[Imports] Post-import quality metrics

[Frederik Ramm]

Hi,

On 19.05.2013 22:59, Paul Norman wrote:
> I'm looking to find some metrics to analyze import quality. Some thoughts I
> had were

In my mind, an import is only good if it helps the community to build a 
better map. It is *not* good if it just dumps more data into OSM for a 
cheap show effect ("look how we have all buildings in X!").

This would mean that the quality of an import might be judged by 
comparing the general editing activity in the area with the activity 
that would have taken place without the import. Is it less - the import 
was bad. Equal - the import was useless. More - the import was good.

Of course it is impossible to *know* what the amount of editing activity 
would have been without the import, but statistics can be your friend 
here - identify roughly similar areas (in terms of size of land, 
population density, and size of mapper community) where no import has 
taken place, and compare.

Bye
Frederik

-- 
Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00'09" E008°23'33"