Hacker News Reader

Comment by enneff | Hacker News Reader

No "real map" would have this error because of the human element: you wouldn't plot a town where there isn't one, particularly if you were also responsible for plotting the roads of the town.

This is a data processing problem. Someone's algorithm screwed up and they didn't bother to have people sanity check the results.

View on HN

you wouldn't plot a town where there isn't one

It happens. About 12-13 years ago, a colleague and I were driving around Ohio to document landuse near a freight railroad line. The maps we had (USGS, AAA maps and local street maps) indicated a small town in one location, but when we drove there, there was nothing, no remnants, no anything. When we got back to the office, we looked into it- turns out the town had been wiped out by a flood a few years earlier and the decision was made to disincorporate the town instead of rebuild. That town appeared on maps that were printed after it was abandoned.

Perhaps this doesn't happen anymore with more recent mapping techniques/technology, but it's not unheard of.

Sometimes mapmakers add fake features deliberately: http://en.wikipedia.org/wiki/Trap_street

But there's a difference between old/outdated data and putting towns in completely the wrong place.

I've used map in the military - of Afghanistan - that were relinquish of old soviet maps. The soviet would have never got in many places of the desert and had mapped villages and roads from aerial photography. Those features would remain on our (US) maps.

As they rushed into the war, the geomatic guys responsible for making maps somehow reused the soviet maps and there was no way to assume those roads and villages didn't actually exist. So they persisted on our maps, up until 2010. Data around main areas would end up getting corrected (patrols would complain about the inaccuracy and the geo guys would update their database). Data in the desert would remain erroneous in many case.

For instance, I could simply look at a village's location on satellite imagery, such as a casual Google Map, and clearly see there was nothing there but maybe some odd looking shapes and shadows due to rock formations. So even hand-crafted maps, in military operation, almost 10 years into the war, can remain inaccurate.

Worst, maps of many military bases in North America are completely wrong as soon as you get off the few main roads listed on it.

I came to rely on satellite imagery whenever you want to go on a somehow non-urban track. It's much more reliable than using a map.

But you might put an island where there isn't one .. Google has. The difference is that Google appear to have quickly acted to remove the island when it became public knowledge that it didn't exist.

http://www.cosmosmagazine.com/news/6234/%E2%80%98missing%E2%...

Te Manukau harbour was removed from New Zealand by Google. And that's a huge thing to get wrong. Neither Apple nor Google have New Zealand sorted at all.

Consumer GPS is somewhat notorious for giving nonsensical directions. See the Daily Mail's endless stream of "but I was just following the satnav!" stories, for example.

No one should see the Daily Mail for anything.

> you wouldn't plot a town where there isn't one

Seems to happen even with whole islands: http://www.smh.com.au/technology/technology-news/where-did-i...

I've heard of this happening before. I remember reading something by Bill Bryson, probably from the 90s, where he returned to the US to retrace some of his childhood journeys. After being used to British maps and their fantastic detail he laments the state of US maps as they miss whole towns and so forth.

The issue is slightly more subtle/nefarious. See this article, for example.

http://www.heraldandnews.com/breaking/article_ea780435-4e3a-...

But this isn't the fault of their GPS putting towns in the incorrect location, this is blindly following your GPS into a snowbank.

But yeah, your original point still stands: you should always use common sense when following GPS. In this case, it would be common sense to assume that the road to a major regional town wouldn't take you into back streets of a national park...

There is a human element in using GPS/smartphone maps too though. You can scroll to the end of the route and see if it looks sensible, rather than just blindly following it.

Reply on news.ycombinator.com