OpenStreetMap

Motivation (or how to get more of it)

Posted by Chaos99 on 8 November 2010 in English (English)

Almost every day I'm out and about roaming my surrounding with a GPS device collecting track data.
I've found a ton of roads and ways not already in the map. As I use OSM data for in-car routing, I also find a lot of missing or
erroneous turn restrictions.

Thats cool right?

No, it isn't.

'Cause all this data ends up in a big pile of files either on my GPS or on my computer, never to be uploaded and integrated to OSM.
My memory fading away with time on where those turn restrictions were and what type of way it was I wandered.
I'm quiet sad about this myself. So I wondered, why don't I find the time to incorporate all the data into OSM?
Well I'm actually doing a lot of data uploading, writing descriptions, posting photos, vote on things and so on.
It's just that it is geocaching.com that I'm on, and not openstreetmap.org. So I still wonder why. What makes geocaching so much more
motivating than OSM? (Follow this Wikipedia link if you don know what geocaching is.)

Well I can only speak for myself. But I'm pretty sure this extrapolates to quite a lot other people too.
The basic activities on OSM and geocaching are quite similar. You go out and carry a GPS with you. Either finding things according to coordinates or finding things to record its coordinates.
Then you return home and record what you have done. Either by writing a log description on geocaching.com, or by tracing your GPS track in JOSM.
I admit that using JOSM is slightly more complicated and time consuming than writing some text into a website form. But that's not the point. I'm a geek, I love complicated things. I do them for fun.

What separates OSM and geocaching for me is what you do after you uploaded your data. On OSM all you can do is sit back and marvel at the map, which will show a tiny new way or mostly just an existing way in a slightly new position and maybe render style. That's it. All your work meant that approx. 20 pixels changed its color in a sea of data generated by other users.
Well of course if you are in the top league of contributers you will show up in some lists, or you may manage to bring a complete new village onto the map (I managed to do that myself a long time ago). But for normal day-to-day editing, you won't get anything noteworthy.

With geocaching on the other hand, you are rewarded for every find. Even without any further work on your own, you will be presented with your number of finds and your finds for every type of geocache. With some work you get a massive load of statistics to put on your profile page that you can compare with the ones of your friends. In which countries have I found caches, how often were I the first to find, what difficulties have I found, how far away from home have I found caches, on which day have I had my most finds, whats the eastern-, western-, norther- or southernmost cache I have found, what the farthest away from home, what was the oldest cache I have found? That's to name just a few. With just a little bit more work on your own, you will receive badges you can present on your profile page. Thats like Bronze for finding more than 250 Micro-Caches, Gold for finding a cache above 1500m or Sapphire for finding a cache every day for 70 days in a row.

I may foresee two objections here:

At first you may argue: "But who likes statistics anyway?" Well, although I won't normally admit it: I do. I already mentioned that I'm a tech geek, right? And I'm sure there are many more on OSM. I'm actually German too, which somehow means I like dull things like statistics, bureaucracy and football. And at least one of these things has to be true, hasn't it?

The second argument may be: "But geocaching is just a game, OSM is serious!" Well again, OSM may have serious fields of application, but that doesn't mean that you can't just handle it as a game as well. Actually OSM may benefit from being more of a game. As all crowd sourced projects it lives and dies with its contributors, so it has to keep them happy. You will always keep a core group of activists even if your project is boring as hell, but look where the German wikipedia is heading with this. So keeping the project fun is a core aspect of keeping it alive. And games mean fun like nothing else. You may want to watch this TED-talk on a similar subject.

So, as a conclusion, I would like to propose to come up with some statistics that can motivate people to keep on uploading data. (Length of ways added, number of different POIs, farthest from home... almost all things from geocaching will translate quite well.)
Design some badges, medals, trophies or whatever to present on profile pages for certain milestones reached.
The How did you contribute to OSM Page of Pascal Neis is a really god start.

This would, at least, motivate me to do more work on the map.
So please give me a lot of feedback on this. Do you think this will generate more motivation? Do you think it's bullshit? Are you able to design some badges or do a statistics page?
I would love to hear your thoughts.

Comment from giggls on 8 November 2010 at 10:42

Different people like different things :)

I don't even have an account at geocaching.com because I think geocaching is a waste of time. Mapping stuff for OSM on the other hand is a very useful thing because of the resulting free data which can be used for countless things.

Its all about free data IMO and geocaching.com stuff is not even free.

Hide this comment

Comment from Chaos99 on 8 November 2010 at 10:56

Well, sorry if this didn't came out clear enough in my post. It's not about promoting geocaching or geocaching.com here.

It's about how to apply things that work for geocaching (4-5 million people) to OSM, so we get even more free map data.

But lets just count that as +1 for people who don't need more motivation.

Hide this comment

Comment from keinseier on 8 November 2010 at 11:21

Best idea ever, I would like to see what I have to do to get some badges ;-)

Hide this comment

Comment from Hawkeye on 8 November 2010 at 12:38

Great idea, I think it would be nice to have a bit of positive feedback like badges and cups etcs.

Hide this comment

Comment from lebsanbe on 8 November 2010 at 13:36

Achievements would be great :D

Hide this comment

Comment from Richard on 8 November 2010 at 15:29

Wikimapia shows why this sort of thing should be treated with great caution. Users can come to value their own "score" over and above the real aim, of making a better map. They then carry out edits that improve their score but don't improve the map (in fact, often worsen it). In OSM, this might be tracing areas which they've never visited and will never do so, or making questionable bulk changes.

We want to encourage people to get out there and map. We don't want to encourage people to change for the sake of change.

So if such a system were implemented, I'd very strongly argue that points should only be earned for areas where the user had uploaded GPS tracks. Such a system would reward surveying effort, and not armchair mapping.

Hide this comment

Comment from Tim Litwiller on 8 November 2010 at 16:31

I for one would love this. I have a phone that 'droid that almost always has 2meter or better accuracy but I keep forgetting to use it when I go new places.
all my usual routes are already well mapped.

Hide this comment

Comment from Tim Litwiller on 8 November 2010 at 16:34

oh -and I totally agree with Richard there could be some minor rewards for "armchair" mapping but the real rewards should be for GPS tracks uploaded and the quantity/quality of detail on those tracks.

Hide this comment

Comment from Chaos99 on 8 November 2010 at 18:45

@Richard I'm with you on that point. Any point- or reward system has to be carefully balanced so that it doesn't encourage behavior destructive to the map. This won't be easy, especially to enforce this on the technical side.

I'm currently looking into what data is available for this purpose.

But as with any purely artificial point system, it's all about the social component. If it is not recognized by others, it's not attractive to the single user. I think OSM contributors are a quite fair and responsible group of people. I think geocachers are worse for that matter ;)

Hide this comment

Comment from Vclaw on 8 November 2010 at 19:56

There are awards that can be given to people who have made a particularly worthwhile contribution to OSM: http://wiki.openstreetmap.org/wiki/Awards
Though they're not used much.

Hide this comment

Comment from Sanderd17 on 8 November 2010 at 20:13

look at this:

http://blogs.thehumanjourney.net/finds/entry/openstreetmap_the_game

I wanted to help develop it, but Mr Reeves didn't have time for this and I don't have the skills to do this on my own. So the project stopped with an idea.

I believe this would work if the gamers would get points if the upload there data for approval. If the data contains too many faults, users should be punished with penalty points.

Meanwhile, gpx traces can be collected.

Hide this comment

Comment from Chaos99 on 9 November 2010 at 07:26

@Vclaw Didn't know they existed. Thanks for that. But they are way to rare to be useful for motivating the masses (like me).

@Sanderd17 Nice idea too, but not quite what I meant. It's not about creating a game and then using its data as a byproduct. (Game design is a very very hard job.) But just to give some gamey aspects to the OSM activity itself.
I especially see the problem in the need to postprocess the data of your game proposal and the problems you face in sparsely populated (by gamers) places, where you don't have enough data to validate inputs.

Hide this comment

Comment from Sanderd17 on 9 November 2010 at 11:34

I don't want to get gamey aspects into OSM itself, it can create the editcountitis disease. Since a game is in most cases a competition (third point in the symptoms list) users would have more chance to be caught by editcountitis. That's why I prefer that the game is separated from the map itself. There has to be a layer between both. And the point's you get in the gamey aspect may not be related to the edit counter of OSM.

Hide this comment

Comment from Chaos99 on 9 November 2010 at 11:59

Woah, I didn't know there's a word for that. Thanks for that link.

The competition (even if it is just between you and yourself) is exactly what causes the inceased motivation. So it is not a by-product, but the target.

I don't deny that this can cause damage. Even if implemented with most caution.
The question then remains: does it do more good than bad? Can this be balanced? How much 'bad' is acceptable? If it motivates 10 people to contibute more, and one of them overdoes it and starts shifting nodes to get last_edit_by status, is this still good? Let's discuss this.

@Sanderd17 How do you think the 'layer between' will help here? And isn't human review the only layer that would be affective? Love to hear your thoughts.

Hide this comment

Comment from Sanderd17 on 9 November 2010 at 13:37

The layer between has to be restricting, e.g. it shouldn't be possible to start shifting nodes, to draw maps of regions where one has never been and the points you earn may not be a one-to-one mapping with the number of edits.

I agree it's possible to have a special edit counter, which only counts e.g. the creation of a way if you have a gpx, or the creation of a POI if your gpx shows you've passed there, but a user can still upload non-existing POI's and it is difficult to distinguish gamers and workers.

If we think of the 'layer between' as the game in my first link (which only accepts certain changes, no shifting nodes or useless tags), then you could reward a user for confirming a POI without changing the map. This find=reward and confirm=reward method will bring better quality to OSM. The game can also confirm that the uploaded or confirmed nodes have been visited.

bear in mind that a typical gamer will fraud if he can get his points up. The gamers will only have a minimal relation with the map, so they won't bother about the quality of the map.

Off coarse, a user could make 2 accounts and always keep confirming his own POIs. But if you only allow a certain order in which POIs have to be found, it will be more difficult. If you only allow that a user can find a POI only once, it will be even more difficult to fraud.

This way, there is human control, but there is no need to do it manually. If a POI gets confirmed twice, it could be uploaded automatically. We could assume that certain (not so important) tags don't contain errors or fraud (like opening hours and others) and upload those tags when they are found for the first time. If a user is already present at the POI, why would he fraud on tags like that?

as a second option, next to the automatic upload, the database of found but not confirmed POIs can be put in a map for OSMers. If an OSMer looks at the map near his house, he can see the found POIs and put the ones he knows on the map without interfering with the game. If he sees an error, he can report it.

This way, every user can choose if he wants to play or not, editcountitis will be reformed to simple competition, the map will get only the positive results of a fanatic gamer and current OSMers can continue working like they did. All those filtering and counting things are difficult to implement inside OSM so therefore, an extra layer (in the form of a game) is needed.

I know the idea is not complete, e.g. if a fraud is discovered by an OSMer, the uploader should get lesser points, but this could cause the gamers to put there POIs on the map themselves, so the fraud doesn't get discovered.

Hide this comment

Comment from Sanderd17 on 9 November 2010 at 13:54

Oh yeah, about "the problems you face in sparsely populated (by gamers) places",

Since points aren't a one-to-one relation with the number of edits, the gamer can get more points if he finds or confirms a POI in a sparse mapped (or sparse populated, we can't see the difference with OSM data) place. And in sparse populated places, OSM will always be worse than in big cities, there's not much to do about it.

Hide this comment

Comment from Chaos99 on 9 November 2010 at 14:15

Hm, I see a conceptual problem here. If you build your system on the assumption that all gamers are evil and will try to fraud, then don't bother creating a game at all. There is no way to build a tamper-proof game system that is still fun.
GPX files can be manipulated just as easy as any other file, so there are worthless as verification device. If players are likely to give false initial tag data to achieve more points, it's just as likely they will just waive through any already present data if they get points for that too. So thats no verification device either.

My assumption is that
1. Gamers and mappers are not two distinct groups. They overlap by at least 90%.
2. Gamers are good and responsible people, with just the same amount of vandals as in any other group.

What about that: Achievments are rewarded with the current database status in mind. So the "Most different amenities in one day" award may be gotten by cheating and entering false data. But as soon as someone recognizes the 'mistake' and corrects it, the award vanishes again. So the evil cheating gamer has just a very short win from his behavior. Maybe his award actually provoked a good mapper (or jealous other gamer) to go out and check the data. Well, thats win-win then. Of course this could lead to edit wars, but chances are slim (with my assumption that vandals are rare) that two nutcases find each other.

I don't think your game idea is bad by the way. I just don't think it serves the purpose I had in mind (motivate existing mappers to do more). It's more of a OSM-based game for other people not currently involved. Also I see that implementing your idea is waaaayyyy more comlicated, which mostly means that it won't be realized soon unless your are a very skilled coder yourself. (Well I'm not.)

To keep track of vandals and users who try to fraud the system, maybe the awards should not be generated by the users themselfes, but by a website where you have to log in with your osm credentials. So if someone goes a bit crazy on the whole thing, it's more easy to stop him or even the whole project from doing more damage.

Hide this comment

Comment from Sanderd17 on 9 November 2010 at 22:25

About the GPX files: those can be encrypted before they are sent. So they can serve for verification.

I know my idea is a lot of work to implement (different apps for different smartphones and a website), I've also said that I don't have the skills to do it properly. I would be able to help here and there, but I can't make the complete project.

I know I can trust the mass of users, but I also know that there is Like you said, a separate website would be good, if you subscribe, you calways a minority who tries to fraud (or in Belgium, it's a majority who does tax frauding :P). You can let the mass catch the fraude (like in wikipedia and OSM). But chances are that if users are attracted for competition, they will fraud more (I know you have a different assumption) and maybe the mass can't handle it anymore or is sick of handling fraudsters.

The game is indeed to attract more users, I thought you meant that (you mentionned the number of geocachers).

A separate website would be good. Maybe, for attracting more OSMers, make something that can compare your edits to OSMers around you, so one can say "I'm the best of New York", or "I'm the best of madagascar" instead of "I'm in the top 100".

I believe that not forcing OSMers to join the gamey aspect of OSM is a must. So if the site were separated, it wouldd be better.

Hide this comment

Comment from Niels Elgaard Larsen on 11 November 2010 at 00:10

You could make games without jeopardizing quality.

For example you select 25 random, unmapped roads in a given area (e.g., Slagelse, Kalundborg, Holbæk triangle). The list is secret. Players go mapping and claim a road when they finish mapping it. If a player claim a road on the list, it is verified that it is a good mapping, the score is updated, revealing the name of the road. The other players have a couple of days to protest based on quality.
The winner is the the player with most claimed roads when all roads are claimed or at some deadline.
The point is that only the roads on the secret list have to be verified, which is hopefully a small fraction.
You can add various ways of giving hints to players. The more hints the larger area you can have.

Hide this comment

Comment from Chaos99 on 11 November 2010 at 07:40

@Sandered17 Encrypted GPX? You mean writing an own logging application for every smartphone OS + ruling out any Garmin or other dedicated GPSr? Don't think so. Sorry.

But I see we can agree on a seperate website. And forcing someone to participate was never my intention. The crux lies in the way the game data/goal needs to be decoupled from the real OSM data. I have to do more thinking on that.

@Niels Another good idea. Some kind of 'task of the day'. The hidden list is also a nice game aspect. I just think it doesn't scale well with global players, as you are by definition very local if you put specific street names in your list. If you go broader (i.e. map 15 post offices) you are loosing the 'hidden list' aspect and therefore the fraud protection that comes along with it.

Just to be more clear of what I aimed at with my proposal (which by the way doesn't claim to be perfect or even finished): I was looking for a way to motivate existing mappers to do more. I was proposing a reward system with doesn't need a change in behavior by the mappers, but rewards them for their normal work. But yes, I see the danger in attracting fraudsters without mapping, but only gaming interest.

But please keep the ideas comming. It just can get better.

Hide this comment

Comment from Sanderd17 on 11 November 2010 at 13:20

@Niels: where do you get the secret list from? you need a database to that. I know, in my area, all databases of roads are closed.

Maybe you can say "trac the roads, at the end of the week/day, 20 roads around the world are chosen (randomly) and verified".

Advantage: worldwide and it keeps the same idea.
disadvantage: someone has to control the roads all over the planet. this would require a large user base. And if the user base grows, the number of roads needs to grow, which means more controlling.

Hide this comment

Leave a comment

Parsed with Markdown

  • Headings

    # Heading
    ## Subheading

  • Unordered list

    * First item
    * Second item

  • Ordered list

    1. First item
    2. Second item

  • Link

    [Text](URL)
  • Image

    ![Alt text](URL)

Login to leave a comment