Rovastar, I’m saying that they do things 1 at a time. So if they add 48 driveways to a stretch of highway, they do it in 48 changesets, which is not optimal. It makes it more difficult to review the changes, it makes it more difficult to analyze history, etc.

It’s of course a problem of finding the right balance, huge changesets aren’t good either, but 1 at a time is not the correct lower bound for the sorts of edits they are doing.

Please think about trying to group tasks into reasonable chunks to ease review for community members. It’s a lot easier to look at a geographically compact changeset with a few dozen changes and understand what it is about than it is to review and understand 3 dozen separate changesets.

Of course I’m making an assumption about how you might proceed based on how driveways are added, my apologies if I’ve made a bad assumption.

Should untrained college kids be banned from artisanal mapping also?

(I get that there is an argument about resources and scale, I just think it’s funny to list characteristics common to lots of OSM participants as scandalous corporate malfeasance)

Is If you have contributed to iD development and managed to refrain from showering anyone who asked for a change with condescension, good for you; sadly it doesn’t change the fact that decision-making in iD has become unstuck from what the community expects of software they imbue with the privilege of being “the official editior”. And the community is not the five people going to lunch with the developer. meant to be ironic?

(what I mean is, I think, despite disagreements, the various people involved are well aware of the nature of the OSM community and it is a bit condescending to dismiss the existence of that understanding and then speak as if you are an affirmative representative of the entire mess that is the OSM “community”)

The worst thing about all this crap is that it is just reaction to change. It’s good for OSM to consolidate tags that mean the same thing, even if in the past it was better for OSM to do something else. We should move on from poorly thought out tag schemes. We should pick winners when multiple tags mean the same thing.

I also think you could have been much more careful in characterizing the situation. iD accepts lots of tagging related pull requests, it isn’t closed off or the work of a few people. And then the situation with third party content is going to be resolved by notifying users and asking for consent. The horror.

Here’s a slightly modified Wizard query that counts amenity=fuel (click through the warning):

http://overpass-turbo.eu/s/DmB

And also one that avoids calling out to Nominatim:

http://overpass-turbo.eu/s/DmC

Note that the preset file is not refreshed automatically: https://josm.openstreetmap.de/ticket/8933

The workaround is to manually delete the cached file.

There’s also a project to collect brands globally, https://github.com/osmlab/name-suggestion-index , additions quite welcome.

The output for JOSM is currently rudimentary, but the suggestions are well used in iD.

Many in the US are tagged as landuse=recreation_ground. I guess it’s mostly my fault.

http://overpass-turbo.eu/s/ACW

Tile rendering servers have been quite busy lately.

Take a look at https://github.com/openstreetmap/iD/commits/master

(go through a few pages)

iD is pretty clearly an active project that welcomes contributions.

Short of a complete and competent implementation of a building tool being rejected, it’s ridiculous (and awful) to blame any one person for iD not having it.

What about all the traffic lights and at grade intersections?

The definition of trunk I try to argue for is the regional road network. The function of these roads is pretty local, mostly providing freeway access.

Why not get rid of the here-comes-a-freeway primary classifications by making them secondary (or whatever makes sense for the local frontage)?

Oh I see, I didn’t read closely enough. I think if you are going to make areas, they should match the type of the road they cover, not just be universally service.

Something may have gone wrong, several streets are mapped as part of a highway=service area:

http://www.openstreetmap.org/relation/7546630#map=19/35.93045/-79.03707

They can also be amenity=doctors.

Another way to get backgrounds images into iD is to use http://mapwarper.net/ .

For small areas even a quick rectification should work well and then there is a tile url available in the export tab for the image that can be pasted into the “custom” background input in iD.

I’ve fixed the dragged node in https://www.openstreetmap.org/changeset/51945038

The error was introduced in https://www.openstreetmap.org/changeset/51891607

If you zoom in close it ends up showing the large changeset bounding boxes. Unfortunately there are a lot of them in this area from POI editors.

I think someone has dragged a node of a building a long distance. I think a place of worship given the color.

Sometimes the history view on osm.org is an easy way to locate the large bounding box of such a change.

I’ve come to look at the import process as an up front quality control step. Of course quality assurance could be done by finding problems with imports and then handling the problems (either by removing and redoing or by fixing the data directly), but that pushes the QA work from the people who believe the new data will improve the map off onto whoever happens to look for and notice the problems.

In this case, one thing the import process probably would have raised is that the county has parcel data with addresses and property use information (the property use is shown in the “Class Code” field at http://ccgisapps.charlestoncounty.org/public_search/ ). For a lot of data users, the address information is more interesting than the building footprints, and some effort to set a more specific value than building=yes is always nice.

It’s likely necessary to use a command like ogr2ogr /share/gis/extracts/test.shp /share/gis/USFSOsm/ -clipsrc -85.2 46.2 -84.5 46.6 to clip out a smaller region of the OSM tagged shapefile, JOSM doesn’t seem to do all that well trying to load the whole file.

As Jack the Ripper says in the diary entry, many roads in US National Forests are built to provide access to a point of interest and will be maintained for use by passenger vehicles. These are sensibly classified as highway=service.