Valid point.

When I did the first pass to specify crossing:markings type (lines, ladder, zebra) I did a query on all the existing crossing=marked in the city. I wanted to leave the existing data alone as much as possible and only add the :markings tag on top- this worked for the most part, until I came across this one node which looked like the markings had been removed. I didn't want to remove the crossing= tag as that would be removing existing data.

Really the main issue here is that it is a data conflict that can only be resolved by a ground survey, which is what my fixme note is for and was added to explain the situation. I just haven't had a chance to make a trip around town to survey fixme's I've added yet, I usually do those in batches.

Hey, just so you know, I intentionally left off crossing:markings from this crossing (node/9241020432) as from aerial imagery it looks like the road got repaved sometime between Bing's image capture and the Virginia state imagery capture. I put a fixme on the node and way to make sure someone knew about this.

I haven't had a chance to survey it yet, but please do not re-add crossing:markings unless you can verify with a ground survey whether or not the crossing got repainted. I know iD likes to yell and makes it easy to add, but there was a reason it was missing from this crossing.

I've also been trying to specify what type of markings there are in crossing:markings, so if you come across one feel free to add that extra detail from imagery instead of just adding "crossing:markings=yes"

This changeset looks pretty good too, the only thing I would advise is that "closed=" is not really a key that's in use. To indicate a shop is closed and a building is vacant, you can change the "shop=" to a "disused:shop=". This allows you to still specify what kind of shop it was, but also makes sure that any of the third parties who use OSM data know it's not currently open, because they may not know about the tag "closed=permanently."

If it's more run down and fully abandoned, you can also use "abandoned:shop" and "abandoned:building", but if it's still in good shape and the building could be re-opened with only some cleanup work then disused is the best tag.
---

Published using OSMCha: https://osmcha.org/changesets/142790174

These edits look good- though it looks like you removed the wikidata tag. Could you add it back? Its value was Q21028360.
---
#REVIEWED_GOOD #OSMCHA
Published using OSMCha: https://osmcha.org/changesets/142692244

Hi there- I saw you requested a review on this edit. highway=pedestrian might be more appropriate here than highway=footway. highway=pedestrian

Hey there! After asking around (because I wasn't sure what to tag it as either) it looks like man_made=cooling (man_made=cooling) is the best tag for a large HVAC unit. I don't think it will render on the map, so just be aware of this. I also don't think iD has a preset for it, so if you come across more, you'll have to manually add the tag at the bottom. However, we always strive for adding accurate tags even if they aren't rendered or preset.

Thanks for your work around E. C. Glass!

Hey there! I addressed your fixme request- the tagging for a pedestrian area is

highway=pedestrian
area=yes

instead of
area:highway=footway

That's why it wasn't rendering. It should appear on the map now!

Hi dufekin,

I still do not see any listed sources for your parcel importing. This makes it very hard to verify if the data you are adding has the appropriate license for OSM.

Please add a source to your changesets and verify that you are allowed to use the data, or you may be blocked again.

As well, I still see multiple issues with the landuses you are adding. Once again I must remind you that legal ownership of land does not always correspond to the actual landuse and that not everything belongs in OSM.

Please revert your changesets and re-upload with the appropriate source attribution.

Hey there! I noticed you've been working on changing road classifications in this area- there's been some work done to standardize how we use the classification system in the US to prioritize network connectivity rather than physical quality of the roads. Generally, the guidelines are to keep major routes like trunk roads classed as trunk throughout their entire length, even if they go from a divided highway down to two lanes.

These guidelines have been written out on the wiki under osm.wiki/United_States/2021_Highway_Classification_Guidance! I recommend taking a look, as the community consensus in Alabama is to follow them rather than what the wiki may say about trunk vs primary based on physical construction.

Downgrading existing trunk routes breaks the connectivity of the region (and also looks fractured when zoomed out since trunk roads render at a higher zoom level than primaries). Just keep this in mind as your changes may be reverted to restore the network!

Other than that, welcome to OSM and I hope you enjoy improving Alabama's map!

Nevermind, I think that standard is more for areas that are routable anywhere in the area. I think this may be correct then?

Hey Nathan! It looks like you're mapping road areas- I believe it will render on the basic osm-carto if you use the combination of

highway=primary
area=yes

rather than the tag highway:area=yes? I've seen it work around in other areas.

Protected public lands can be tagged as boundary=protected_area with the appropriate protect_class=*.

leisure=park implies there are leisure activities to be found there, that the public can visit freely. Applying it to other parcels just to force it to show up on the renderer is bad practice (see osm.wiki/Tagging_for_the_renderer)

Regarding the airport property, I don't know what the best practice would be here. I would lean on either not mapping it or combining it with the surrounding landuses. Again, not everything needs to be rendered on the base map so putting a tag that doesn't accurately describe what's there just so that bit of land fills in a certain color is not good practice.

Hi there, thanks for the response. My concerns are that you are importing parcels without verifying things beforehand.

Are you certain the data found on the Charlottesville GIS portal is public domain? While it may be hosted openly, we still need to verify we are allowed to use and import it.

Secondly, OSM should be as much of an "on the ground" map as it is of legal parcel ownership- blindly adding parcels without regard for what is actually on the ground might be a tempting shortcut, but you end up with things I and others have brought up with tagging combinations that don't make sense, weird geometry, and misleading data.

If I go to a plot of land that is labeled a park, I expect a park to be there- not a patch of forest that is technically owned by the city that they colored as "the same color as other parks" but is really just a legal distinction. And certainly not a wastewater treatment plant (like on other changesets you've done!)

I see you have a description that sounds like a guidepost on the junction node. Could you make a changeset that removes the motorway=junction tag and replaces it with a guidepost? The link I posted in my earlier comment should guide you in how it needs to be tagged.

I'm not sure if trail junctions need to be mapped unless there is a physical marker present. If there is, then you can add a guidepost information=guidepost or a route marker information=route_marker

I'm not aware of any standardized tag for named trail junctions.

At least ask the person who changed it in the first place before editing. I don't mind if you're out there in NHD import land where things haven't been touched in years but changing someone's work without consulting local mappers sucks.

I'll go out there this week and do a resurvey. I do know that at least one of these has a width tagged as four meters, which does not seem like it should be as low as a creek.

I also get your point about "professional geologists" but we aren't here to blindly copy the topo maps.

Also you removed a fixme I had placed on node/9621641803 without actually fixing it. These apartment complexes need unit numbers added. Please don't remove fixmes without actually fixing what was wrong...

Hi RunTrails,

Please do not use the USGS map here for stream/river classification, I've previously gone through and taken care of the classification based on aerial imagery and scouting the area. Stream vs river is subjective but these creeks get wide enough to not be easily traversed over on foot and therefore fall under natural=river.

I find the USGS maps *consistently* under-classify streams/rivers when compared to the OSM definitions.

Hi dufekin,

What is your source for these edits? It's impossible to know just from the imagery you've listed that way/1062911927 belongs to the town. Please list your sources and verify that they are licensed correctly so that OSM has permission to use it.

Hi Natbu2384,

Please take a look at osm.wiki/Tag:highway%253Dliving_street. This type of street is very rare in the United States, and is not for regular residential roads. Usage of this tag should only be for roads explicitly signposted where pedestrians have primary access over cars. It is understood that in the US pedestrians have right-of-way on residential neighborhood roads like these, but this is not enough to classify as living_street.