Thank you for fixing this relation. I was stumped. What was the problem with it that was preventing the rendering? It was confusing that the boundary text label was showing in the Openstreetmap-carto style still. - Joseph Eisenberg
See https://github.com/gravitystorm/openstreetmap-carto/issues/3822#issuecomment-527563520

gibranhalil,
In this changeset you added an aerodrome node with name "Unnamed Bime Valley1" - please put this information in "note=" rather than as the name; the name should be the local, Indonesian name for the airport.

Also, this aerodrome was already mapped as an area: way/514947915 (See way/514947915/history)

Be careful not to duplicate data when you add new airport nodes.

- Joseph Eisenberg

gibranhalil,
In this changeset, you added the name Benban and the ICAO code WXU1 to this aerodrome way/52009337, way/520093372/history

However, a week or two earlier you added the same name and code to this nearby airstrip:
way/603539534

They can't both be the same name and code. What was the source for this information?

Thank you for helping to improve the map in Papua.

-Joseph Eisenberg

Gibranhalil,
In this changeset what is the source for the name, elevation and ICAO code of Benban aerodrome? (way/603539534)
Please mention the source of data like names and codes for airports in the changeset comment. Thank you,
Joseph Eisenberg

Hi gibranhalil,
icao= WA0P for way/609054400, Okhika airport.
Could you please include the source for codes like this?

Hi gibranhalil,
I see that you added icao= WXW7 for way/51157017, Okpahik airport.
What is the source of this code?
I thought that all ICAO codes for Indonesia start with WA, WI, WR or WQ?
Perhaps this is a local Indonesian aerodrome code?
See: https://en.wikipedia.org/wiki/List_of_airports_by_ICAO_code:_W#WA_WI_WQ_WR_-_Indonesia
way/51157017

Hi AndrewBuck,
I noticed that you mapped a large number of features as barrier=embankment in Libya and Egypt a few years back.

I wanted to let you know that Openstreetmap-carto (the map style used for the "Standard" layer on openstreetmap.org) is planning to stop rendering barrier=embankment. We do render man_made=embankment, which is intended for one-sided slopes and is much more common.

See: https://github.com/gravitystorm/openstreetmap-carto/issues/3744

Other tags to consider:

200 times barrier=earthworks (https://taginfo.openstreetmap.org/tags/barrier=earthworks)
or perhaps barrier=earth_bank - 184 times (https://taginfo.openstreetmap.org/tags/barrier=earth_bank)

Some mappers use embankment=yes alone for two-sided embankments.

Happy mapping,
Joseph Eisenberg

No, it's ok to add the names if they are right, I don't mean to discourage you.

But please also add a node with place=village or place=hamlet and the name of the desa or dusun, if you know it. I want to add more village names, so that's why I was wondering if you had a good source for finding them.

Google maps and Apple Maps don't have any reliable village names here in Papua yet, for example.

Re: "we consider it a fairly temporary tag that can be removed as people further enhance the data over time"

Good to know. But I've gotten different advice from some of the HOT guys from Jakarta, Indonesia. They asked me not to remove the tags, even if I had edited the way and reviewed it, for at least the next few months, until their project is complete in November. See: changeset/72632172 - last comment from Iko IRM-ED at the bottom.

In this changeset the node/1308641758 (place=town Kenyam) was deleted, I think this was an accident.

- Joseph Eisenberg

I checked way way/703367059 and way/703367058 which were added in this changeset based on facebook's RapidiD program. Both of these are small streams, not roads. I've fixed them.

Joseph Eisenberg

Hi fatur,

I notice you added a bunch of names for roads, like Jalan Jabanda - Yuruf. I assume those are the names of the two villages connected by this road?

Where are you getting the names of these roads and villages? Is it from a government database, or local knowledge, or another map? I'd recommend adding the villages as place=village (a node), if you are sure about the names. But if you don't know for sure, it's better not to add the names to the roads either. Hopefully you have a good source that I can use too, because I would like to add more names.

Sorry, I think I meant to ask about this one: way/70095657/history (I've edited it, but it was new in this changeset).
Would it be possible to add a tag to changesets that are created with RapidiD? This changeset says "created_by=JOSM/1.5 (15155 en)" not RapidiD, and the changeset comment is "adjusted and modified existing ways and tags to better fit digital globe imagery" which isn't really correct, since a several new roads were added, and RapidiD isn't mentioned. But I thought I should ask, because the HOT/HDM guys here in Indonesia said that they were using "import=yes" on ways that were found by the RapidiD program.

Thanks for clarifying that this was the case. It looks like the geometry recognition is quite good! Too bad there isn't any automated way to pick the correct highway class - this is quite hard for humans too.

Perhaps there could be a special changeset tag for things found by RapidiD? Putting the import=yes tag on the individual OSM ways isn't a great idea, since anyone can change these (like me - I was deleting them because they seemed wrong, since the changesets all said "from DigitalGlobe imagery", which isn't an import in the normal sense. I'm leaving them alone now that I know what they are). A special changeset tag, or at least something in the changeset comment, would make it clearer.

Hi Aron Martin,

Thanks for noticing that way/584068527 has been upgraded from a footpath to a real road! However, I have 2 suggestions:

1) This is not a track, because it actually connects villages to the main (primary) road

2) Please don't reuse the route of a footpath, when the road follows a slightly different route. In the DigitalGlobe (or now Maxar) imagery, you can see that the road is many meters away from the old footpath.

Thank you,
- Joseph Eisenberg

It looks like you accidentally deleted node/1308646188 - Ibele (place=village) in this changeset? I've fixed it. Please be careful with that delete button.

JOSM warnings are not "errors", so they don't have to be fixed if there isn't clear enough information. But yes, you are probably right that it is a culvert.

Re: way/597236097 (I wrote the wrong number above) - this is the main road from Kota Wamena to 3 kecematan, and leading to Kabupaten Yahukimo. There the bridge washed out to the south (way/584006123), but the road is supposed to continue into a 4th kecematan across the river - you can see the road south of the sungai (way/391472048). Now people walk across the river on temporary bridges, and then walk the rest of the way to reach 8 or 9 other kecematan in Yahukimo Regency. So it's the main route south. Many people will walk 2 or 3 or even 4 days to get to the end of the road, where they can take a minibus to Wamena. Bukan sama seperti Jakarta, ka? :-) They are extending the road with a new bridge across to the other side of the river, which I've just roughly mapped (no aerial imagery yet, just survey): way/713090211

I believe this meets the standard that you wrote about before: a highway=secondary is a main route that connects kecematan to other kecematan, crossing the borders, no? Actually, since it's the main route into Kabupaten Yahukimo from Kabupaten Jayawijaya, it could be highway=primary, but since it ends near the border I didn't do this - I think highway=primary should usually reach another town, not just villages.

Juga saya mau minta maaf karena saya berbicara dengan cara terlalu langsung dan kurang sopan-santun. Saya harus coba menyesuaikan diri dengan budaya orang Indonesia (pasti orang Papua agak lebih kasar / langsung) - and I'm not used to having anyone help out in this area besides a few hobby mappers adding rivers.

VLD081, thank you for adding way/71801342 (way/71801342/history) - it looks nicely mapped, though not complete yet. Did you use the facebook RapidID program for this one? Just checking.
Joseph Eisenberg
Wamena, Indonesia

Hi Charles Frank,

Thank you for adding some more roads in central Papua.

BTW, did you use facebook's RapidID program for these, or trace them yourself? Wondering about the import=yes tags. Just curious.

Also, I've changed way/701792573 (way/701792573/history) from service to residential. The last building might be a service building, but there are two houses on this road earlier, so I believe it's a public residential road. The last few meters might be a service road, if the last building is related to the water feature (small hydroelectric?), but there is no barrier, and it's quite wide (not just a private driveway), so in this case I believe highway=residential is probably correct.

Oh, also: in this area Maxar Standard imagery is now the newest source, so you might need to look at that to see what I'm talking about. I think DigitalGlobe may no longer be available here?

Hello again!

About way/701838444 and 701838445 - first off, thank you for finding and adding these! They look new.

By the way, did you use facebooks RapidID system for this? Is that what import=yes means?

But I'd recommend using highway=residential for both of these. You can see that the first (way/701838445) connects to 2 different extended family compounds: each one is surrounded by fences with 4 or 5 small houses (round huts with brown, thatch roofs), so there are probably 30 people living on this road, in 2 small neighborhoods. And the footpaths after this road reach a couple more extended-family compounds. It should be highway=residential, not highway=service, in my understanding.

The other road (way/701838444) connects to a neighborhood or village with at over 20 houses and a big church, and there are about 5 different fenced areas of houses for extended families, so this is a hamlet or maybe even a place=village by local standards. It should be highway=unclassified (or highway=residential is ok, since it is so short)

Thank you for helping! (I've made the changes)
Joseph Eisenberg
Wamena, Indonesia