ZLima12: see node/111272510 before v9.

Zlima12: This was part of a nationwide effort to move whatever was in deprecated exit_to tags to the modern destination tags. I did not verify that the existing data was correct.

OSMCha shows dashed ramps on the streets/dark layer, but I'm not seeing anything like that in the current OSM database.

I've fixed all the ones I could find.

My apologies; JOSM crashed during my edit, and I thought I fixed all the errors that introduced, but apparently I missed some. I'll dig deeper next time that happens.

Please put a comments on your changesets indicating what they're doing, especially large ones like this; it's hard for others to determine from hundreds of edits what is going on.

These islands look mountable (just paint) as well:
node/95712768
node/271831024
I have local knowledge these are mountable:
node/767536813
node/2958897476
In general, I assumed that any island with trees/shrubs, bare earth, or (shadows of) sign posts visible on sat imagery was non-mountable--and that accounted for the vast majority of nodes marked as mini_roundabouts. Some cases were indeterminate, so I left them as-is even though the odds are high they're wrong too.

I looked at each one (of thousands, so hours of work) via sat imagery. Many cases, including the one you cited, were actually roundabouts, which have a non-mountable island, rather than mini_roundabouts, which have a mountable (often just paint) island. About as many cases were actually turning_circles or turning_loops. Actual mini_roundabouts in the US are exceedingly rare; I found only a couple dozen correct examples.

Unfortunately, the standard layer doesn't (yet) render junction=roundabout on a node, but I've opened an issue in GitHub to get that fixed. Regardless, one should always map the correct logical feature rather than map an incorrect feature that happens to get rendered better.

No se como eso ha pasado, pero yo le he corregido en changeset/43407793. Lo siento.

I can (and did) delete the old relations if i have that list, but it'd be even better to fix my query so i don't miss an existing relation that i can reuse.

I reused all the old multipolygon relations (not many) that came up with the query I'm using:
(way[boundary=administrative]({{bbox}});relation[boundary=administrative]({{bbox}});node[place]({{bbox}}););(._;>;);out meta;
How do I change that to prevent/fix the problem you're talking about?