Google Contractors Vandalised OpenStreetMap Data

Fascinating. After the whole Mocality story, we were greeted by another story of Google misconduct. This time, it’s OpenStreetMap, which claims that users connecting from the same Google IP addresses in India as in the Mocality incident are vandalising OpenStreetMap data. Google has confirmed to ReadWriteWeb that two contractors acting on their own behalf while on the Google network were responsible. Another, less serious instance of Google misconduct, perhaps, but OpenStreetMap’s handling of this issue does smell fishy.

So, what’s going on here? OpenStreetMap is a collaborative project to create community-driven maps of, well, the entire world. These maps can then be used by navigation systems or other products. In essence, it’s more or less the Wikipedia of street maps, and a few years ago, I remember being invited to several OpenStreetMap mapping walks or bike rides or whatever, but since that sounded about as interesting as watching paint dry, I always declined. The project’s come a long way since those early days, since now they’re using aerial photography, and several large data donations have been made.

Today, though, Mikel Maron (OSMF board member), Grant Slater (OSM sysadmin), and Steve Coast (OSM founder) claim that Google has been sabotaging OpenStreetMap’s database (like Wiki vandalism). “Preliminary results show users from Google IP address ranges in India deleting, moving and abusing OSM data including subtle edits like reversing one-way streets,” Coast writes on his blog, “Two OpenStreetMap accounts have been vandalizing OSM in London, New York and elsewhere from Google’s IP address, the same address in India reported by Mocality.”

Coast states they’re in contact with Google, and the search giant has already let out a statement to ReadWriteWeb. “The two people who made these changes were contractors acting on their own behalf while on the Google network. They are no longer working on Google projects,” Google states.

Should we believe Google? Well, apart from the fact that companies ought not to be trusted, malicious behaviour by companies does have to make some form of sense – and it usually does. Google prioritising its own services in search results makes sense since it promotes the use of Google services over competing services. The Mocality incident makes sense because Google was abusing Mocality’s data to sell stuff.

Two accounts changing mapping data, out in the open, for everyone to see, easily reversible through the Wiki-style editing system? It simply makes no sense. There’s nothing to gain. Google has way more effective means at its disposal if it really wanted to harm OpenStreetMap. On top of that, Google is an OpenStreetMap sponsor; they have sponsored OpenStreetMap conferences, made donations, and OpenStreetMap also participated in Google Summer of Code. Why would Google pour money into a project while simultaneously vandalising it in this manner?

Hence, it would indeed seem that we’re looking at malicious contractors working in their own time, but it does raise questions about just how tight of a control Google has over its overseas operations. Contractors or no, in their time or no, the connection with Google is still there.

On OpenStreetMaps’s side of things, I’m not sure this issue was handled with the same kind of grace as, say, Mocality handled its trials. Interestingly enough, even the OSM team itself isn’t particularly unanimous on this front, as there appears to be some serious dissent within the team about how this matter is being handled. Tom Hughes, an OpenStreetMap system administrator who first discovered this incident, isn’t happy with how this issue is being handled, calling the blog post “grossly irresponsible and wholly inappropriate”, hinting that OpenStreetMap is trying to squeeze some press out of the Mocality incident (see the comments to the blog post).

“The board of OSMF are making mountains out of tiny pimples here. It seems that they want this to be some sort of organised corporate malfeasance on the part of Google which is why they have tried to link it to the recent Mocality incident where there was indeed clear evidence of such behaviour,” Hughes writes, “The reality in this case is that there is no evidence that this is any different to the numerous other incidents we get all the time where users either accidentally or deliberately make bogus edits. The only difference in this case is that there happen to be two accounts (though we do not know if that is two people) and the user or users involved happen to (presumably) work for Google.”

Applying the same rules to OpenStreetMap as we just did to Google, you may wonder what OpenStreetMap has to gain from riding the publicity train all the way to link station like this. OSM doesn’t seem to have too much to gain here, especially since Google is a supportor of their project, and thus, the saying ‘biting the hand that feeds you’ comes to mind.

However, we can’t ignore the fact that one of the authors of the blog post, Steve Coast (who also happens to be OSM’s founder), is a Microsoft employee working for the Bing Maps team. While the resulting bad publicity is something Google can only blame itself for, the rather bombastic presentation of it all by such an open project leaves a bad taste in my mouth, especially when taking the Microsoft connection into account.

In any case, it seems Google has some serious work to do in India, because this kind of stuff is wholly unacceptable. This and the Mocality incident are still peanuts compared to the kind of stunts Apple and Microsoft pull and have pulled, but with virtually every Google service having dozens of easily accessible competitors, they really can’t afford alienating themselves from their users.