You appear to be forcing me to write a driver’s manual sentence by sentence! 
This is kinda getting off the main topic of the thread, but the proper solution to “lane that comes to a stop” is bloody well not to proceed until you have space on the other side to clear the intersection. Failing to do this is not only how bad drivers get stuck in the middle of intersections, but also how they get boxed in sitting on railroad tracks when a train is coming.
Anyone that always leaves an entire intersection worth of free space in a busy city environment isn’t driving safely.
It’s nothing at all like the train track problem, where you should only enter if there is at least a car-length (plus margin) of space on the opposite side.
Ehh, in Texas, they’re probably just driving legally. In general, if you can’t exit the intersection on the other side, you can’t legally enter it.
I’ve seen enough idiots screwing up already heavy traffic by blocking intersections that they aren’t leaving any time soon that I don’t enter an intersection unless I can clear it, and I’m a total scofflaw.
There’s an intersection near me that’s about 360 feet across. Twenty-ish car-lengths. And past the intersection is a curve with a stoplight behind it that could be backed up–no way to tell for sure from that far back (I mean, you can’t even really see the other side of the intersection, let alone anything much past that).
I’m second in line at the stoplight. The first car goes, but I can’t tell for sure that I’ll be able to exit. Are you saying that I should stay behind the line, stationary, until the lead car successfully exits the intersection?
But even if I did that I still couldn’t guarantee anything! Because while I was waiting, other cars may make a right-turn-on-red and fill any gap I left.
Yes, of course it’s stupid for people to block the intersection when it’s an obvious high risk. I’m talking about the rare edge cases here.
And you’re a human, and you pays your nickel and you takes your chances. If you can’t exit the intersection, you’re technically breaking the law here.
As to that particular intersection, I’d petition for the lights to be better coordinated if you can’t expect the next intersection to clear. I know of similar intersections in Dallas, and I avoid them like the plague from any direction. Seems like it’s risky by design, and the only way to win is not to play.
Well, if I got a ticket I’d fight it, arguing that there’s no reasonable way to anticipate the obstruction, and that waiting would have caused even more of a problem. People behind me would get annoyed and maybe have gone around me. In fact I don’t think it’s possible to even clear the intersection before the first light has turned red (it’s so wide that there are two sets of lights with different timing).
And in my experience, you’ve got a 50/50 chance of fighting it working out in your favor, mostly based on how that particular judge feels that day. I’ve made what I felt were totally reasonable arguments in traffic court and lost, and made what I thought were pretty out there arguments and won (hey, I kept my license).
Woah, it’s like it’s designed to create problems. Good luck with that problem, self driving cars. I’m glad I’m not developing you.
Strongly disagree. The width of a city street isn’t much of a buffer to optimize traffic flow, but more to the point, no such space is a buffer at all if you only “borrow” it and it has to be cleared every time the light changes. So on the contrary, the practice of barging into an intersection when it’s not obvious that you’ll be able to clear it before the light changes may well leave you blocking the intersection, with irate cross traffic honking at you and angry drivers trying to get past. Now that would be a classic example of a driving safety problem, directly caused by the behaviour you’re trying to defend.
I’m talking about exceptional cases, not the typical case. The typical case is that you make a judgment call about how traffic is moving and either hang back or not depending on how it looks. If traffic is moving smoothly, you keep a normal following distance, which is smaller than many intersections. But sometimes you’ll get it wrong and block traffic a bit. Almost always this is when traffic is heavy to start with and all drivers have some expectation that they might have to wait a bit before the intersection fully clears.
You yourself emphasize “predictability”, but this means driving like other people on the road, even when technically against the law. Just like driving the speed limit in the fast lane of a freeway when the flow is actually 15 over the limit. You are the unsafe, unpredictable one in that case.
In the intersection I described above, waiting for the lead car to clear means a 100% chance of irate drivers behind me honking and trying to get past me, whereas keeping a normal following distance means a <1% chance of that happening, since I’d only block the intersection when I guessed wrong about being able to clear it. So yes, based on the odds I’ll pick the latter.
Actually, here I agree with your points. For driving speeds, it’s generally safest to be keeping pace with the general traffic flow. The exception I make is when too many morons are driving too fast for the conditions, such as poor visibility or, worse, poor visibility combined with icy/snow road conditions. These are the morons who often end up in the ditch, sometimes upside down. In that situation I simply drive in the rightmost lane at a safe speed and let the daredevils go by at whatever speed they want in the outer lanes. It they have to change lanes to get around me because they want to drive 120 km/h in a blinding blizzard, that’s their problem, not mine.
On the intersection matter, this was the first time you mentioned what appears to be an extraordinary special case, apparently caused by what sounds to me to be very bad road design. So I can’t comment on that specific situation, but my earlier comments were prompted by the number of idiot drivers that barge into the intersection even when the traffic ahead isn’t moving, because dammit, if they didn’t they might have to wait until the next green light cycle. Then they end up with their fat-ass Ford F-150s sitting in the middle of the intersection blocking all the cross traffic when the light changes. These assholes make me wish that my car was equipped with cannons, James Bond style! 
Indeed. There are no blizzards here but we do have significant downpours, and fog can be a problem. I’m quite happy to putter along in the right lane, following a semi at a safe distance if possible.
It’s unusual but not that extraordinary. It crosses a major expressway with 3-5 lanes in each direction (depending on how you count them), plus an elevated light rail track in the middle, plus it crosses at a 45 degree angle. Here’s the street view:
The green traffic light you see is just the “intermediate” light. The end of the intersection is waaay further out, next to the buildings and trees on the other side of the elevated track. That motorcycle rider can’t even see past the truck in front of him!
Yes, I certainly agree with this. It’s especially infuriating when you see them do it intentionally just to make the light, even though they’re inconveniencing hundreds of others.
Another total failure of my Tesla’s self driving, albeit a safe one. Stopped at a light in the leftmost of two left turn lanes, the car panel went red with “take control immediately”. No obvious change in surroundings. After pressing the brake and rotating the steer wheel a bit, the error message was something like “automation stopped due to system failure”.
I actually restarted the self driving, and it didn’t complain. Until after a few seconds it again totally failed. Same messages.
Anyway, the self driving isn’t good for my commute. I’ll keep trying it on the weekend.
So, incredibly, it is the 7th anniversary of this thread, and I need an assessment. Was the OP right? I, for one, am not hearing much about autonomous cars in the news like I was when this was a big item back when. So, I’m thinking the OP was right. What say you?
There are Waymos that are fully autonomous that are only allowed to drive in some very limited areas. Teslas are largely (but not near fully) autonomous on regular roads and highways but not unmarked rural roads and that’s in the US and Canada only. The major manufacturers that aren’t Tesla don’t have anything more than intelligent cruise control.
So the OP was right not to buy it.
I enjoyed this review from a guy who attended the recent Tesla event. I don’t know this guy, his bio says he’s a web and podcast producer, but scrolling his videos he seems like a big EV (and other tech) fan who has…questions.
Surprising that you haven’t run across MKBHD before. He’s a big name. Though IMO his EV stuff isn’t really as good as his phone reviews (where he started). He reviews cars from a techie perspective, not as a gearhead.
But honestly I don’t get the confusion. Almost all their design decisions make sense to me. I don’t think people are really fully comprehending that this is actually a fully autonomous vehicle, and can’t possibly be anything else, regardless of how skeptical you are of FSD. If FSD doesn’t work, the car won’t ship. Period.
Like, he’s confused about the wireless charging, saying that it’ll be hard to position over the mat. Yeah, hard for a human, but humans aren’t driving it. It’s trivial for self-driving.
He also sorta embarrassed himself on X by claiming that wireless charging can only be 75% efficient:
Tesla bought a whole company specializing in wireless vehicle charging with ~93% efficiency. 75% might be true for the phones he knows about… not cars.
On the plus side, he said he’d shave his head if Tesla ships by 2026. So that’ll be entertaining, at least.
We talked a bit earlier about price, and whether we should expect them to hit the $25k target. One tidbit that came out from some of the event attendees, but was also confirmed in the Tesla Q3 results, is that they’re targeting 5.5 mi/kWh. That’s an excellent efficiency figure (the Model 3 is ~4 mi/kWh).
They’re also going for a ~200 mi range. Combined, that means it only needs a 36 kWh battery pack. That’s in comparison to the 80 kWh pack on the Model 3. So the single biggest-ticket item should be about half the cost.
And of course it’s just physically smaller, has fewer luxury items (like no glass roof), and is missing some parts that normal cars have (like the steering wheel…). All these things and more should help keep costs down. Tesla’s average cost of making a car is already only $35k (and this includes all their higher end models), so hitting <$25k for a far smaller and simpler vehicle seems… not unreasonable.
Of course, 200 mi kinda sucks compared to >350 for the Model 3. But it’s a robotaxi. It doesn’t need that much. If you want a long range car for personal use, buy one of the more expensive models.
A couple weeks ago I took our EV on it’s first road trip, from Phoenix-metro to the Grand Canyon south rim and back. I stuck to 65 and less because range really drops above that speed. That was mostly semi-truck speed anyway didn’t have to w=and I was often between two of them so the zoomie-zooms didn’t have to plan hard how to get around me.
I must say it warmed the cockles of my heart when, coming off the Colorado Plateau in I-17 I watched the range meter go from 198 miles to 227.
How was the charging situation? With our Tesla, the chargers in Flagstaff and Tusayan make South Rim trips easy, but EV charging near the North Rim is almost nonexistent. For North Rim trips, we rent an ICE car.
Well, being a naysayer for wildly ambitious tech isn’t exactly a hard job 
But I think the goal posts have moved in the last 7 years, and not in a pejorative way. Similar to how people expected flying cars right around the corner in the 50s. Well, flying cars are here, they’ve been invented, prototypes have been built, and I believe there are some in production. In the meanwhile, we discovered that we don’t actually have much of a use case for flying cars. Wheeled vehicles got a lot better, more comfortable, efficient, etc. Speed limits and infrastructure all improved. Cheap air travel became a thing. And thus, the killer application for flying cars never materialized.
For self-driving cars, 7 years ago I pictured getting into a car that I own and store in my garage, opening a book in the back seat, and spacing out while the car takes me wherever I want to go. What’s changed is that we’ve gotten a lot of the benefits of self-driving cars without actually having self-driving cars. We have cheap ride-sharing apps that have caused some people to rethink the need for cars entirely. A lot of us who never dreamed of being able to work remotely now work 100% remote. And now with Waymo, the concept of a self-driving car might give way to the concept of a self-driving taxi. Thus, the killer application for self-driving cars may never materialize.