Add several esoteric surface types to the exclude list

[nickspeal]

Class I bike paths don't have the following surfaces:
brush,wood,mulch,decomposed_granite

I found several paths in the Tahoe Area showing up as Class I bikeways because they had these surface designations, not on the exclude list.

I think any arbitrary surface input is possible in OSM, so in the future we might want to move to an opt-in list vs an opt-out list, but this is fine for now.

[brendannee]

Maybe we should update to instead accept only no surface tag or a list of valid surfaces that are bike able? What are the full list of surfaces that should be allowed?

[brendannee]

@nickspeal What do you think of the approach I proposed above?

Let me know if you'd like me to regenerate all bike route overlays with this approach.

[nickspeal]

Thanks for the followup. I managed to set up my local environment to be able to iterate on a few different queries here, and am working to come up with a combo that yields the paths we want.

I have a path forward that looks pretty viable: Dropping the exclude list and adding an allow_list.

Also, for Tahoe at least, Gavin and I think it makes the most sense for Class I paths to only include ways with bicycles=designated (instead of yes). We were dealing with the problem of relying on surface because we've seen bridges with surface=wood both in town on a Class I path and in the forest on hiking trails. Distinguishing based on connectivity to dirt vs ashphalt paths is not viable.

Before going forward with this though, I need to go update a few ways around the Tahoe basin that are incorrectly marked bicycles=yes instead of bicycles=designated.

I'm not sure if we want to do the same things in other non-Tahoe areas, or if we should just update the Tahoe query.

[brendannee]

Glad you were able to build this on your own to test it out. Can you make a PR with an update to the recipe that you are using? I think it makes sense to keep this Tahoe-specific now in the Readme, but I'll try it out sometime for SF and see what the results are.

Let me know if and when you'd like me to pull the latest OSM data and reprocess the overlays.