Ah, thanks, Marcos.

Looking at that I think it may make sense for me to make sure that when i review a road in my area and know it to be accurate, I remove the TIGER:reviewed tag if there are no other TIGER tags.

One thing i don’t verify is road geometry. I’ll keep an eye out for road geometry problems. My GPS devices are always set to multi-band, so they’re accurate enough to see if a road geometry is wrong or at least i could do a confirmatory check.

Unfortunately doing a quick glance at Garmin’s maps vs OSM maps it looks like on the web interface garmin is mostly using google map data, so a section where i identified that the map geometry was wrong actually looked fine in OSM when i imported the GPS track.

I think the bike computer itself has OSM data as i’ve noticed that road names disagree sometimes between the garmin computer and web interface and they need to be corrected in OSM.

I also ran into this at first and noticed that it was due to TIGER imports from ~18 years ago.

Apparently, these imports were from government GIS databases that didn’t always just includes roads and paths meant for transportation, but also planned/approved roads that never actually were created (i see this today, for example, with a planned foot traffic bridge that has been routable in garmin for 3 years and isn’t yet completed), old logging roads, or tracks that were miscategorized during automatic categorization on import.

I’ve corrected my share of them. and i’ve also gone through and corrected surfaces as accurately as i could around my area as a lot of my friends use OSM sourced maps to create cycling plans (such as ride with gps or garmin).

Sadly, i think the reality is that a large complex mapping project like this isn’t going to be very accurate without absolutely massive work and that’s harder to get in more rural areas especially.

At any rate, enjoy the journey, find the best routes, and try to label things where cycling is disallowed or private drives correctly for the sake of the landowners or public lands.