On Mon, Dec 31, 2012 at 12:34 PM, Serge Wroclawski <emac...@gmail.com> wrote: > On Mon, Dec 31, 2012 at 12:15 PM, Jeff Meyer <j...@gwhat.org> wrote: >> Why does "imports" mean "not crowdsourced," if the crowd determines which >> imports are source-able? > > Crowdsourced means that we survey the data indivdually. > >> Why doesn't public-domain data that the crowd has funded count as >> crowdsourced? > > Because the folks collecting that data aren't part of our community; > they're often low wage, low skilled workers looking to do as little as > they can. That doesn't mean the data is all bad, but it's not from our > community. > > That doesn't mean all imports are bad; in fact, I think some imports > can be good, but there is a distinction to be made.
Data quality issues aside, there is also a large technology difference between traditional GIS and OSM, and it matters. Our data model is completely different and converting between the two in a way that doesn't create strange artifacts on the OSM side (both in tagging and in geometry) is not easy. Our lack of the concept of layers is also important. Everything is shared by everyone. This means some data that is sometimes useful but also very noisy can't be as easily ignored by people who don't need it. This needs to be considered when assessing how a particular data set might impact the community. I'm working on an address import right now so I'm not entirely anti-import but I am also definitely not about to say that we should import everything, just because it is there. We need to be selective and enforce high quality standards. Toby _______________________________________________ Talk-us mailing list Talk-us@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-us
