Please revert this changeset. It is vandalism

These buildings are incorrect. Please do not add buildings to the map if they do not exit.

Nurmianti, kelihatanya ada beberapa gedung di sini yang tidak cocok dengan bentuknya benar-benar.

diahputri, terima kasih atas kerja keras Anda untuk menabah bangunan di kampung Hom-Hom, Wamena.

Untuk bikin gedung lebih rapi-rapi, bisa pakai "q" unuk bikin gedung persegi, kalau anda pakai editor iD (openstreetmap.org/edit).

Juga ada opsi lebih mudah untuk membuat gedung dengan cepat: pakai program JOSM dengan plug-in Buildings: https://josm.openstreetmap.de dan tambah https://josm.openstreetmap.de/wiki/Help/Preferences/Plugins con ​"buildings_tools" untuk dibujar gedung dengan persegi dengan 2 klik aja.

Terima kasih,
Joseph Eisenberg

way/687074200 - I'm not sure if this is a wetland, meadow or intermittent lake. See ESRI vs Bing vs Mapbox imagery. It would need survey

way/687074198 - this line is a firebreak. I'm not sure how to tag that. And I can't tell if it's grassland or bare ground from aerial imagery; would need on-ground survey.

gibranhalil, what was the source for you IATA code at this airport? KMU is wrong, it's Kismayo Airport in Somalia according to wikipedia https://en.wikipedia.org/wiki/Kismayo_Airport

This is in regards to relation/8140878: relation/8140878/history

I did not intend to delete this feature. I actually attempted to extend it further east, but when I tried to upload JOSM reported that it was empty. I'm not sure what went wrong, but I had a poor internet connection at the time, so it is possible that data was lost when I attempted to make the changes. I will see if I am able to fix it, by downloading the last version and merging with the current data.
---
#REVIEWED_BAD #OSMCHA
Published using OSMCha: https://osmcha.mapbox.com/changesets/76134372

I'd recommend mapping this as two separate features. The cliff should just be a line, and I don't think it should have the name Masade - that's the name of the mesa/plateau, not the name of the cliff, right?

The feature with the name Masada can be mapped as a multipolygon or closed way, and also tagged with historic=* site, for the areas that area included in the historic site. Adding tourism=attraction is fine, though not really needed.

Also see osm.wiki/Proposed_features/Tag:natural=mesa

> "Without this track it is not obvious how to get to the residential zone - so it is also important topologically."

OK, while I didn't want to make assumptions, I will try to guess how you are currently using highway=track from your previous comments, since I have not yet seen a clear statement of the definition for highway=track in Lithuania.

From the comment above, it appears that you treat highway=track as a higher class than highway=service, but below highway=unclassified (and =residential maybe?) and you would like highway=track to render sooner than highway=service on your maps.

It doesn't appear that there are any specific physical characteristics that you can use to define a track: it can be paved or not, it's usually narrow but "forestry tracks are much larger/wider as they need huge forestry vehicles to pass", and it can be short or long.

Am I missing something else?

But this could be highway=service, as used in way/319125591 and many other examples in Lithuania. It's short, as are most of the other examples of roads currently tagged highway=track which lead to residences.

It's still not clear to me how highway=track is defined in Lithuania.

You mentioned that it should not be wide enough for two vehicles to pass one another, but it can be paved. However, highway=service (and highway=residential) can also be narrow, and can be paved or not.

So how does one decide whether to use highway=track in the case of way/230684594, but not in the case of way/319125591 way/319125591 and other similar situations?

The definition at the wiki page is clearer, because it focuses on whether the road serves as access to agricultural/forestry areas - this is highway=track. In contrast, a road which provides access to developed properties like a house, hotel, or power substation is a highway=service.

Oh, that explains it! I only changed the leisure=park tags, I didn't really check the others in this edit.

How do you and other local mappers currently decide if a road to a small hamlet or a couple of houses should be a highway=residential or a highway=service or a highway=track? If it is paved with stones, bricks, concrete or asphalt would you still use highway=track for the example above?

> "If such a small residential area would cause a requirement of residential or even unclassified road it would mean changing 99% of track roads..."

I've checked several areas along the border with Latvia (where I already had data downloaded), and it looks like a majority of the highway=track there are used for access to areas of woodland, forest, or farmland

Here are some examples of highway=track which are access to agricultural or forestry areas:

1) way/86399532
2) way/313101139
3) osm.org/#map=17/56.38954/24.75204
4) osm.org/#map=15/56.3898/24.8920
5) osm.org/#map=14/56.2454/24.9899

All the above locations show roads which appear to be used only to access forestry or agricultural areas, which fits the definition of the tag highway=track since early in the history of OSM:

(Though I did see many tracks which lead to residences, and I would usually map those as highway=service (or highway=residential), they did not appear to be a majority)

(I apologize in advance for writing in English)

About way/230684594 (way/230684594) - this appears to connect a small residential area to a highway=unclassified road. I would suggest using highway=residential - "used for roads accessing or around residential areas but which are not normally used as through routes (which would usually be classified highways or unclassified highways)." highway=residential

In contrast, highway=track "represents roads for mostly agricultural use, forest tracks etc" - roads which are created for tractors, logging trucks and similar vehicles, and might also be used for recreation, hunting and such, but not to access residential areas.

In contrast, highway=service is "Generally for access to a building..." etc - not for access to farmland, orchards, or forestry areas.

Based on these wiki definitions, as well as how the tags are used in many parts of the world* I believe way/230684594 should be highway=residential.

*(my personal experience is mainly limited to the USA, Canada, Indonesia, Singapore, PNG and Australia, but I've also look at a number of other places in Europe and the British Isles, including Latvia and Estonia, though not Lithuania I'm afraid)

What happened was that a previous changeset/73039521 (changeset/73039521) added natural=coastline to the ways across the mouths of the tidal channels. The user, Farras, was trying to fix some relations like the island of New Guinea multipolygon and perhaps the local government boundary relations. (I may have accidentally broken them). But Farras also added natural=coastline to ways like this: way/623823365/history which duplicated the coastline. I appreciate that dmgroom_ct fixed the duplication, but it would have been better to remove the coastline from 623823365 and similar, rather than from the tidal channels, which should be mapped as outside of the coastline (waterway=tidal_channel)

I added to the confusion by not removing the natural=water tags from many of the tidal channels when I added the coastline to them. I had been planning to propose a way to tag the water area of estuaries and tidal channels, but I can see how it can be confusing.

It appears the problem with the coastline was created with changeset/73039521 by user Farras -changeset/73039521

But it's important to be careful before editing so many ways to remove the coastline all at once. There isn't a standard tag for estuaries, so I've been mapping them as riverbanks and coastline - this practice is pretty common in other parts of the world. In this case probably some of the water areas should actually be changed to water=tidal_channel + salt=yes + tidal=yes since some are not actually rivers (though sometimes this is debatable and would require local survey to check).

Farras,
With this changeset you added natural=coastline tags to the administrative boundary ways which cut across the mouths of rivers and tidal channels along this coast. This broke the coastline in several places. If you are using JOSM, please check the validator first to make sure there are no problems with the coastline before submitting a changeset.

In this case, the coastline around Timika extends far north and east because of the mangroves, which are salt-water, marine environments. Please see waterway=tidal_channel for how to map the coastline around mangroves. Thanks!
- Joseph Eisenberg, in Wamena

Sebenarnya jalan ini ditutup untuk perbaikan jalan? Saya dengar masih bisa lewat. - Joseph Eisenberg

Please don't remove the coastline from these tidal channels and estuaries in the mangroves. The coastline should not be below the mangroves, because they are salt-water, marine ecosystems. See waterway=tidal_channel
For example, way/623885638 (way/623885638) is the edge of the mangroves along a tidal channel, which does not connect to any rivers. It's entirely limited to the salt-water, intertidal mangrove environment. It's incorrect to use natural=water in this case, since that tag is only for inland water bodies, not marine water.

dmgroom_ct,
Please don't remove the coastline from these tidal channels and estuaries in the mangroves. The coastline should not be below the mangroves, because they are salt-water, marine ecosystems. See waterway=tidal_channel