OpenStreetMap logo OpenStreetMap

Pokemon Go trash

Posted by BushmanK on 28 January 2017 in English. Last updated on 29 January 2017.

As it was recently discovered, Pokemon Go users have found (or, at least, they think they have) that OSM data affects Pokemon “nests” location. Some of them immediately started adding fake map features. Here are several examples: Washington, Indiana Washington, Indiana, United States, this user https://www.openstreetmap.org/user/forever2darkness/ , fake residential areas and footways. (Reported to DWG)

Charlotte, North Carolina Charlotte, North Carolina, Unites States, this user https://www.openstreetmap.org/user/Fishboy35 , fake parks, with the obvious names like “not an actual park” (Reported to DWG) And guess what? He got a welcome message for adding these fake features https://www.openstreetmap.org/changeset/45538011 > Hi and welcome to OpenStreetMap! You’ve been adding some great detail to the map since you’ve joined. We count on local knowledge to make the map a better place! That said, please make sure you’re only mapping real objects. Things you add go into the live database, which is depended upon by businesses, governments, and even humanitarian organizations. If you have any further questions about mapping, don’t hesitate to give me a shout! Best, Ethan aka FTA

It was sent automatically by https://www.openstreetmap.org/user/FTA . How nice of him.

Assuming that they are interested in adding fake lines and areas, not nodes, this Overpass query could be used to load objects, last touched by a specific user, to look into one’s activity. http://overpass-turbo.eu/s/lzL Keep in mind, that not all changed objects are added by such users, some of them are only edited. Do not delete all of them without examining them.

One important thing: I am not saying here that all mappers, who are Pokemon Go players, are adding fake objects. I am also not saying that they can not add real ones. I’m only saying that to prevent adding fake objects in the future, it is important to delete these as soon as possible after they were added. And only after that anyone could try educating these guys about OSM.

Update: a certain DWG member I’ve contacted preferred to preach about being polite with new users and teaching them to be nice instead of helping to remove that trash. It seems very “cooperative”.

Update #2: It looks like at least some of obviously fake objects were removed by DWG reverting robot.

Discussion

Comment from SomeoneElse on 29 January 2017 at 01:20

Taking the points raised here from the top…

As it was recently discovered, Pokemon Go users have found that OSM data affects Pokemon “nests” location.

I’m not convinced that that has been discovered. It’s certainly true that some fake stuff is being added (as you mention, some helpfully marked “not actual park”, but also some water features such as https://www.openstreetmap.org/way/469266458 ), but I don’t think that anyone’s found the “smoking gun” yet showing that Pokemon game locations actually derive from OSM. We know that the game map claims to use OSM data in South Korea, but that’s a different issue.

So the next question is, what to do about it? There are two bits to it - the new users, and the data.

In the case of the data, if (as here) it’s obviously fake, then it makes sense to revert it so that those cars in Troutville are no longer parked in a pond. However quite a lot of the Pokemon Go new edits that I’ve seen are actually trying to map real-world features. They may not be making a very good job of it, but they’re not in any sense “vandalism”.

The other issue is the new users. In this case, the fake Troutville pond mapper has made exactly 4 edits. They likely have absolutely no idea what OSM is; they only know it as “source for in-game items” (and whether that’s true or not is an entirely different issue). Personally I’d like to think that there’s a non-zero chance that we can explain to them what OSM is and have them become a “real” OSM mapper. Ethan’s response seems entirely reasonable to me in terms of trying to guide them on that path - it may not work; but if it doesn’t we’ve lost nothing, and it doesn’t stop us fixing the data,

Finally, with regard to how to fix the data I suspect that a JOSM revert-based approach is likely to work better than an overpass find-and-fix one, as I suspect that that will miss stuff that other people have edited in the meantime - JOSM revert also flag up where there have been subsequent edits (which might be valid) that need to be considered.

Comment from BushmanK on 29 January 2017 at 02:34

@SomeoneElse,

The first question: are you aware of this Reddit thread? Because if no, then now you are. It doesn’t actually matter if Niantic is actually using OSM data to generate Pokemon Go “nests” - it is enough that there is a rumor that it is. I mean, enough for Pokemon Go players to start adding obviously fake objects.

I am perfectly aware of that kind of view, that if a person adds tons of fake information and just one real McDonald’s, it is probably a potential new mapper. But again, if we are trying to give data consumers a reliable map (or “the best map in the world”), removing numerous fake objects is more important than trying to teach misbehaving kids. That has never been an objective of this project. And nobody here has an obligation to do that. If you, personally, would like to do that - neither me nor anyone else will stop you. But that shouldn’t be done at expense of data quality. So, removing trash first, teaching - after that.

Regarding of welcome messages - let’s imagine a simple situation: there are several kids who discovered that the hallway of your apartment building is nice and warm, so they can hang out there, smoke weed and drink beer when it’s cold outside. One of them, out of boredom, fixed a door knob, but after that, they left empty beer cans, scratched their names on the wall and so on. What would you do: call the police first or express your gratitude for fixing a door handle? I’m not trying to convince you here, I’m trying to figure out, are you living in the real world.

And what I believe is that there should be another rumor, saying that all kinds of fake stuff will be more or less immediately removed, so they shouldn’t even try adding it. Because, regarding of “what should we do” - there is no way to tell if something is real or fake if there are no other active users in an area (like in these two cases above) or these kids aren’t “smart” enough to make their fake objects so obvious.

Comment from Glassman on 29 January 2017 at 03:22

I send out welcome messages to new users in Washington State. Even ones that added crap. I encourage them to add only what can be verified on the ground. Most of the edits are good faith attempts to add data. I’ve spent hours verifying parks against local database. It turns out that most are accurate. Maybe they don’t match shapefiles from cities or county shapefiles but they do add in parks. I’ve also seen a surprising number of edits from these Pokemon users. Typical new users do one changeset. I’ve seen a number with over 10 edits. It turns out that most of the vandalism is easy to spot. Parks polygons over residential neighbors is real common. Trails that couldn’t possibly be correct.

Do they have quality problems - yes. But mostly because they are not familiar with OSM. Abbreviated street name, paths/footways not connect to adj. streets, and untagged ways.

The last new person changed a street into a waterway. But interestingly he also correct the street name. Local knowledge is useful. I removed the waterway and removed the bogus footways he added. End result, we have better data.

I’m as frustrated as you with vandalism. But I think potentially we can gain some active contributors if we help them.

I’m well aware of the post on Reddit. See my comments on the OSMUS Slack Channel. I also know that the number of new users has increased significantly. In one day, I contacted as many new users as I used to in a month.

I’ve scheduled a meetup for our area for next weekend. One of the topics is editing for Pokemon.

Rather than fight Pokemon, we need to welcome them with encouragement to become active community participants.

Clifford

Comment from BushmanK on 29 January 2017 at 03:54

@Glassman,

Human beings are social animals, after all. All social animals have a concept of negative response on actions, unwanted within a group. And of a positive response on ones that benefit it. This is completely normal for people to say, that unwanted actions are “bad” while useful ones are “good”. Adding fake objects to gain something in Pokemon Go (probably) gives a person some profit (Pokemon creatures he wants), so that is a positive feedback. By sending encouragement letters, you can increase useful outcome, because these individuals don’t really have an intent to vandalize data - your encouragement doesn’t contradict anything they want. However, there is nothing in it that prevents them from continuing adding fake stuff because they haven’t received a negative response that “cancels” their profit. And that should be something significant because Pokemon creatures are important for them.

Your approach seems more like a method of teaching non-social animals, who have no concept of negative corrective actions and perceive them as a threat. Cats, for example. You can easily teach a cat to do some tricks by rewarding it for something you like, but you can’t make it stop peeing in your slippers with any kind of negative feedback - it just thinks you are an aggressor who has something against it. Fortunately, the vast majority of people are not like cats, even if sometimes they are quite far from being smart.

So, please, don’t imagine things like “it’s a quality problem” - no, it’s an intentionally and deliberately added fake object that even says in its name tag that it’s not an actual park or it’s a “Pokemon Go path”. Think about an actual motivation. They want their imaginary creatures closer to their home or school, that’s all.

As I said above, anyone is absolutely free to exercise being an all-loving person. But not in expense of data quality.

Comment from Warin61 on 29 January 2017 at 04:47

Local to me :- One has clearly state in their changeset comment that they are a Pokemon Go person and that is why they are adding stuff.. fortunately most of them are adding real stuff… some can do with some guidance, true of most new mappers. On the whole, so far, they are ‘good’. A few I have deleted entries .. and they have taken that Ok, and are now adding good stuff rather than making ‘bad’ errors. One looks to have been removed as a user… which is not the way I would go, I’d prefer more communication. Some of them may not realize that OSM is actually used by things otehr than Pokemon Go.

So … I think some communication and most of them should come ‘good’. Certainly OSM can use more active mappers here.

Comment from BushmanK on 29 January 2017 at 05:09

@Warin61, I appreciate your experience, but I think I’ve already addressed the same point you’ve made and I also have to remind that a single random case can not serve as an argument that disproves a tendency or a broad statement. In this case, the statement is: “There are Pokemon Go players adding fake objects, and there will be more”. So, the fact that somewhere Pokemon Go players have added something real, doesn’t contradict this statement at all.

There will be more, at least because more people will become aware of that rumor (even if it’s not true) and eventually, more people will start playing. I’m guessing here, but I think if it is actually about parks, some places could have way less real ones than other ones have, and these places are at higher risk of fake edits because people who live there want to have their creatures anyway.

What I am not saying is that we should somehow eradicate every mapper and every edit linked to Pokemon Go. Immediate removal applies, if it is possible, to fake objects only, as a measure to keep the database free of rubbish. Sometimes it could be impossible to do manually in a limited time, because, say, there are 100 changesets with 300 fake objects and 10 real ones. If someone would volunteer to do that manually and quickly instead of mass-reverting that, that person is welcome.

Comment from Alberto J (Deathan) on 29 January 2017 at 05:37

Hello, Pokemon Go player here. I have recently joined OSM because of the rumors of it affecting the game. I just wanted to contribute to the conversation that not all of us are providing incorrect information. I joined this Open Source just two days ago and have made around 40 edits, all of them strictly parks around miles and miles of my home location. I have edited around 20 parks, added new ones that are labeled on Google Maps and always provide the address and all information. Before I encountered this post I expected people to abuse the Open Source because of the rumors and have kept an eye out for anything out of place, so far I have found zero bad edits. My hopes for these contributions are accurate and truthful places for Pokemon Go players to explore and enjoy, I would hate to see a park with with barely or any Pokemon activity. My whole experience with Pokemon Go has forced me to learn about my surroundings and city, and now OSM benefits from my experiences. In my opinion it all benefits from each other as long as these contributions are truthful. If I do find anyone making these wrong edits I will make sure to fix and report their mess. Have an awesome day everyone!

Comment from BushmanK on 29 January 2017 at 06:03

@Alberto J (Deathan),

Good to hear that you have a plan to add the real parks, but do you know that Google Maps must not be used to improve the OSM? And you just said that you’ve added parks labeled on Google Maps. Unfortunately, you can’t do that.

As you probably noticed, I didn’t say that all Pokemon Go players are adding fake data, so, no worries - it’s not about any kind of discrimination.

And thanks for your will to keep OSM clean.

Comment from forever2darkness on 29 January 2017 at 06:26

Hi Pokemon go user here. I will admit that i did add some fake data to osm. To be honest i didnt really know that osm was used with so much stuff. Anyways not all stuff i added was fake. For instance the residential areas i had mapped were actually there. And most of the parks were too. I made the mistake of thinking you could mark anywhere you can walk in a park a footpath. I am so sorry for any trouble ive caused. It just sucks to see that alot of stuff isnt mapped. Ive looked in 3 towns and nothing was mapped except 2 parks.

Comment from Glassman on 29 January 2017 at 06:40

@forever2darkness - In the US we need more mappers in rural areas. Mostly we attract people in large cities, many with an active community. Our OSM meetup group has over 700 members. I’m real excited when I find an active mapper in rural Washington State.

One nice thing about mapping in rural America - you don’t have fights over how something is mapped. You are the expert. So go fix any bad mapping you did early and go improve your town. Join us on talk-us mailing list or join the our Slack channel.

If you need help finding your old edits just let us know.

Clifford

Comment from forever2darkness on 29 January 2017 at 06:44

Alright. I’ll try. Like i said i marked residential areas and that got deleted. Just to be clear that would be like houses, right? Anyways i wont abuse osm anymore.

Comment from pitscheplatsch on 29 January 2017 at 14:48

@All There are many changesets of forever2darkness, which should be corrected. See http://resultmaps.neis-one.org/osm-discussion-comments?uid=5201761

Comment from BushmanK on 29 January 2017 at 17:12

@forever2darkness,

I’m guessing, @Glassman could provide you a better help on this topic, but what I could suggest doing, is to carefully study how things are mapped in some European town, say, in the UK, Netherlands or Germany. That will definitely give you an example of how everything should look. No rough contours, no false features and so on.

And thank you for showing up here for a dialog - too often, people prefer to avoid it.

Comment from unglued94ta on 29 January 2017 at 17:15

Good morning everyone, from rural middle Tennessee. I too am a Pokemon Go player. I came here originally just to see what all was marked, so, IF the rumors were true, I’d know of a better place to hunt. Much to my surprise, there are a lot of things missing in my town (population 10k). So I joined to help the cause. If it helps the OSM project and other Pokemon Go players on the back end, then it’s a win win for both parties. I have only marked 2 locations as of right now, but will add more I’m the future. I added a playground and a park. There are more parks and several cemeteries that aren’t marked, that I’ll add next. My goal is to be a worthy contributor here.

Comment from FTA on 29 January 2017 at 17:18

Hi,

I think there’s some great discussion in the comments here. We have a major deficit of mappers in the US and the younger generation is a great group to try and keep on board. They are active, have great memories, and have a lot of free time.

Many of these users have added real and useful data–they just stray at some point and add fake data, so my intention is to steer them back in the right direction. If they continue to add fake data after being told not to, then that’s where blocks can be useful (possibly equivalent to a spray water bottle in the cat analogy).

I’ll also dispel a few notions possibly implied here.

You happened to catch the tail end of my DWG work for one evening where I had commented but not reverted. I had been reverting other fake edits that evening prior to this comment.

My commenting process is not automated, where something is never reviewed after I comment. In fact, this changeset was in my followup queue for the next time I had a block of my personal time to volunteer to the DWG, which was this afternoon. So please don’t conflate friendliness in the changeset comment with ignorance or an oversight.

It’s often the case the DWG gives mappers some time to respond before reverting.

Cheers, Ethan aka FTA

Log in to leave a comment