[Talk-us] Coconino National Forest boundary isn't rendering anymore?

[stevea]

 Paul White <pjwhite1010 at gmail.com> wrote:
> Does anybody know why the Coconino National Forest doesn't render on osm.org anymore? I don't see any recent changes that would've messed anything up but it's gone. I also noticed that the Klamath National Forest is gone, as well.

I'm glad to see august and more-technical members of OSM (Paul Norman, Joseph Eisenberg...) chiming into this thread.

I am the most recent author of this relation.  I made minor changes to the tags on the relations, not the members or their roles.  Specifically, the edit History (click View History link at bottom of object "pane") displays the previous set of tags (and seems to have rendered to the o.p.'s liking), which included:

boundary=national_park + boundary:type=protected_area

while the present tags exclude those, but include:

boundary=protected_area + protect_class=6

I did this because boundary=national_park is not a valid tag on a USFS National Forest per our evolving wiki https://wiki.osm.org/wiki/United_States/Public_lands , which prescriptively suggests this tagging.

I believe it is safe to assume that the previous tagging of boundary=national_park was incorrectly applied because it rendered, and that the somewhat clumsy and collides-with tag boundary:type=protected_area was added to be more consistent with the newer tagging scheme of protected_area, though it excluded the associates-with tag of protect_class=6 which my newer tagging added, along with the "proper" key of boundary, not boundary:type.  If you followed all that, thank you.

The particular combination of boundary=protected_area + protect_class=6 does render (as a thin green line and an occasional name=* value along edges).  And again, boundary=national_park renders, though differently than boundary=protected_area + protect_class=6 — and rightly so, as these ARE different entries:  a national park is not a national forest and vice versa.

> If anyone knows how to fix this, let me know.

I believe there isn't anything to "fix" here:  what appears to have happened is that a wrong-tagging which rendered with a certain appearance was corrected to be "more properly" tagged, and this renders, but differently.  As these are issues which may continue to be evolving (relatively newer tagging schemes like protected_area compared to national_park, as well as rendering support, or lack thereof, for various values of protect_class), it is possible I lack full clarity into either the present exception of or intended effects of these tags and the Carto renderer.  Here, I only offer my best explanation of present tagging and rendering effects, not future ones.

SteveA