Cleanup of name= fields describing buildings

Communities Indonesia
1

Hello!

A user in the OSM US Slack pointed out that this mapper has been adding name=rumah and name=bangunan etc to structures they have added. I have sent them a message about using the name tag more appropriately and cleaned up the local area to their recent edits. However, this overpass query shows that this is somewhat common across the region and will need some more work helping various mappers. A couple other common name= values that I suspect better as regular building tags are “pabrik”, “bandunan”, and “sekolah”.

Beyond that, the presence of these may point to areas that need even more cleanup. Scrolling around a bit (and a lot of mechanical translation) leads me to believe this is one such area: Node: ‪Campakoah‬ (‪1308701335‬) | OpenStreetMap. I did a mechanical translation of all of the name= fields in this area, trying to get a sense of it, you can see it here.

I am happy to do the cleanup edit (JOSM makes this super easy) but it may be easier for someone more local to investigate and see what things need to be taken care of. It’s also probably helpful to have someone send messages to local mappers and talk with them about the general nuances here. After all, it’s a super easy mistake for mappers to make!

All that said, it’s totally possible I have massively misunderstood some local mapping convention… so definitely let me know if that’s the case.

1 Like
2

Hi, I’m kinda tackling a similar issue at my scale in France. Could you share more details about the cleanup process?

Do your name values can be 1:1 conveyed through a building tag ?

2 Likes
3

Sure!

The general idea is to use the JOSM editor in conjunction with Overpass queries to do bulk review and fixup. Unfortunately, this kind of cleanup just isn’t well suited for iD or Rapid. It’ll roughly follow the work pattern outlined here. That’s aboud fixing up address information but the download and edit loop is the same.

  1. Find a tag you think needs cleaning up
  2. Use Overpass to determine the extent of the issue
  3. Coordinate with other mappers to make sure you’ve understood the cleanup correctly
  4. Edit in JOSM as above. Generally across a set of changesets you can individually review… this part is actually pretty fast for simple tagging edits.
  5. [optional but can save you headache later] look through some object histories and message users that have been adding the incorrect tagging to help get them back on track.

Definitely let me know if you have any questions or if there’s anything I can help with.

3 Likes
4

To answer your second question, it’s possible that the name tag may need to be replaced by a set of tags, one being an upgraded building tag. The problematic area I pointed to in my original post has a large number of non-building features that would need likely need tag updates.

In general, taking textual descriptions out our the name tag is good for a lot of reasons but the big one is that it makes the data more consistent and query-able.

5

In short, they are trying to encode important geospatial-related information using plaintext natural language instead of OSM’s tagging conventions.

The problem is that not all “local geospatial patterns” are properly represented with the current OSM tagging standards. Therefore, they don’t really have much choice but to use plaintext natural language. They simply want to indicate that there is an “X” in this region, but there is no accepted tagging convention to encode that information accurately.

I believe that mechanically “mass-fixing” these tags without proper knowledge of local culture and context will result in significant information loss.

The solution to this problem is probably to develop standardization for local feature mapping tags. This way, there will be no information loss, and all this information can be repackaged with a uniform tagging scheme, making it easier to process by both humans and machines.

It’s quite a colossal task, but I think we could start with these steps:

  1. Identify all tagging errors related to this problem.
  2. Identify common mistakes that could be resolved by simple tagging cleanup.
  3. Identify several edge cases that require the creation of a new tagging scheme. This is quite challenging as we must closely consider the current OSM tagging consensus and whether this new tagging scheme will be accepted in the community or not.
  4. To prevent future mistakes, we may need to create a specific iD tagging preset that incorporates the new tagging scheme devised in step #3.

For example

Simple tag cleanup

  • name=rumah : residential house
  • name=bangunan : building
  • pabrik : industrial factory
  • bandunan : might be a typo (bangunan = building), or might be a local language that i dont know. There are several widely used local language in Indonesia, and sometimes people tend to name a things in their local language instead of using the official national language of the country.
  • sekolah : school

Might need to devise a new tagging scheme

Node: ‪Campakoah‬ (‪1308701335‬) | OpenStreetMap and
TranslatedIndonesianNameFields - Google Drive

That spreadsheet is truly fascinating. Machine translation can’t fully grasp its meaning because some local terms don’t have a direct one-to-one translation to English words. We might need several wiki pages to explain the concepts behind these terms.

I will attempt to explain (and define) several of the local terms listed in this spreadsheet later. For each definition, we have to select an OSM tagging scheme that closely matches the intended meaning. If none exist, we may need to propose a new tag.

3 Likes
6

Thank you for your thoughtful reply. I’ll start by stating that I agree that a mechanical edit of all of this would be completely incorrect. The sheet was only for my own use to try and make sense of things and maybe get some insight. You’re totally correct that mechanical translation would produce undesirable results.

Getting better tagging presets available in iD and other editors, as well as wiki documentation (in accessible languages!) for classes of things that local folks clearly want to make notes about is a large but worthy task. The tagging system can always be better refined and more accessible. For example, even here in the US there’s a gap the various landcover and landuse values that don’t really appear in continental Europe.

While I cannot help very much with things requiring local region or language knowledge, I do have a reasonable amount of experience with tooling that makes discovery/tagging/reviewing quick and as close to error free as a mapper 10,000km away can be. I am happy to work though issues identified to get things cleaned up a bit.

Let me know how you think it best to proceed. I’m happy to start working on the “simple tag cleanup” items in a few days, to give folks more time to digest and comment.

1 Like
7

For now, I think it’s safe to perform a mass cleanup for the name=RUMAH/rumah/Rumah tag (as stated in this Overpass query). “Rumah” means house, and it should be tagged as building=house.

Here’s the pseudocode that I came up with:

  • Retrieve all OSM objects that contain the tags name=RUMAH, name=rumah, or name=Rumah.
  • Process each downloaded OSM object:
    • Remove the tag name=RUMAH, name=rumah, or name=Rumah from the OSM object.
    • Check if the building=* key already exists in the OSM object.
      • If it does not exist, add the tag building=house.
      • If it already exists and its value is “house,” do nothing (it’s already correct).
      • If it already exists and its value is yes or residential, replace the value with “house”.
      • If it already exists and its value is anything else, do nothing.

I conducted a quick analysis using the Overpass query, and here are the possible values for the building tag:

image

This process will:

  • Add the building=house tag to 9646 OSM objects that did not have the building=* key before.
  • Change building=yes to building=house for 320 OSM objects.
  • Change building=residential to building=house for 11 OSM objects.
  • Delete the name=rumah tag for the 1941 OSM objects that are already properly tagged as building=house.
  • For the remaining OSM objects that are not houses (e.g., industrial, greenhouse, retail, hotel), they should not be tagged as building=house. Do nothing.
2 Likes
8

I’ll start on the mass cleanup. The rules above look sensible and I’ll let you know if I discover any edge cases etc. I also will let you know if I discover another location like Campakoah. So far, that’s the most densely name= tagged area I’ve seen.

9

I managed to get a bit of cleanup done. Here’s some notes that may or may not be useful as we keep working through things.

Areas with more name fields potentially needing review:

Common name= fields I saw on things working through the houses list:
ways with name “gang” (alley?)
ways with name “kios” (stall?)
ways with name “gudang” (warehouse?)
ways with name “ruko” (shop?)
ways with name “Bangunan Masjid” (mosque?)
ways with name “Kebun” (garden?)
ways with name “Sawah” (ricefield?)

1 Like
10

Changeset : 136885252

This is Malaysian territory. I don’t know much about their local convention there.

Changeset : 136885370

Just like Campakoah, we need manual, case-by-case analysis around here.

Also, after I analyzed its changeset history, it seems that there was a government sponsored humanitarian mapping program around here. One of the OSM user that contributed heavily around this area is “fasilitatorteknispemetaanmitra”, that means “Participatory Mapping Technical Facilitator”. It might be related to this or this. There are some tagging conventions that I dont understand around here, and I don’t want to remove them until I fully understand it. I already tried to contact the community in the Telegram group, asking if anyone know about this program.

Changeset 136886043

  • rumah warga : building=house
  • gudang : building=warehouse, a commercial building for storage of goods
  • kamar mandi : amenity=toilets, but I can’t verify whether this mapper is trying to map a public toilet, or someone else’s private toilet.
  • pos ronda : building=guardhouse , a small building constructed to house guards. It literally means “the ronda outpost”. The ronda is a night watch that patrols the environs of a village or neighborhood. The term “ronda” in Indonesian is a loan word from Portuguese “ronda”, meaning “patrol”. The ronda is typically performed by male heads of households (or their sons) from each neighborhood. Pos ronda is really prominent in Indonesian villages, that serves as its most basic civil defense infrastructure. See (Rafael, 2018) for further description.
  • gang : highway=living_street , “gang” is the lowest hiearchy of Indonesian road system where the local pedestrian get the highest priority and the vehicle speed limit is very low.
  • kios : building=commercial
  • ruko : building=commercial
  • bangunan masjid : Mosque, building=mosque + amenity=place_of_worship + religion=muslim
  • kebun : landuse=farmland + crop=* (specify the crop type for each kebun. i.e. : “kebun singkong” → crop=cassava )
  • sawah : landuse=farmland + crop=rice
  • Also, just like Campakoah, we need manual, case-by-case analysis around here. I probably need more time to process all of this.
4 Likes
11

So, I want to know who is responsible for name=* tag around Campakoah.

First, download all name=* tag by using JOSM :

image
image828×571 109 KB

Then, file → save as → OSM server files (*.osm).

Then, process this osm file by using python script.

from xml.dom import minidom
import datetime

f = minidom.parse('campa.osm')
actors = dict()

nodes = f.getElementsByTagName('node')
for i in nodes:
    tags = i.getElementsByTagName('tag')
    for j in tags:
        if (j.attributes['k'].value == 'name'):
            timestr = i.attributes['timestamp'].value
            timestr = timestr.split("T")[0]
            #print(timestr,i.attributes['user'].value," : ",j.attributes['v'].value)
            act = i.attributes['user'].value
            if (act not in actors):
                actors[act] = dict()
                actors[act]['names'] = []
                actors[act]['dates'] = []
            actors[act]['names'].append(j.attributes['v'].value)
            actors[act]['dates'].append(timestr)


nodes = f.getElementsByTagName('way')
for i in nodes:
    tags = i.getElementsByTagName('tag')
    for j in tags:
        if (j.attributes['k'].value == 'name'):
            timestr = i.attributes['timestamp'].value
            timestr = timestr.split("T")[0]
            #print(timestr,i.attributes['user'].value," : ",j.attributes['v'].value)
            act = i.attributes['user'].value
            if (act not in actors):
                actors[act] = dict()
                actors[act]['names'] = []
                actors[act]['dates'] = []
            actors[act]['names'].append(j.attributes['v'].value)
            actors[act]['dates'].append(timestr)

topactors = dict()
for i in actors:
    topactors[i] = len(actors[i]['names'])

topactors = {k: v for k, v in sorted(topactors.items(),key=lambda item:item[1],reverse=True)}
for i in topactors:
    print(i,topactors[i])
    for j in actors[i]['names']:
        print(j,end=", ")
    print("")
    print("--------------------------")
    print("")

Here is the result :

12

Animal Keeping

landuse=animal_keeping + animal_keeping=goat : Map an area that is used to keep animals, particularly horses and livestock.

Example : KANDANG KAMBING, Kandang Kambing, Kandang Kambing (PAK DARSONO), Kandang Kambing (Pak Suryono), Kadang Kambing (BU BADRIAH), Kandang Kambing (PAK GERI), Kandang Kambing (Pak Wartoni), KANDANG KAMBING (PAK SISWOYO), KANDANG KAMBING (PAK HAMIDUN), KANDANG KAMBING (KOSONG), Kandang Kambing (Kosong) (Pak Sunaidin), Kandang kambing, KAMBING, Kandang kambing (Pak Ahmad Suwanto), Kandang kambing, Kandang kambing, Kandang Kambing (Pak Munarso), Kandang kambing, Kandang Kambing, Kandang Kambing, KANDANG KAMBING (PAK HAMIDUN), KANDANG KAMBING (PAK YUDI), KANDANG KAMBING (PAK SYARIFUDIN), KANDANG KAMBING (PAK SYAHRONI, KANDANG KAMBING (PAK PARYO), KANDANG KAMBING (PAK SARKAM), Kambing pak sukirman, Kambing pak samyudi, Kambing pak jarwoto, Kambing pak wasirun, Kambing pak sudiarso, kambing pak karnanto, KAMBING PAK SURDIYATO, kambing pak rusno, kambing pak karso sasmoyo, kambing pak karso sasmoyo, kambing pak karnanto, kambing pak rasmudin, kambing pak misroni, kambing pak wahyono, kambing pak kusmono, kandang kambing, sapi ( pak yasroji ), ternak kambing ( pak mirsono ), pak warmudin ( kandang kambing ), Pak sudardi ( ternak kambing ), Kandang Kambing ( muswanto ), kandang kambing, kandang kambingpaksarmanto, Kandangkambingpakkusrianto, Paktohirinkambing, pakrusmonokambing, Kambing pak tirto, Kambing, Kambing pak wahyanto, Kambing pak wito, Kambing pak wandi, Kambing pak mistono, Kambing pak saryono, Kambing pak munaryo, Kambing pak ramidi randat, Kambing pak kasmidi, kambing pak sukmanto, kambing pak tarno, kandang kambing pak tarmin, Kambing pak rt, Kambing pak sukiman, Kambing kosong pak miarto, kandang kambing pak ahmad nur kholis, kandang kambing pak surwanto, Kambing pak supardi, kandang kambing pak trio, kandang kambing pak daryanto, kandang kambing pak sukarjo, kandang kambing pak karyanto, kambing pak surmin, kambing pak sukirno, Kandang Kambing, Kandang Kambing, Kandang Kambing, Kandang Kambing, Kandang Kambing, Kandang Kambing, Kandang Kambing, Kandang Kambing Pak solihun, Kandang Kambing ( pak jaelani ), Kandang kambing Pak warmudin, Kandang Kambing ( Pak kusmidi ), Peternakan Lele dan Kambing, Pak Kasmidin (Peternak kambing dan sapi), Kandang Kambing

Aquaculture

aquaculture=* : Used to specify the aquatic organism farmed in an aquaculture feature tagged landuse=aquaculture.

Example : natural=water + landuse=aquaculture + aquaculture=fish

[20,799]

Example : kolam lele kecil (PAK WARSONO), Kolam Pemijahan Ikan Lele, Kolam Pembesaran Ikan Lele, Kolam Pembesaran Ikan Lele, Kolam Pembesaran Ikan Lele (Pak Yudi Warsono), Kolam Pembesaran Ikan Lele, Kolam Pembesaran Ikan Lele, Pembesaran Kolam Ikan Lele, Pembesaran Kolam Ikan Hias (Pak Dasiman), Kolam Pembesaran Ikan Lele (Pak Toto), KOLAM IKAN MUJAIR, BAWAL (PAK HAMIDUN), KOLAM IKAN LELE (PAK HASAN), KOLAM IKAN LELE (PENGELOLA PONDOK), KOLAM IKAN LELE (PAK RISMAN), KOLAM IKAN LELE (PAK RT), KOLAM IKAN LELE (PAK YUSWERI), KOLAM BIBIT LELE (PAK KASRUDIN), KOLAM IKAN GURAME, KOLAM IKAN LELE (PAK SUTIRNO), KOLAM IKAN KONSUMSI MUJAIR, GURAME(PAK YUSUF), KOLAM IKAN KONSUMSI MUJAIR, NILA (PAK SUNARYO), KOLAM (KOSONG), KOLAM IKAN LELE (PAK FITRIANTO), KOLAM IKAN KONSUMSI (PAK SUPANDI), KOLAM KONSUMSI IKAN MUJAIR (PAK DARMANTO), KOLAM KONSUMSI IKAN LELE (PAK SARYO), KOLAM IKAN LELE (PAK NARYO,SRI), KOLAM KONSUMSI MUJAIR(PAK WANDI), IKAN LELE DUMBO, TERNAK IKAN PAK SUHRI, IKAN PAK MARYONO LELE DUMB, IKAN PAK KADUS lele dumbo, IKAN PAK KADUS mujahir, ikan, ikan, ikan, kolam ikan mujahir pak kades, kolam ikan lele dumbo pak sadirun, ikan lele pak saeful rohman, ternak lele ( Pak edi purnomo ), Pak Mustofir (Penjual Ikan termasuk Laut), Ikan Mujair ( Pak wismanto ), Kolam Lele ( Pak kusmidi ), pak anton ( penjual ikan ), pak narmudin ( ternak ikan ), empang lele Psk Mistarto, Pak Saiful Mustakim (Kolam Ikan Lele), Kolam Lele Pak, Kolam Ikan Pak Kusno, Kolam Ikan lele pa mundirin, Kolam Ikan pak jahiman, Kolam Lele Pak Riyatno, Kolam Kosong Pa Solimin, Kolam LELE, BUDIDAYA IKAN LELE, ternak lele ( pak maryono ), Peternak lele Pak hadirin, Kolam lele pak hadirin, Peternakan Lele dan Kambing, Pak Wasrudin (Peternak Lele), Pak Kasmin (Peternak Lele), Kolam lele, 1 kwintal ( kelompok miniaturtarus), campuran dari beberapa Rt, Ternak lele pak karyono

Poultry

farmyard=poultry : a farmyard area used for raising poultry, usually chickens, for meat or eggs.

[987]

Peternak ayam (Hj.Musringah), KANDANG AYAM PETELOR KOSONG (PAK HAMIDUN), KANDANG AYAM PEDAGING (PAK HASAN), KANDANG AYAM (PAK HASAN), KANDANG AYAM BANGKOK TERNAK (PAK NENDRO), KANDANG AYAM BANGKOK TERNAK (PAK SUBEKTI)), KANDANG AYAM BANGKOK TERNAK (PAK SUPANDI), kandang ayam, mie ayam bakso ( pak syarif ), Ayam, ayam potong pak jurit, Peternakan Ayam, jual gorengan, ayam mateng (ibu makhyuri), mie ayam pak maino, Ayam Potong, KANDANG AYAM U MUSRINGAH, Kandang ayam, ayam, Kandang Ayam Makam, Kandang Ayam Makam, Kandang Ayam petelor ( Pak wuhar Rt 1/3 ), Kandang Ayam dan Lele, ENTOG (PAK EDI SUKAMTO), KANDANG ENTOG (PAK WARSONO), KANDANG ENTOG (PAK DWI), KANDANG ENTOG (PAK WINARDIN), KANDANG ANGSA (PAK SUPANDI)

Cowshed

building=cowshed : a building for housing cows.

[21,625]

Example : KANDANG SAPI, kandang kambing, sapi ( pak yasroji ), pak maryanto ( ternak sapi), kandang Sapi pak sukarjo, Kandang Sapi milik 2/6, Kandang sapi milik 2/6, Kandang sapi pak tarsuni, Kandang Sapi pak daryun, kamdang sapi pak basirun, kandang sapi pak sakiman, kandang sapi pak tasruni, Kandang Sapi, Pak Kasmidin (Peternak kambing dan sapi)

Fast Food

amenity=fast_food : The food has a short preparation and serving time, usually because it is industrially prepared food and requires very few additional preparation steps.

Example : PRODUKSI SARAPAN KELILING (PAK SUPRI), PAK SUBANDRI (BAKSO,SOSIS BAKAR, MAINAN ANAK-ANAK), IBU RIHANAH (AYAM BAKAR,GORENG), IBU SUDIRAH (PEJUAL AREM-AREM), Warung Mie Ayam, Penjual Pecel dan Rujak (Ibu Asminatun), Sarapan (PAK YUDI), WARUNG JAJANAN (PAK WANDI), PENJUAL AYAM GORENG KELILING (IBU SULENI), Warung Seblak, IBU SITI AMALIAH (PENJUAL JAJANAN KUE), GUDANG RT 2/4, TEMPAT JUAL JAJANAN PASAR SERABI (IBU KUSPRATINAH), WARUNG JAJANAN (IBU MUNIROH), pak rohiman ( bakso bakar ), nursholeh ( kupat gorengan lontong ), jajanan tradisional pak turino, penjual ketupat bu sanmurni, Rumah Penjual Onde-Onde (Ibu Nur), bubur sumsum, cenil, kluban pak marsudin, Rumah Penjual Makanan Klanting (Pak Tarno), Rumah Penjual Tahu Isi Brontak & Arem-Arem (Ibu Pingah), Penjual Aneka Snack (Bu tuti), jualan rempeyek bu rasiti, jual gorengan, ayam mateng (ibu makhyuri), warung bakso bu ana & warung kopi pak ari setianto, Serabi bu aminah (jajanan pasar), mie ayam pak maino, Warung Makan, Penjual Soto, PENJUAL JAJANAN PASAR, Warteg, Penjual Mie Ayam Ceker, martabak mini ( pk heru), Penjual jajanan tradisional ( ibu surni ), Warung Makan (Pak Yetno), Warung Bakso & Mi Ayam Pak Widiarto, Warung Makan Pak Khoeri & Mba Nur Kholis, Warung Jajanan (PAK PARTO), Bakso, Warung Jajanan, Mie Ayam MIsno, Warung Jajanan, Siomay & Jajanan lainnya, Warung Jajanan (Ibu Adimiati), Warung Jajanan Mendoan Warkop ( Mislamudin ), Warung Makan Berkah (Pak Suparjo RT 2/6), Toko Jenang Wajik Ketan Jati Rasa Pak Arifudin

Food Production

product=food : Used to map entities that produce food.

Example : Pembuat Tape (Pak Darsono), PRODUKSI KERUPUK (IBU BADRIAH), PRODUKSI JAJANAN PASAR (PAK MARGITO), Penggilingan Tepung dan Kopi (Pak Sutarmo), Pemesanan Roti Untuk acara Bu Riswanti, PRODUKSI KOPI BU SUHANTI, PRDUKSI KUE BU NURUL KHOMARIAH, OLAHAN MAKANAN SINGKONG BU ROSIDAH, produksi jajanan pasar, produksi snack, bu kasmini ( jiwel kriyik ), keripik pisang ( mas hamid ), penjualan sampai purbalingga, ternak Lele, produksi teh jawa ( pak sahyuni ), ternak lele, keripik tempe ( pak nasir), penjualan supermarket, penjualan sekitar purbalingga, produksi tempe pak trio, Penggilingan Padi

Convenience Shop

shop=convenience : A small local shop carrying a variety of everyday products.

Example : warung SEMBAKO (PAK HAMIDUN RT1/4) PENGEPUL GULA, PEDAGANG SEMBAKO (PAK DARMO), Warung Kelontong (Pak RT Kusworo), Warung sembako, PENJUAL SEMBAKO (IBU SUYANTI WAHYONO), WARUNG SEMBAKO, Toko Sembako Pak Sarno, warung jajanan kelontong lengkap alat listrik (, toko sembako bu mumfingah, WARUNG Kelontong, Warung sembako dan jajan, Warung Sembako (Pak Toto), warung bu rasimah, warung pa Solimin, Warung Sembako (Samsoni) (Tusuk Sate), Toko Kelontong (Pak Rino Utomo), Warung EVRI, Warung Najwa

Greengrocer

shop=greengrocer : A shop which sells fruits and vegetables.

Example : PEDAGANG KELAPA DAN SAYURAN KELILING BELANJA DARI PASAR (PAK SARYO), JUAL SAYUR BU ERNAWATI, JUAL SAYUR BU JARIYAH, pak sarmanto ( jual sayur keliling ), tokobuah ( pak turyono ), jualan sayur bu sumarti

Sawmill

industrial=sawmill : A place where logs are turned into wood products including lumber and plywood.

Example : PABRIK KAYU LAPIS, Industri kayu ( pak han su )hensu, Pabrik kayu pak fauzi, Tukang Kayu pak roni, PABRIK GERGAJI KAYU, Perajin Kayu ( Pak Badir ), PERAJIN KAYU (PAK WARSONO), PERAJIN KAYU DAN BAHAN BAGUNAN, PERAJIN KAYU (PAK EDI SUKAMTO), PERAJIN KAYU (PAK RIANTO), Perajin Kayu, Servis Elektronik, Pengrajin Kayu (Pak Akhmad Sukhiri)

Sugar Production

product=sugar : Used to map entities that manufacture sugar

Example : RUMAH PENGEPUL GULA JAWA, PABRIK PRODUKSI GULA, pengepul gula jawa pak karyanto, PakSutiman (Penderes), Pak Kartoyo(Penderes), produksi gula

Face Mask Production

product=Face_mask

Example : PENGRAJIN CONNECTOR MASKER (IBU SUNITAH), Rumah Pengrajin connector masker (Bu Listi)

Coffee Shop

shop=coffee : Shop focused on selling different kinds of coffee

Example : WARKOP, warkop bu irma, kopi jawa tulen (produksi & warung) ni medi

Canal

waterway=canal : man-made open flow waterways used to carry useful water for transportation, hydro-power generation or irrigation purposes.

Example : Irigasi, irigasi mati, irigasi mati, Irigasi, Irigasi, Irigasi, Irigasi

Drain

waterway=drain : Artificial waterways, typically lined with concrete, used to carry superflous water like storm water or grey-discharge.

Example : drainase dari jalan mampet jadi masuk rumah, jalan rusak (drainase kecil, jadi kurang berfungsi), DRAINASE RUSAK, DRAINASE RUSAK, DRAINASE RUSAK

Note

note=* : A note is used to inform other mappers about non-obvious information about an element, the author’s intent when creating it, or hits for further improvement. A note is generally used on OSM object to describe real-world things like “the bus stop is actually here but the signpost was knocked down in September 2009”. The length of a note is limited to 255 characters.

[3,776,899]

Example : BELUM ADA DRAINASE SEBELAH UTARA JALAN, JALAN RUSAK DARI RUMAH PAK RT, DIBANGUN RABAT BETON DARI PINGGIR RUMAH PAK YUDI, BELUM ADA DRAINASE DARI RUMAH BARU, DRAINASE KURANG BESAR DAN MAMPET, kurang penerangan 1 atau 2, penambahan talud sekitar 50 meter dari pos ronda menuju kolam, diadakan irigasi disebelah sawah suksara, irigasi mati, irigasi mati, perlu penerangan yg terang, PERBAIKAN JALAN, SAMPAH DISELOKAN, BANJIR KARENA SAMPAH, DRAINASE MINTA DIPERBESAR (1), TALUD SEPANJANG JALAN WISATA, drainase rusak ± 50 m barat jalan karena taludnya rusak, plan gudang RT, dibuat jalan tapi belum dapet ijin dari pemilik tanah, butuh penerangan, butuh penerangan, butuh penerangan, butuh penerangan, butuh penerangan, butuh penerangan (2), drainase kiri belum adaa, kanan rusak, tiang listrik kurang 3, butuh talud sepanjang jalan utama, penerangan rusak, jalan rusak, jalan rusak, penerangan rusak, penerangan rusak, banjir jika drainase tidak ada, jalan rusak, drainase dari jalan mampet jadi masuk rumah, jalan rusak (drainase kecil, jadi kurang berfungsi), kurang penerangan (cuma 3), Kurang Penerangan titik 1, rawan longsor, BUTUH PENERANGAN, BUTUH PENERANGAN, DRAINASE RUSAK, DRAINASE RUSAK, DRAINASE RUSAK

1 Like
13

Mass-cleanup process

First, extract the OSM object ID by using regex :

image

image
image760×352 11.2 KB

image
image1364×681 140 KB

Open JOSM → File → Download object - > Paste the string of OSM object ID that already extracted in the previous step.

image

CTRL+A to select all the objects, then mass edit the tag by using objects panel.

image
image1254×416 129 KB

Results

Since the majority of the name=* describing buildings in Campakoah are agriculture-related places, those tags are suitable for mass-edit cleanup and should be given priority right now.

Goat shelter

Fish pond

Poultry

Cowshed

Meanwhile, the rest of the tags around Campakoah are too few in number to be automated.

1 Like
14

Wonderful work. I’ve been working through things a bit at a time. I will have more time to focus on this after the State of the Map US conference this week.

Thanks again for doing a ton of heavy lifting here!

1 Like
15

Changeset : 136885252

OpenStreetMap

Changeset: 136885252 | OpenStreetMap

OpenStreetMap is a map of the world, created by people like you and free to use under an open license.

This is Malaysian territory. I don’t know much about their local convention there.

Fixed.

1 Like
16

An update:
I’m very close to completely done with the “rumah” tagged items. I will start working on the other issues identified above.

Question:
I came across edits from this mapper (https://www.openstreetmap.org/user/aswatullah%20saputra/history#map=17/-4.02371/119.62513) It appears they have tagged everything they added with aswat### (ex: aswat1, … aswat184)… I suppose is it possible this is some local addressing scheme?? Maybe??

What do you think?

1 Like
17

My money would be on a kid sticking their name on everything on the map that they edit! Only 9 edits, from 4 years ago - I’d just remove the rubbish names.

1 Like
18

Okay, fixed up easily enough. If someone figures out it’s something else they just need to revert Changeset: 136992512 | OpenStreetMap

19

Just hit a patch of “buildings” that are much more accurately landuse=residential. Manual review is a hassle but does turn up things!

1 Like
20

Would someone message this user (RAHMAYANA D131221028 Teknik Lingkungan 22 | OpenStreetMap) about their use of name= tags? I’ve attempted a few things including getting a block from DWG but they persist. Their edits are generally fine otherwise so I presume it’s a language barrier issue.