[Talk-us] Drop the tiger:reviewed tag from roads
OSM Volunteer stevea
steveaOSM at softworkers.com
Fri May 11 20:08:50 UTC 2018
The legacy of TIGER-tagging will persist in OSM for a long, long time. That is the reality of the import we did, rough/sloppy data and all. This legacy serves as many lessons to be learned regarding the practice(s) of wide-scale imports. If it sounds like I'm saying "we made this bed, so now we must sleep in it," you are correct. There are no easy solutions, though there may be better ones.
As TIGER data remain a dominant source of road/highway data in the US (plus MANY improvements), obfuscating their source in the guise of "cleaning up their history" does not help. In fact, a wholesale deletion of tags different than we delete them now (and have) hinders the continuing clean-up/improvement of these data. I elect to continue to clean up noisy/messy/sloppy TIGER data where/when/as I can. When these data reach a state of "good enough that I would enter them into OSM" (as good OSM Contributors, we share such judgements) I remove the tiger_reviewed tag. I support others who do so, too. Largely speaking, this is how we'll "solve" this, although solving with smarter/better solutions is certainly welcome.
Slowly, slowly indeed, we clean up TIGER. It will take years, it may take decades. I 100% support talking about strategies (especially better ones) to do so, I support the chip-chip-chip away at messy data that need improving as we have since TIGER finished uploading. However, wholesale deletion of tags, as doing so contradicts the workflow we have evolved, no, I do not abide that. Should we want to improve that workflow, I'm listening. But (politely, Clifford) I reject that the tiger:reviewed tag has lost all meaning.
We can be more clever, we can be more zealous, but let's not be more blind.
SteveA
California
More information about the Talk-us
mailing list