Hello all,
I was hoping to open a discussion on current best practices for dealing with old TIGER tags. I know this has been covered here in the past, but I think it’s been a few years and it seems possible that methods have shifted. I’m from Portland (Oregon J) and on many streets, TIGER tags have been completely removed, while on others, people have left only the tiger:county tags. Sometimes, you see tiger:zip_left and tiger:zip_right too, while in other cases there are zip_left and zip_right keys instead. In many cases, the following 6 TIGER tags remain: tiger:cfcc, tiger:county, tiger:name_base, tiger:reviewed, tiger:zip_left, and tiger:zip_right. The approach that is preferred at TriMet (where I work) is that if we are able to check the geometry of the street against fairly recent imagery (improving it if needed) and verify the name of the street, from either our local jurisdictional centerlines or the latest TIGER TMS layer, then we remove all of the TIGER tags. We see that as being adequate to remove the TIGER:reviewed tag (especially when multiple mappers have edited the way since the initial import, which is typical). We think that the other TIGER tags are not needed as they’re mostly comprised of information that isn’t really appropriate for the street ways (zip code and county, which take up less space and are easier to keep up-to-date when maintained as separate boundary polygons) and attributes that can be derived from other fields (e.g., prefix, basename, suffix). While it could be handy to have these address components broken out, it adds bulk and requires updating several fields when a name is changed. Further, the TIGER attributes are mostly really outdated at this point as they come from 2005 data and have rarely been updated by mappers. What do you all think about this? Thanks much, Madeline
_______________________________________________ Talk-us mailing list Talk-us@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk-us
