Has this large-scale and, I presume, mechanical edit been discussed and documented anywhere? If not, it would be against the rules and should be reverted.
No, but in this particular case I wouldn't expect the changes to be controversial in any way, and I endeavoured to check as many of the changes as possible (although this was difficult due to the large number of country relations that were in the dataset). None of the changes impact rendering on the main osm.org map, although I would expect them to significantly improve rendering on localized maps like those of Wikimedia Maps.
• Who would I be discussing this change with? Most of these tags were added by a very small number of users who (possibly incorrectly) imported names from Wikipedia or their own data tables, or were added because of the iD language code bug (see below).
• name:xx keys are based on IETF language tags, at least according to the OSM wiki. All of the language codes involved have relevant codes, so I would expect the keys to use those codes.
• I have chosen to capitalize "Hant" and "Hans" because "Latn" is far more commonly used than "latn", and IETF language tags use the uppercase letter.
• Other variants in use in OSM tags: "zh_hant", "zh-hant", "zh-trad" and "zh-traditional"; "zh_hans", "zh-hans" and "zh-simplified". Their meaning is hopefully unambiguous.
• "zh-min-nan" is not an ISO 639 code and is based on the Wikipedia language code for Southern Min. I changed it to "nan". (It is technically difficult to change a Wikipedia edition's language code, as doing so introduces a large number of technical issues that have still not been resolved for the language edition for which this has been done. The Southern Min Wikipedia predates ISO 639-3 by five years, so there was no standard language code for it at the time of its creation. Nevertheless, iD uses the Wikipedia language codes for multilingual names (the issue has been open for almost four years https://github.com/openstreetmap/iD/issues/2457), introducing tagging errors for several languages.)
• "zh-classical" is not an ISO 639 code and is based on the Wikipedia language code for Classical Chinese (I have no idea why these were imported, to be honest). I changed it to "lzh".
• I changed several "zh_py" tags to "zh_pinyin". The latter is much more commonly used and it is impossible to misinterpret the former as meaning something else.
• I made most country codes uppercase. There are very few objects with these tags, and most of them are unnecessary. I opted to uppercase them where they were used in the dataset to match the IETF language tags.
• "Minor" fixes also include key changes for name:pt-br, name:pt-pt, name_zh_classical, and several other keys. I could not use the JOSM validator because of the large number of country relations, although it may be possible by using the reverter on this changeset.
The automated edit guidelines at https://wiki.openstreetmap.org/wiki/Automated_Edits_code_of_conduct say:
Documentation should be placed on the wiki and the proposal should then be discussed on a suitable mailing lists:
Either talk (a general purpose mailing list)
or imports, when discussing imports or issues with previous imports
or if your edit affects only one country or territory then the national-language mailing lists, forums, or other standard communication methods for the territory affected by the change
or if your edit affects only a town or small region, the local mailing list, forums, or other standard communication methods of the area affected by the change.
So thank you for the list of things that you decided to edit and the rationale behind it - but next time it would be really good to post that upfront and ask for comments.
OpenStreetMap is a map of the world, created by people like you and free to use under an open license.
Hosting is supported by UCL, Bytemark Hosting, and other partners.