Has Google Maps used my OSM updates for their Map?

General talk
1

I’m not sure if that’s allowable or not, but that’s my observation. I might be wrong–maybe someone can show me that Google has had the main trail built out this whole time, and I simply missed it–if so I’m happy to be wrong.

I’ve been trying to build-out OSM in a friend’s area, roads, navigation, even foot paths. One area I’ve spent some time on is nature trails, I believe there’s value in mappinng known foot paths, so that’s what I’ve been doing. When I first started, Google Maps showed a very rough, single line, for the trail I’d been mapping, except the North portion where it properly showed a loop. It also only noted 2 of the main trails by name (Central, North).

Over the past few weeks I’ve spent some time mapping the trails, and trying to get them as accurate as possible in OSM, so re-visits occur. I’ve also added all the proper names for the trail where available. Today, I looked at Google Maps’ version just to compare the “aesthetic” differences (because it’s fun to compare your own hard work to that of a global conglomerate’s and see that you’ve done more lol), and I noticed that Google Maps suddenly has built out more of the trail, added more names, and even marked one as a normal footpath (for some odd reason, because it’s not). What really caught my eye was one of the utility trails I mapped.

I mean, maybe they’re allowed to copy OSM, and maybe they have people walking the trails (this is doubtful). They aren’t using trail maps given by the local City government, because the City doesn’t keep trail maps, the only one they have is outdated and doesn’t mark which bridges no longer exist–I’ve been doing that in OSM. If Google had someone walking the trail, they wouldn’t have drawn a bridge in place which doesn’t exist, but that’s apparently what they’ve done.

Here’s a link to the google view: https://www.google.com/maps/@36.742661,-76.2340444,16z

Here it is in OSM: http://www.openstreetmap.org/#map=16/36.7408/-76.2351

2

Google maps has significant crowd sourcing these days. In the same way that some people think it OK to copy from Google to OSM, I suspect there are GM contributors that think it OK to go the other way.

As such, it wouldn’t surprise me if this has happened, without Google’s authorisation, or encouragement.

It is also possible that your mapping, has led a Google maps contributor to independently survey the trail.

If the trail has been copied by one of Google’s contributors, I’m not entirely sure who has the right to issue the DMCA take down notice to Google, i.e. whether it is the original OSM mapper or OSMF. Generally conglomerates will only accept copyright violation notices from the owner of the rights.

3

Google definitely hasn’t copied from OSM, at least in this case. This can easily be seen by using Map Compare. It’s easy to see that the trails don’t match up in many places and many that are in OSM are missing in Google.

4

Thanks for the responses from you both. That map compare tool is neat. There definitely are deltas, I was really curious if there’s a way to see Google Maps ‘back in time’ specifically on those foot paths, because like I said in the OP, they never had the trail built out the way they do now–it’s still missing a lot, but they’ve added details which aren’t even on the official brochures.

I didn’t know they had independent contributors though, that is the likely explanation. Thanks again.

5

@hadw google does not have significant crowd sourcing of linear features since they have turned off map maker.

@taz2015 your question is nearly a FAQ: but in any case: there are no recent verified cases of google using OSM in any way. There was -one- historic incident in which a third-party supplier used OSM data to fulfil a request by google (I would be surprised if that company still exists).

In general, outside of the limited areas were they collect their own data, google simply purchases the data from governments and other geo-data suppliers. Similarities in geometry are typically simply due to the fact that the objects mapped are the same, and in some cases that we are using the same third party sources.

Naturally keeping an eye on things is good, but the hurdle to show real copying is very high.

Simon

PS: @taz2015 nice work!

6

Thanks for the info, and the kudos.