3 ms·

I'm so tired of these stories that pretend like the the map has 100% culpability for the decisions the drivers make. If a road looks closed or dangerous with yo

I'm so tired of these stories that pretend like the the map has 100% culpability for the decisions the drivers make. If a road looks closed or dangerous with your eyes, you shouldn't ignore all of that and drive it anyway because you saw it on a map. The Google engineer had a great defense in this article: > "[People are suggesting] if you can't get driving directions and road closures right 100% of the time, you shouldn't offer any. Surely that would be a worse trade-off than the status quo?" he wrote. "Consider that we don't know a priori which mistake is potentially dangerous." Anecdotally: I often drive through a familiar and paved mountain pass. In the past year, it has become increasingly difficult to convince Google Maps to draw a route through that pass, even though it's significantly shorter than going around the mountain. I get the impression that Google has caved to these pressures and started preferring paths that are lower risk, even if the roads in question aren't that risky to begin with.

deanCommie 5y ago
I personally completely agree, but let's play devil's advocate: Many safety regulations exist for precisely these circumstances: grey areas where an intellectual consideration of the situation would cause the safety feature to not be needed, but people rush, are careless, or just stupid. And we deem that even hurried careless stupid people deserve to be safe. There is a second factor: How do people treat navigation direction (from Google Maps or otherwise)? I think when they first came out, they were an assist - an augment to the existing driver's careful consideration of their route. Not so anymore. I have friends that will absolutely never overrule Google Maps under ANY circumstances. Even if they are visiting a city that they don't know, and are driving with a local, and the local says "Oh don't take that road, it'll be terrible, turn here intead", they'll say "no, I'd prefer to follow the nav". Some of this is occasionally justified. Local wisdom and old wives tales about faster roads and shortcuts do not incorporate active live traffic conditions like Google Maps can. But a lot of it is simply that the thought process has been reversed. People are defaulting to trusting the sat nav to make the navigation decision COMPLETELY - and they focus on the moment to moment driving conditions. This level of trust has been EARNED by the quality of the product. Which means it's entirely reasonable for 2 people to face a snowy mountain road, have one of them say "This looks sketchy, maybe we shouldn't go", and have the other respond "Yeah but Google Maps wouldn't have recommended it if it was. So it must be safe". And that would be a logical counter-argument. So no, the Google Engineer's defense isn't actually very good. In situations like this, yes, it MIGHT be better for a mainsteam product to default to "no safe options", similar how you can check "no toll roads" in your settings and only get directions for free roads. Suggesting sketchy back roads should be a setting that you opt in to explicitly. Of course the next step is detecting what is a sketchy mountain road and what isn't, etc. "Paved" might be the threshold I would use. "Include unpaved roads" is the setting I would add - defaulted to off.
dap 5y ago
> The Google engineer had a great defense in this article: >> "[People are suggesting] if you can't get driving directions and road closures right 100% of the time, you shouldn't offer any. Surely that would be a worse trade-off than the status quo?" he wrote. "Consider that we don't know a priori which mistake is potentially dangerous." That's a terrible argument, and it's even worse in context[1] because the person he's replying to didn't start from the position that "Google Maps shouldn't offer any routes ever". They asked if it were reasonable for Google Maps to express the known uncertainty in specific situations and the engineer seemed to imply that wasn't workable. It's frustrating because (as people keep pointing out) these product choices can result in death (regardless of whose fault you think that is) and the response sounds like they just don't care. I'm sure the problem really is hard. It's also hard to believe Google couldn't do better. It's known that roads in certain areas can be especially treacherous in bad weather. The NWS posts official warnings about bad weather, including time ranges, affected areas, etc. It's not hard to imagine some heuristics like "in the Tahoe area [really], when the NWS has posted a winter weather warning, strongly prefer the main roads". Sure, this is harder to scale. But if that's the issue, then it's not "we couldn't possibly do better". It's "we're not willing to invest in what it takes to do better". (I get that even this idea might be flawed for reasons I don't know -- but I'm certainly not convinced from the argument in the article that there's just no way to do better.) [1] https://twitter.com/BuschnicK/status/1475750318294061060 https://twitter.com/BuschnicK/status/1475750318294061060
AlotOfReading 5y ago
My father and I have an annual tradition where we find a country and go get stuck in. I typically end up as the navigator and manage Google maps (among others). Having a dedicated navigator is the only safe way to use Google maps outside routine conditions, in my opinion. It often recommends incredibly dangerous detours that aren't visible in the navigation UI because they're off-screen. The only information you get is "alternate route available 3 minutes faster" that you have to decline in 10 seconds or it will autoselect. If you're lucky enough to have a dedicated navigator, they'll still have to know to review the entire change every time and have pretty extensive knowledge of the area to know those conditions ahead of time.

deanCommie 5y ago

I personally completely agree, but let's play devil's advocate: Many safety regulations exist for precisely these circumstances: grey areas where an intellectual consideration of the situation would cause the safety feature to not be needed, but people rush, are careless, or just stupid. And we deem that even hurried careless stupid people deserve to be safe. There is a second factor: How do people treat navigation direction (from Google Maps or otherwise)? I think when they first came out, they were an assist - an augment to the existing driver's careful consideration of their route. Not so anymore. I have friends that will absolutely never overrule Google Maps under ANY circumstances. Even if they are visiting a city that they don't know, and are driving with a local, and the local says "Oh don't take that road, it'll be terrible, turn here intead", they'll say "no, I'd prefer to follow the nav". Some of this is occasionally justified. Local wisdom and old wives tales about faster roads and shortcuts do not incorporate active live traffic conditions like Google Maps can. But a lot of it is simply that the thought process has been reversed. People are defaulting to trusting the sat nav to make the navigation decision COMPLETELY - and they focus on the moment to moment driving conditions. This level of trust has been EARNED by the quality of the product. Which means it's entirely reasonable for 2 people to face a snowy mountain road, have one of them say "This looks sketchy, maybe we shouldn't go", and have the other respond "Yeah but Google Maps wouldn't have recommended it if it was. So it must be safe". And that would be a logical counter-argument. So no, the Google Engineer's defense isn't actually very good. In situations like this, yes, it MIGHT be better for a mainsteam product to default to "no safe options", similar how you can check "no toll roads" in your settings and only get directions for free roads. Suggesting sketchy back roads should be a setting that you opt in to explicitly. Of course the next step is detecting what is a sketchy mountain road and what isn't, etc. "Paved" might be the threshold I would use. "Include unpaved roads" is the setting I would add - defaulted to off.

dap 5y ago

> The Google engineer had a great defense in this article: >> "[People are suggesting] if you can't get driving directions and road closures right 100% of the time, you shouldn't offer any. Surely that would be a worse trade-off than the status quo?" he wrote. "Consider that we don't know a priori which mistake is potentially dangerous." That's a terrible argument, and it's even worse in context[1] because the person he's replying to didn't start from the position that "Google Maps shouldn't offer any routes ever". They asked if it were reasonable for Google Maps to express the known uncertainty in specific situations and the engineer seemed to imply that wasn't workable. It's frustrating because (as people keep pointing out) these product choices can result in death (regardless of whose fault you think that is) and the response sounds like they just don't care. I'm sure the problem really is hard. It's also hard to believe Google couldn't do better. It's known that roads in certain areas can be especially treacherous in bad weather. The NWS posts official warnings about bad weather, including time ranges, affected areas, etc. It's not hard to imagine some heuristics like "in the Tahoe area [really], when the NWS has posted a winter weather warning, strongly prefer the main roads". Sure, this is harder to scale. But if that's the issue, then it's not "we couldn't possibly do better". It's "we're not willing to invest in what it takes to do better". (I get that even this idea might be flawed for reasons I don't know -- but I'm certainly not convinced from the argument in the article that there's just no way to do better.) [1] https://twitter.com/BuschnicK/status/1475750318294061060 https://twitter.com/BuschnicK/status/1475750318294061060

AlotOfReading 5y ago

My father and I have an annual tradition where we find a country and go get stuck in. I typically end up as the navigator and manage Google maps (among others). Having a dedicated navigator is the only safe way to use Google maps outside routine conditions, in my opinion. It often recommends incredibly dangerous detours that aren't visible in the navigation UI because they're off-screen. The only information you get is "alternate route available 3 minutes faster" that you have to decline in 10 seconds or it will autoselect. If you're lucky enough to have a dedicated navigator, they'll still have to know to review the entire change every time and have pretty extensive knowledge of the area to know those conditions ahead of time.