predicament too often you probably need to check and evaluate yourself, but how do you program a machine to throw caution to the wind when it's necessary? What happens when the chaos of the world puts its full will to bare against the perfect order of mechanical procedure?
Finally, there's your point of automate everything on the roadway. Yes, that would be ideal, and for the truckers that would complain about the loss of freight jobs I say change is inevitable; but you can't argue that at least for a while you're going to have automated vehicles sharing the road with traditionally controlled vehicles. For this, a solution needs to be made. Personally, on the Interstates I would prefer to see automation only lanes with their own entrances and exits that travel parallel to the non-automated lanes. This would eliminate the chaos processing the cars would have to perform at high speed.
What I'd miss most when diving into automated travel is the decompression that comes with my daily commute to and from work. That's an hour each way when there's nothing to think about but me, the road, and the performance of my bike while taking in the scenery with the four cylinders humming nicely at 2800 RPM, 5th gear pushing me along at 45MPH.
Dave, are you a programmer?
If you are, what steps would you take to ensure that an automated vehicle would not allow itself to get caught in The Rolling Box? Let me go over the scenario with you.
You are cruising inside of a truck's slip stream at about 65 mph (automated vehicle, it's going to ride closer to the truck than recommended for humans by default and thus can take better economic advantage of the slipstream.) You have heavy bumper-bumper traffic moving to your left at about 75-80mph with less than a car length between them and a jersey barrier on your right, half a foot from the shoulder line so no room for your vehicle to go there. A Tractor trailer is coming up on you from behind and matches speed with you about a truck length behind the truck in front of you. Now they're starting a convoy shift so the truck that was in the tail for the last 10-20 miles will move to the front to start giving the front truck an economy boost through slip streaming. The truck performing the shift gets to be right beside you in his passing maneuver, and now you are in the kill box. Most likely, the truck will continue on and take his place at the front of the convoy, but at this critical moment when he is right beside you there's countless things that can go wrong, and every single one of those things WILL KILL you (especially in that little POS frame).
Another way to end up in the same box is to have a truck pull up beside you first, then a third truck will pull in behind you and complete the box... though this one tends to be deliberate and happens more when a driver pisses off a trucker... such as when a driver slipstreams a trucker without getting permission.
So, Dave. What does your car do? How does it recognize that it's getting boxed in, especially in the latter case when it looks every bit like the truck is just going to pass you and the truck in front of you...but decides to build the box instead?
Ok dude... I'm a programmer in a company that supports Agriculture, which means that I get to meet a lot of farmers. If I can get a bunch of farmers pissed off about the kind of data service they're getting from their ISP, then it shouldn't be too difficult to get your businessmen, doctors, nurses, etc to realize they're not really getting what they pay for.
Of course, if you can't persuade people in a way that they would understand then maybe you need to look at yourself and see where your articulation is failing you. Perhaps you don't feel confident in your own knowledge and understanding of what this whole customer/Netflix/Comcast relationship entails. Rather than face your own lack of understanding, you'd prefer to give up before you even start and externalize that which you lack, projecting it upon the people around you with the rationalization that if someone of your brilliance cannot understand the concept, how can anyone else around you understand it. I completely understand this, and in a way it is true. It's a self fulfilling prophecy of a sorts. If you can't understand a concept it is a near impossibility for you to articulate the concept to a person who has less of a technical platform to use as a basis than you do.
So, rather than stating that I've lost 99% of the people with my psychoanalysis, please let me know how I can help you with your understanding of this subject that you may be armed with the persuasive weaponry that you can use on those around you that you may increase the size of the army that can stand against Comcast?
So what? Educate them. Show them the Correlation: User pays Comcast, Netflix pays their direct carrier, Comcast wants Netflix to pay them directly to guarantee QoS, Netflix gives in giving user better experience and subsequently a couple months down the line increases the rate on their service. This rate increase is to directly pay Comcast for the added service charge, which can be effectively presented to the user that Comcast is double dipping from the user via a third party (Netflix), and if Comcast hadn't tried to play the dirty pool, it's unlikely that their Netflix rate would have gone up.
There's a way to spin anything to get someone riled up. If they still cannot see the issue with this or worse, justify it with "If it makes my service better I'm all for paying a little extra", hit them with "Comcast didn't do anything to make your service better, they just stopped actively making your service worse to try and milk money Mafia style outta you and Netflix." If they still don't get it, just tell them that they're a part of the problem with jacked up data rates and leave it at that. If they don't understand by this point, they're hopeless and nothing will work.
One thing about all this is that we cannot give up. We need to make every effort to try to educate the people that will listen and understand why this is so bad. If we give up at the first sign of resistance, we're as bad as they are. If we give up after hammering the strongest point we have in the argument and they still won't change their beliefs, only then should we move on, possibly to revisit once there's stronger arguments. Advocating for change is hardly ever easy.
If I lose power steering or braking I just have to steer or brake harder - ok, it sucks, but if you can't do that you shouldn't be driving
That's fine and dandy for hydraulic power boosting systems, but what about these cars coming out now with electronic fly by wire systems that don't seem to have any mechanical linkage between the control interface and the control surfaces?
I have a GM vehicle that has an electrical issue that regularly drains the battery where it needs to be cranked and run daily, or else it will need to be jumped. That is an issue in and of itself that GM hasn't really took seriously that isn't the focus of my discussion here. One such time I was running late for an appointment and found the car was yet again dead in the driveway, and I had to roll it down the hill to make room for me to get my motorcycle out of the garage. I discovered just how harrowing this car would be if it ever electrically died on the roadway.
When I unlocked the ignition, my break pedal went to the floor without any resistance and there was very little movement in the front wheels for the wide swaths I was making with the steering wheel...again with VERY little resistance lock to lock. I put my foot on the break, shifted to neutral and slowly released the Emergency/Parking brake. Without the e-brake, the breaks were pulsing and would not hold the vehicle, leaving a very jerky ride down a long hill. The steering wheel also had no means of controlling the direction, but luckily the drive is long and straight. Before I got to the end of the driveway, I slowly reengaged the emergency break and tried to shift out of neutral...NOPE. Transmission was locked into neutral now, AND my ignition couldn't be turned back to lock.
I then got out of the car, applied power from my bike's battery tender to the posts under the hood (the actual battery is under the passenger seat), deactivated the alarm that just woke up, was able to move all the control interfaces as expected with proper resistances, shift back into park, and retrieve my key.
I will be so glad when I can unload this thing and put the money into a new front suspension for my '96 Caprice Classic, like I wanted to do in the first place. Never ever ever buy a Traverse.
"Money is the root of all money." -- the moving finger