Hi, On 22.11.2017 04:16, Yuri Astrakhan wrote: > Pierre, I suspect the number of QA-tool-driven changes are as big, if > not much bigger than changes from the organized events and paid editing. > I agree QA tools should be regulated, but are you sure we want to do it > in the same document, and significantly increase the scope?
This is something that was discussed at length while drafting the policy, and you are certainly right, it *is* a difficult area. The spirit of the policy can largely expressed in "responsibility" terms; the policy, by and large, applies whenever the person being responsible for an edit is not the person making it. Most QA tools still require the user to take responsibility. If the QA tool says "here's a road that crosses a river without a bridge or ford or anything, please check on aerial imagery and apply correct tagging" then the responsibility clearly lies with the user. Even if the QA tool says "this road is tagged highway=residentail, should it perhaps be highway=residential instead?" the responsibility still lies with the user. You could go so far as to say: A QA tool that doesn't require the user to take responsibility is not a QA tool, it is a distributed mechanical edit and as such, covered by its own policy already. (Of course if I now set up "the great bridge fixing event" where I invite people to help me fix all these problems in one weekend, and provide detailed instructions to absolute newcomers on how to fix bridges, then there might be a point where responsibility shifts to me and I am now "directing" these people to use the QA tool to fix things.) Bye Frederik -- Frederik Ramm ## eMail frede...@remote.org ## N49°00'09" E008°23'33" _______________________________________________ talk mailing list talk@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk