[HN Gopher] Google Maps may have led Tahoe travelers astray duri...
___________________________________________________________________
Google Maps may have led Tahoe travelers astray during snowstorm
Author : nradov
Score : 12 points
Date : 2021-12-28 21:37 UTC (1 hours ago)
(HTM) web link (www.sfgate.com)
(TXT) w3m dump (www.sfgate.com)
| PragmaticPulp wrote:
| 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 wrote:
| 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.
___________________________________________________________________
(page generated 2021-12-28 23:02 UTC)