While we're on the subject of maps-related bugs, I was recently borrowing a new Tesla Model Y and took it on a RORO ferry. After the crossing, the car’s GPS was convinced I was still at the port where I had departed from. I restarted it a couple of times, but nothing. I drove off using Waze on my phone instead of the car's navigation. The map on the car kept moving relative to the direction I was driving, so the navigation was showing me driving into the sea and eventually started complaining that it would be impossible to find a charger.
Approximately 5 hours later, just as I was about to arrive, the car finally managed to figure out my correct location.
Exciting trip, not a huge fan of Teslas, but their charger planning is really nice. It was very unpleasant to suddenly lose it.
I just genuinely wonder how such a bug can actually occur, surely you'd update the GPS fix more often than every couple of hours. Hard to imagine the car just suddenly couldn't get a GPS fix for hours either. But if it did somehow totally lose the ability to use GPS, the car must have a pretty good dead reckoning system given how well it was responding to my changes in direction.
On a vaguely related note, driving 3000 kilometers through Europe in an electric car was surprisingly nice. Certainly didn't affect the length of the trip nearly as much as I'd have expected, but it was certainly super annoying to try and figure out the optimal rate of travel on the Autobahn. Charging at Tesla's supercharges was vastly more expensive than I expected, the "fuel" costs weren't much lower than what you could easily reach with a diesel car.
>On a vaguely related note, driving 3000 kilometers through Europe in an electric car was surprisingly nice.
I did 2 cross country road trips here in the US (~5000mi/8000km total) and had a similar experience. The nav's automatic charger routing did a great job, and we had 0 issues with charging.
The most annoying thing about iPhone Maps for me is the inaccuracy of the direction indicator. I've googled and it doesn't seem to be a universal complaint even though I've experienced this on multiple devices over the years even after following instructions to re-calibrate multiple times. Maybe I've just been unlucky.
It intensely frustrating to use maps for direction finding while walking or cycling when the app is telling you are facing one direction when in fact you are facing another.
Tangentially, why won't computerized maps - e.g., Apple Maps, Google, etc. - show a scale? Generally they'll show one as you pinch/zoom, but it disappears when you release your fingers.
Sometimes I just want to eyeball distances, and not plug in addresses and request routes for everything.
I know not everyone always wants it, but a scale is a fundamental feature of mapping! Scale almost defines mapping.
Edit: One reason I ask is because I suspect some technical limitation.
I thought it'd be cool if the scale were shown by some channel, like coloring roads (pink at neighborhood scale, blue, green as you zoom out) so you could intuit the scale at a glance.
With map simplification and symbolic markers (roads are the same thickness at multiple scales) it can be really hard to figure out distances. I was once going to a less populated area, based on my local intuition I thought the distance between two locations was a maybe 30m walk max if I missed the bus, but it turned out it was a 3h walk.
I’ve had that bug a few times. I think it’s related to some roads being closed and/ or under reconstruction. I’ve seen this happen multiple times on the same route, and it always fixed itself after I passed the construction site.
Maps has gone backwards on CarPlay, too. Hysteresis is completely broken - I stop at a traffic light and the map will jump back and forth between two resolutions, often once a second, which is extremely distracting.
I also love that Find My says my partner was "last seen" at home "8,912 days ago".
Apple just doesn't give as much of a shit any more.
I made a comment earlier that was rightly flagged for its tone, and I would like to restate the technical point more constructively.
The post author, Dan Piponi, clearly knows about fractals, but his post raises the question of whether asking Fermi questions in interviews is actually effective. I am skeptical that such questions would have prevented this type of bug.
I suspect the issue stems from small measurement imprecisions accruing over long distances, which is—in my view—tied to the fractal nature of roads traversing natural landscapes.
However, as others have pointed out, it may also be tied to road closures: if closed segments are set to a higher length internally (to discourage routing), these values might be getting summed up blindly over longer distances.
None of these issues would have been prevented by being good at estimating quantities alone.
Apologies again for the unconstructive tone of my previous comment.
Here's a Fermi-ish question: given the fractal nature of typical landscape, should a truck driver budget for 10 times as much time as would be predicted by using the known typical speed of the truck and the distance as the crow flies?
(Aside: the nearest grocery store to my house is 1 mile away but the shortest route by car, as measured by the car odometer, is 10 miles.)
22 comments
[ 3.1 ms ] story [ 43.3 ms ] threadIf you put all of your blood vessels end to end they would go to the moon and back.
Approximately 5 hours later, just as I was about to arrive, the car finally managed to figure out my correct location.
Exciting trip, not a huge fan of Teslas, but their charger planning is really nice. It was very unpleasant to suddenly lose it.
I just genuinely wonder how such a bug can actually occur, surely you'd update the GPS fix more often than every couple of hours. Hard to imagine the car just suddenly couldn't get a GPS fix for hours either. But if it did somehow totally lose the ability to use GPS, the car must have a pretty good dead reckoning system given how well it was responding to my changes in direction.
On a vaguely related note, driving 3000 kilometers through Europe in an electric car was surprisingly nice. Certainly didn't affect the length of the trip nearly as much as I'd have expected, but it was certainly super annoying to try and figure out the optimal rate of travel on the Autobahn. Charging at Tesla's supercharges was vastly more expensive than I expected, the "fuel" costs weren't much lower than what you could easily reach with a diesel car.
I did 2 cross country road trips here in the US (~5000mi/8000km total) and had a similar experience. The nav's automatic charger routing did a great job, and we had 0 issues with charging.
It intensely frustrating to use maps for direction finding while walking or cycling when the app is telling you are facing one direction when in fact you are facing another.
Sometimes I just want to eyeball distances, and not plug in addresses and request routes for everything.
I know not everyone always wants it, but a scale is a fundamental feature of mapping! Scale almost defines mapping.
Edit: One reason I ask is because I suspect some technical limitation.
With map simplification and symbolic markers (roads are the same thickness at multiple scales) it can be really hard to figure out distances. I was once going to a less populated area, based on my local intuition I thought the distance between two locations was a maybe 30m walk max if I missed the bus, but it turned out it was a 3h walk.
The story arc is about a lost airtag that is living it's secret life in Mexico.
The link is for a post where Apple decides the distance to it's location in Mexico is greater than the circumference of the Earth.
I also love that Find My says my partner was "last seen" at home "8,912 days ago".
Apple just doesn't give as much of a shit any more.
A geostationary orbit is ~26,000 miles. Dollars to donuts, that's where it is.
The post author, Dan Piponi, clearly knows about fractals, but his post raises the question of whether asking Fermi questions in interviews is actually effective. I am skeptical that such questions would have prevented this type of bug.
I suspect the issue stems from small measurement imprecisions accruing over long distances, which is—in my view—tied to the fractal nature of roads traversing natural landscapes.
However, as others have pointed out, it may also be tied to road closures: if closed segments are set to a higher length internally (to discourage routing), these values might be getting summed up blindly over longer distances.
None of these issues would have been prevented by being good at estimating quantities alone.
Apologies again for the unconstructive tone of my previous comment.
(Aside: the nearest grocery store to my house is 1 mile away but the shortest route by car, as measured by the car odometer, is 10 miles.)