Survey on the OSM Wiki

Hi all,

I am a researcher at University College Dublin and I am conducting research on how the OpenStreetMap vocabulary of tags and keys evolved.
For this reason, I am analysing the OSM Wiki website (
It would be very helpful if you could answer some of these questions, based on your experience with OSM:

  • Have you defined/updated definitions of tags and keys in the Wiki?
  • Is the editing of the wiki mostly an individual, online activity or is there collaborative editing of groups of users?
  • What were the first keys/tags to be defined?
  • Were there specific milestones/events/meetings/parties that marked the evolution of the keys/tags?
  • What were the “difficult” keys and tags that required a lot of discussion? And the “easy” ones?
  • Do you have any examples of conflicts/disagreements/edit wars in the Wiki?
  • Were there mistakes in the tag/key definition process?
  • How do you solve the conflicts in the vocabulary?
  • What lessons did you learn from the development of the Wiki?

If you want to contact me privately, please drop an email at andrea [dot] ballatore [at] gmail [dot] com.

Thanks in advance for any information!

My home page:

Yes, I have created several proposals and contributed to a lot of other wiki pages.

I guess that the vast majority of wiki edits are performed by individuals, not in a group. I have on very rare occasions edited the wiki with other contributors in the same room, but for me that’s clearly the exception.

That was before my time. But my impression is that tagging was even more informal back then and there was very little effort to document a tag before using it. I believe that this experience still shapes the views of some people in the community - especially (but of course not exclusively) on the English-language mailing lists, where you will likely find considerable opposition to the idea that the wiki is important, and especially to the proposal votes.

In my opinion, bringing in other national communities on the topic might help to balance out the responses.

There has mostly been a gradual development over the years, with no big leaps. Occasionally, some larger challenge which has remained unsolved for a while is tackled more systematically by a dedicated group of people. Examples are the Karlsruhe Schema, a solution for house number mapping primarily developed by the local user group in Karlsruhe which is now the standard tagging for addresses, and the [2nd 3D Workshop Garching](2nd 3D Workshop Garching), where a group of application developers and power users agreed on a standard for 3D building mapping. It should be noted that such groups are in no way official and the results are used simply because they work well, they are not binding by any means.

It took years of time and and several proposals to come up with tagging solutions for lanes and Conditional restrictions. It’s still not clear whether the solutions will catch on.

But while it’s easy to understand that these topics and others like them are inherently complex, we also have problems with topics that seem to be really straightforward such as “How do we map cycleways?” There are still different styles around and I’m not sure whether it will ever be resolved to everyone’s satisfaction. Look at this page for an attempt to create an overview. I fear the fact that it’s so common that almost everyone has an opinion on it makes standardisation very hard - because it would involve telling large parts of the community that their traditional approach, along with all the mapping they have done over the years, is now “wrong”.

For “easy” tags, I’ll use one of my own proposals as an example: step_count. Surprisingly uncontroversial.

I’m using one of my own proposals as an example again: This was not well received, and there were even fake accounts abused to cast opposing votes - something I’ve never seen anywhere else in the OSM wiki.

Another source of conflicts are deprecations of existing tagging, e.g. look at this (relatively tame) squabble:

Or this more civil example, following a discussion of the topic on the tagging mailing list:

Too many to count. :wink:

We don’t really have an effective means to do that. Nobody has the authority to declare some style of tagging deprecated and mass retagging is only tolerated within very strict limits, so in some cases all we can do is wait and hope that all of the conflicting tagging styles except one fall out of favour.

Coming to terms with the OSM Wiki definitely teaches patience and an ability to live with “organically grown” ontologies. But I believe the most important lesson is an understanding of the subtle nature of influence in OSM and communities like it.