In the history of OSM around two million mappers have contributed to the data. OSM has always been proud of its potential for diversity. Mappers are generally free to map what they want and how they want, although of course the community genrally strives for alignment with the OSM Wiki .

One of the pillars of its success was the hyper-localness and hyper-specialisation OSM reached in some areas. Regional mapping practices, ground-collected information and topic geeks are no rarity.

In our analysis we looked at 'special' (i.e. rare) map edits and if they are contributed by certain 'special' mappers. Special or uncommon behaviour like the usage of uncommon tags may spark suspicion . For example Pascal Neis linked it to vandalism .

But these special cases can lead to interesting insights as in this qualitative study of ours . Without a qualitative analysis e.g. by looking at a certain tag in detail, it is nearly impossible to distinguish erroneous from valid information. In fact it is the rare information that only certain mappers have and add that makes OSM potentially hyper-local and the sparkling and colourful map we know. And love, even though, as data analysts, we sometimes throw our hands up in dispair. For example when we did this analysis on deleted OSM elements .

Methodology

The OSM unicorn factor we present here was calculated using the OSHDB . It describes the tendency of a user to make edits to rare tags or rare map feature types . For example supermarkets are more frequent than hospitals. And because hospitals are rarer, there are less users that have touched them making these users special. They contributed to the diversity and colourfulness of the database and the maps. The same is true for placenames. They often represent highly local and unique information. ... If its not Hauptstraße, ul. Polna or Rue de l'Êglise.

For a detailed description of the methodology, please refer to the Git-Repository linked above containing a thorough manuscript.