OpenStreetMap
Open street map should include concepts from Gamification, to improve participation:
- Leader boards
- Chat between users editing nearby areas.
- Regional leader boards.
- Badges and stars.
Wikipedia has most of them - Can we implement the something similar?
Discussion
Comment from Sanderd17 on 16 January 2013 at 09:27
There’s an OSM fight: http://osmfight.neis-one.org/ Or personal statistics (which can be compared): http://www.hdyc.neis-one.org/ In the stats, you can also see the biggest editors of the day, week or month: http://www.openstreetmap.org/stats/data_stats.html
So there is a bit of gamification, but we shouldn’t advertise it too much. As it’s not our meaning to produce as much changes as possible (with armchair mapping, you can improve data quickly and cause a lot of changes), but it’s our aim to make sure people go out and map what they see outside. Data that can’t be simply derived from satellite images, and that can’t be found on other maps.
The chat could maybe be handy, but I think that the internal messaging system is good enough as it is now. In any way, I never work via the OSM site anymore, so you would have to include the chat in all editors available, I don’t think this is possible.
Comment from smsm1 on 16 January 2013 at 10:16
You may be interested in ITO World’s OSM Mapper where you can create areas and see who has been doing a lot of mapping in each area. http://www.openstreetmap.org/user/Sajith%20VK/diary/18403
Shaun McDonald ITO World Developer
Comment from robert on 16 January 2013 at 11:21
Many have been throwing around ideas like this for a while. Problem is, if you don’t get it right, people will just “game the system”. And it’s very difficult to get it right. A naive way of doing it would give people points for mapping things - but if a mapper were highly driven by points alone (kind of the point of gamification) they would just add thousands of benches in a park or something, getting thousands of points. Which is clearly not what we want.
Getting “points” to correlate only to actual valuable work is very difficult and anyone playing OSM solely as a game will very quickly find & exploit any cracks in the system where you can gain as many points as possible without doing much useful work.
It’s tricky.
Comment from Sanderd17 on 16 January 2013 at 12:29
I agree with you, Robert.
It doesn’t only depend on the things added, but also on the resources available.
I remember, when we had no images available, I mapped a single building by doing GPS tracing while walking straight towards the building corners. This was multiple days work. Now, I can add hundreds of buildings per day. But buildings without extra information (like address) don’t hold much value.
Now, there are still regions without good images, so where building are a lot of work and should be valued highly, but in a lot of areas, this isn’t the case anymore.
So even if you would succeed to award more points to one feature than to an other (which would be a lot of work), it would have to change per region.
I think the tools we have are good. They’re not too easy to find, so people who can’t be bothered with gamification don’t get them “in their face”. But for people who really want it, they can look up the stats.
Comment from robert on 16 January 2013 at 17:37
Of course, having said that, it would be great if someone were to find a neat way of making gameification work for OSM. I’ve been throwing a few ideas around myself but haven’t come up with anything totally solid (and have all my development time being spent on other projects).
Comment from Hjart on 18 January 2013 at 21:30
As someone who has already spent thousands of hours in JOSM and so probably would top a few lists, I would be very wary of gamifying anything OSM. Because how would you really measure the “value” of a contributor going out of his way to provide quality (takes his time to draw accurately rounded curves or spot all sorts of mistakes, etc.) relative to someone who (apparently) just cares about throwing lots of nodes and ways onto the map (I’ve noticed a few examples of that). At this point I just can’t imagine a way to effectively differentiate between the two without throwing insane amounts of manpower at it, in which case it sort of seems moot to me.
Throwing more nodes and ways at the map doesn’t necessarily make it better.