Mapper performance

From OpenStreetMap Wiki
Jump to: navigation, search

Measuring mapper performance

Some basic statistics about mapper performance are kept in the API, including number of changesets submitted, number of GPS traces uploaded.

Better performance versus poorer performance

Some aspects of mapper performance appear easily measured and compared. One can presume that, all else being equal, Mapper A with 100 edits is more-beneficial to the project than Mapper B with 10 edits. One might be tempted to say that Mapper A is ten times more beneficial, based on number of edits. It becomes more difficult to compare mappers when more statistics are included. If mapper B has contributed more GPS tracks, does that outweigh any shortfall in edits? Are additions more beneficial than modifications and deletions, or not?

The temptation to compare mapper performance based on one or two statistics overlooks the many other possible scales that indicate performance. There seems ample possibility to examine too small a set of criteria and to emphasize a particular type of mapper over other types.

Does is matter?

Does mapper performance matter. In some ways, no. If mappers enjoy contributing to the project and benefit the project, then why worry?

On the other hand, if it can be found that the data and project benefit, from specific mapper actions, or mappers with specific qualities, it would make sense to know what these actions are, and to encourage them.

Measuring mapper performance

Some software and services measure, visualize and / or compare mapper performance. Start a list here.

Measuring mapper performance

  • yosmhm
  • hdyc
  • mapscore (never published)
  • UserStat

Comparing mapper performance

  • mapfight

Combining accounts and performance

  • UserStat