As far as I can tell, all validator issues have been fixed since this changeset was made, and river relations have been created using the new ways. Following our discussion I've also updated the editor to add a link to osm.wiki/LINZ in all changesets. let me know if there are other issues with the import process

Hi, I’ve re-read the import guidelines and I believe it has been followed properly. To address your 3 points:

* This import is not 'disguised', it never claimed to be just for NZ. Data from LINZ has been imported for the Ross Dependency, and parts of the Pacific since 2012, when it was initially discussed with the community (I was not involved at the time). You can see all the regions on the wiki page (osm.wiki/LINZ)

* documentation is available at osm.wiki/LINZ, and every changeset has the attribution= tag which links to the Contributors page, which links to the LINZ wiki page. However, adding source=xxx to every single imported feature is frowned upon nowadays, so we shouldn’t do this. As you suggested, I will add another changeset tag to directly link to osm.wiki/LINZ.

* data is not being 'dumped in complete disregard'. The import is done using a modified version of RapiD, which requires you to manually check and add every feature. For melt streams, this involved merging the starts/ends of streams with lakes and the coastline, and skipping streams that duplicated existing data. The validator warnings from iD in this specific changeset were noted and cleaned up shortly afterwards. Please let me know if you see any examples of badly conflated data.

relation/13266223 will be the relation to replace the one deleted in changeset/89947873

If there are two official names, both of which are commonly used, then the name= tag in OSM should contain the name in the local language(s). In New Zealand, places with two names are most commonly written with a / between them, or with one of the names in parentheses.

If you consume OSM data and only want English names, then you can use name:en=* and fallback to name=* if there is no name:en=* tag.

also reverted, see changeset/111217494
---
#REVIEWED_BAD #OSMCHA
Published using OSMCha: https://osmcha.org/changesets/111217442

This is incorrect and has been reverted. Both islands have dual names (see https://gazetteer.linz.govt.nz/place/54454 and https://gazetteer.linz.govt.nz/place/54452)
---
#REVIEWED_BAD #OSMCHA
Published using OSMCha: https://osmcha.org/changesets/111217494

I've reverted this too
---
#REVIEWED_BAD #OSMCHA
Published using OSMCha: https://osmcha.org/changesets/81656615

reverted along with changeset/110883921
---
#REVIEWED_BAD #OSMCHA
Published using OSMCha: https://osmcha.org/changesets/110883978

It is perfectly acceptable to map driveways. I've reverted this. If you don't like the look of it, use a map style that doesn't render driveways
---
#REVIEWED_BAD #OSMCHA
Published using OSMCha: https://osmcha.org/changesets/110883921

Hi, if you've read access=designated you may have noticed the perfect examples at the bottom of this page. The wiki recommend bicycle=designated and foot=designated for "A combined footway and cycleway". In New Zealand we call this a 'Shared Path', see https://www.nzta.govt.nz/walking-cycling-and-public-transport/cycling/cycling-standards-and-guidance/cycling-network-guidance/designing-a-cycle-facility/between-intersections/shared-paths/. A 'Shared Path' is a combined footway and cycleway explicitly designed for walking, cycling. Both parties have equal right of way.

You asked why =designated is useful over =yes. This is because =designated explicitly states whether the path is designed for that mode of transport. Consider the Kāpiti Expressway: You can cycle on it so it could have bicycle=yes. But it is definitely not designed for cycling; it's a 100km/h expressway. This is very different to the nearby Shared Path which has bicycle=designated because it is explicitly designed for bikes. Routing algorithms will therefore prefer =designated over =yes.

If a large number of data consumers find a tag useful, we should not remove it without discussing with those consumers (e.g. the tagging mailing list).

I have a question for you – why are you removing this tag in the first place?

once again, why are you removing the tags bicycle=designated and foot=designated? These tags provide valuable infomation, as discussed in changeset/107465048

This is wrong, I've reverted this and your other recent edits. Street signs were updated around July 2021. The official name has also be changed to Tūwharetoa Street (ū instead of u).

Hi, in New Zealand addresses in OSM are regularly updated based on the data from LINZ. The tags addr:housenumber, addr:street, and addr:suburb / addr:hamlet are automatically populated by the data that LINZ has for that address. For more info, see osm.wiki/Import/New_Zealand_Street_Addresses_(2021)

The value of addr:suburb / addr:hamlet comes from the LINZ dataset (https://data.linz.govt.nz/layer/53353) which uses Fire&Emergency NZ's localities database (https://data.linz.govt.nz/layer/104830).

I agree that it is strange that Fire&Emergency chose "Tauranga" as the name of the suburb for the CBD. In most other cities, the suburb name for the downtown area is "Hamilton Central", "Auckland Central", "Dunedin Central" etc.

If this is important to you we can update the code to use "Tauranga Central" as the suburb name for this area

Hi, this changeset combined the address points for individual flats into a single node for the property. Example: instead of 12A, 12B, 12C, ..., 12X, 12Y, 12Z Example Street (26 nodes), these were combined into a single node for "12 Example Street".

This was only done for places where the address nodes for the flats were either:
(a) all in exactly the same location; or
(b) haphazardly scattered around the apartment complex; or
(c) where the location of the individual units was guessed by the importer and therefore is inaccurate.

Note that this will never happen for cases where local mappers have correctly spread out the nodes (e.g. osm.org/#map=19/-36.97423/174.92146). If a local decides to map the individual flats in the future, the tool will recognise and respect this.

For more info on the LINZ Address Import see here: osm.wiki/Import/New_Zealand_Street_Addresses_(2021)

updated the remaining ones in changeset/109838072

yea, I'm not sure if motorway is the right tag. There was a note open here for a long time but no one commented. highway=trunk + motorroad=yes seems logical, to match Takitimu Drive which is a similar situation

also the surface is not concrete, it's a deteriorating gravel/mud path

Hi, ways 840886759, 722836095, 722836096, 246588892 are definitely not lit - it's one of the dodgiest places in taka

see discussion on changeset/108517669

This appears to be vandalism:

- Meaningless changeset description
- Suspicious username
- Fake edit (definitely not motorway, not separated carriageway, not restricted access, cycling/horse riding/tractors allowed)
- User has a history of vandalism

I've reverted this. Feel free to discuss if this was a legitimate edit