[Tagging] [Talk-us] Large fire perimeter tagging?

stevea steveaOSM at softworkers.com
Sun Sep 27 19:59:28 UTC 2020

On Sep 27, 2020, at 12:43 PM, Peter Elderson <pelderson at gmail.com> wrote:
> Clifford Snow <clifford at snowandsnow.us>:
> I'm not sure there would be a consensus agreement to revise the wiki to indicate landuse=forest should be used for timber production.  Thoughts?
> I am sure there would not. landuse=forest just means the area has trees. I think there is some consensus about that. 
> natural=forest: same.
> The idea that natural=wood is for natural woods and landuse=forest is for managed forests has too little practical support.

Peter offers this with no evidence, so I question its validity.  Furthermore, I offer that something like Approach 3 or Approach 1 (from our https://wiki.osm.org/wiki/Forest) has been used by me in North America since 2009 (and I am not the only one; I collaborate with other OSM Contributors on "good use of OSM tagging in wooded areas" in areas I map so we avoid conflict and harmonize on our use of tags).  So, "too little practical support" simply isn't true:  I and my mapping activities (along with others) simply contradict this assertion.  We have been contradicting this assertion for many, many years.

> Since there is no consensus about other aspects than "there are trees", data users and renderers will stick to this.

Here, Peter predicts the future after reaching the false conclusion that "there is no consensus about other aspects."  There is slow, emerging consensus that landuse=forest, natural=wood, landcover=trees, key managed=* with values yes or no... most certainly need much work, but "what exactly OSM should or will do" is a long way from having consensus established.  We have absolutely no predictability about what data users and renderers will "stick to," let's not kid ourselves.

I repeat myself, but what can be said is that trees, woods, forests and how we tag them need a lot of work if OSM is going to comprehensively capture the very wide semantics about these in the real, global world with a finite set of tags to capture these semantics.  We'd do well to improve these, but I'll agree with anybody who says "this is difficult work."


More information about the Tagging mailing list