Or the data is just old. Years ago, the junction between Route 1 and Route 36 in Iceland was very visibly wrong on Google Maps. If you pulled over and looked over the bushes, about a quarter mile south of the intersection, you could see the old Route 36 that was still in Google's road network, with crumbling asphalt and all. And if you looked at satellite imagery, of course there was no trace yet of the newer and more efficient fork.
The 80% accurate statement is about the maps themselves. Google maps will show a road that just doesn't exist. Or you will be driving down on a road that isn't on the map. Or the map will show the road going through, even though you are at its end, looking at a cliff. It will show intersections that do not exist.
In some cases, I can tell they used old data. I know one road in particular that did exist 30 years ago, but the forest service deliberately closed off when a rock quarry closed, and it has been re-forested. It still shows on the maps. Another that shows on the maps was planned to be built by the BLM, but never was... yet the map shows it. So I'm not sure exactly what data set they built their maps from, but clearly it did come from data, not actual exploration... because the dirt roads in the American west are just not accurate.
Google maps also gives dangerous directions in several places in Iceland, directing traffic over old mountain roads that save a kilometer or two but take much longer because they are poorly maintained. It also makes no note of closed roads, though many roads in Iceland are only open seasonally, and this information is easily available.
This reminds me of how in some ways online maps are far inferior to what the old road maps like the ones Rand McNally printed thirty years ago.
The Google map doesn’t have symbols to distinguish major roads from minor ones. For instance, in the US a printed map would typically have different symbols for free limited-access highway, toll limited-access highway, multilane divided but not limited-access highway, major two-lane road, minor two-lane road, local road, gravel road, and dirt road. Google has none of this; it just has lines of vaguely different width and color intensity.
So in the US I know that if the route has an Interstate highway I know what to expect. Typically a US highway is at least a major route, though you can’t count on it. If it’s a state highway, who knows. The map tells you little about the size or condition of these non-Interstate routes.
So if Google routes me down a non-Interstate route, sometimes I will look at the street view to gauge it. If it shows a big divided highway, I’m good. But sometimes it shows a two-lane road with no shoulders, multiple driveways, and hills blocking the view. Completely unacceptable for driving more than a few miles if there’s absolutely any alternative, yet Google will send you down these roads for two hundred miles if it thinks you’ll save ten minutes—-which you won’t the instant you get stuck behind some slow dump truck.
I don’t know this situation in Russia but years ago on a paper map in the US, that deadly route would have been shown as a minor road on a paper map, in it was on there at all, while the major, safer route would have been shown prominently.
I just marvel at how tech takes us forward in most ways but how the old tech had superior elements that simply don’t get replicated even decades later. (Another example is how old print maps showed rest areas and even distinguished them from simple no-restroom pull-offs. This information is not on Google at all.)
I was going to a location yesterday that you couldn't search for. "The junction which splits the road going to A and B". I was very confused by the new Google Maps which essentially didn't show enough detail for a good overview and orientation.
As a counter anecdote I marked a road as changed (one end was closed off) and it was fixed in a few days. I have made quite a number of edits to Maps so to me this is not at all what I have experienced. Maybe Google is different (IE. worse)) in the US than Scandinavia.
Oh, this is crazy. So Google still provides directions across that bridge, and has data collection from May showing (qualitatively) the bridge is out. But when you go to StreetView, you can't hand-navigate across the bridge in the 2023 data, but if you snap back to the 2012 data, you can... And Google will let you navigate back and forth, then snap back to 2023.
Looks like the revision in 2023 didn't remove the road from the 2012 dataset, and Google's map resolution algorithm is chaining back to an older revision to find the connectivity that justifies the path. What a mess.
Despite all of Google' effort, they still rely too much on USGS published maps. For example, just 6 miles from the GooglePlex is a road in Newark with a gate [1], but Google doesn't know you can't drive on it.
They're doing a good thing, but unless you can see that Google has driven the route and has published street view, you can't entirely trust them.
It's not, but in this case I suspect it's giving us a clue as to how the algorithm broke. Here's my imagined failure mode:
1. Google collected new data in 2023
2. That data didn't include a turn down the busted road (because why would they bother, they can see the bridge is out from up the hill)
3. Since there was no turn down the busted road, there was no new data collected for the busted road
4. They ran a fusion step to build a connectivity graph from snapshots of traversible world. Here in particular, that would have been necessary because there isn't much visibility on the road from the sky either, due to the tree line.
5. Their algorithm considered street-view-collected data to be highest-quality. Especially new street-view-collected data.
6. The connectivity graph went "Hm, I'm missing new data for this road segment... Oh, but I have some from 2012. I'll just assume we missed collection here and fuse the old data in to build a coherent map."
7. This fusion step killed all the "Hey this road doesn't exist" edits because it kicked out a new, "pristine" map from real-world measurements. In essence, the fusion step improperly updates the system's belief on recency of data so that the 2012 data is treated as fresh as the 2023 data and overrides older (2022 etc.) reports the bridge is out.
Steps 6 and 7 here are bugs, but it's a subtle enough bug that I can imagine the Maps team failing to catch it until disaster hits.
I don't think gmaps actually does that. We have a 20 meter stretch of road near our house that connects two other roads. Google, for at least the last five years, has refused to believe that this road can be traveled, having everyone take a detour. Hundreds of cars are taking this non-existing road every day, but Google still hasn't taken the hint. (Probably preventing quite a bit of through-traffic in our neighborhood, so we're actually rather happy about this. :-))
Yes, it is. There is a comic that is internally shared that describes it well. A Googler comes to a fork in the road. On the left, there is a road that is worn out, with a sign saying "deprecated, don't even think about it"; on the right there is a brand new road still being constructed with a sign "under construction, danger!".
Google Maps has tried to route me through closed off bridges in the relatively recent past. I don't have a huge amount of confidence in the validity of its data.
I live in a reasonably rural part of the UK in an old farmhouse with historic public rights of way criss-crossing the land around here. One day google decided that a footpath through a wooded area just wide enough for a quad was public highway and started giving driving directions through it. There's no place to turn, the gradient is steep and, once you're committed, the road surface is, errr, mud. Cue umpteen drivers getting stuck, panicing or grounding out on the occassionally fast flowing ford at the bottom.
I'm guessing image rec on some new satellite imagery convinced it it must be a road.
No amount of reporting a mapping problem to google has changed anything. When I reported something similar in Apple maps a couple of years ago, the change was reflected within a couple of weeks.
It could simply be a map projection issue - where the traffic data projection is slightly off by an order of 1000 meters. This datum shift could be easily fixed by Google.
Another example: a lot of rural areas have patches of gravel roads. It's not obvious from Google maps where the roads go from paved to gravel (and they sometimes do it in random sections).
Naively looking at a map, you'd think a roundabout route had potential to be shorter, but not realize the road was gravel.
Same in Netherlands. Google Maps has recently become worse and worse. Small pedestrian roads are missing, and the navigation is getting stupid. Google applies the one-way limitations for cars also on cycling routes.
Not sure if it's a US vs Norway thing but Google Maps keep confusing me by saying "keep left at the fork". Like what fork?! Oh, you mean don't exit the freeway...
reply