I did not say "changing back and forth", it only changed a few times in 13 years. Of ~90 edits, the node name was changed 3-4 times as was the relation name. Looking at the relation/36989, it started with an English name in 2008, which became Italian in 2009 and latin in 2010, then again Italian from 2016 to 2018. Admittedly, I care less for the "name" tag of the relation, because rendering is done from the node. There is no automatism to put 2 names in "name" just because both names are official, indeed the official name should go in "official_name". Double naming in "name" is a last resort if it is required to respect a language minority or for equally important languages to have both. Putting a very long name in the "name" tag is counter productive because it will either lead to no name at all being displayed, or other important names being omitted because of the occupied space.

Hi, I noticed that you added landuse=religious to a whole country. Can you explain why? Is a museum "religious" landuse? A bank, a supermarket or a post office? A park? A police station? The administration of a portfolio worth billions of dollars? Should we also add landuse=religious to all of Iran?

Have a look: https://pewu.github.io/osm-history/#/node/424311883
before your edit, the name of this node was Italian only, since 2016-08. Before this, it was Latin since 2013, before it was Italian.

Ciao VioMaffe, ieri hai cancellato il museo delle memorie a Corviale, perché?

regarding the double naming, it is an exceptional case here as well, because Latin is the official language but it is a dead language, no mother tongues here. I’m unsure and would discuss it on the tagging mailing list.

I am tending towards place=country, or is there already another node with the tag?

if we decide to remove any of city/town/village it could get place=suburb IMHO, while it is very small by inhabitants, it is very important on a global scale, and I guess it is also important regarding for example economical indicators like the GDP (compared to other suburbs). It used to be a place=country tag on this node, which is undoubtfully correct as well, maybe it is an option?

Ciao Stefano Trinca, ho notato che hai riclassificato il Parco Naturale Regionale Appia Antica da nature reserve (riserva naturale) a park (parco urbano). Per come noi interpretiamo il tag "park", il parco dell'appia antica non può essere considerato un leisure=park, perché ci sono anche aree residenziali e industriali (AMA) al suo interno. Poi sembrebbe che qualcosa è andato male con questa way: way/842337730 perché è stato aggiunto name=B? Non ti preoccupare, ho già sistemato...
Un saluto
Martin

Grazie Davide per la risposta, stavo per mappare nohousenumber=yes all'ingresso di Via Adorno, ma secondo dei fonti non utilizzabili, il civico è 60 e la targa c'è. Ho aggiornato la mappa con i tuoi commenti, grazie ancora, Martin

agostini:
the issue I see with the Agostini is that you did not delete place=city_block, you deleted the whole object with all translations and the wikidata link and wp link (which both are about the Augustine Order and not about this facility, agreed, but could have been retagged to wikidata:operator or something alike). I am not sure the relation is about the same thing, because one way tagged place=city_block, the other is just a landuse with a name. From looking at this, eventually amenity=monastery with community tags would be a better representation. Anyway, both objects are far from perfect, there are even religion and denomination tags missing. Removing the node in favor of the area is fine, but the tags should be transferred.

vatican city
Thinking about it, maybe neither town, nor city are fitting, because the vatican is actually just a part of Rome (not of Comune di Roma or Roma Capitale, but of Rome as a settlement), it is not a town on its own. On the other hand, the situation is so exceptional (a country as part of a city) that developing specific tagging is maybe not suitable either. I'm quite sure the number of inhabitants is not the most important property to look at, if you wanted to classify the Vatican. Generally we also look at functional features, is there an airport, a train station, a university, banks and offices, monuments, museums, theatres, cinemas, industry, supermarkets, shops, ...
The vatican has some of them (even world class), while others are completely missing. Functionally, it is completely dependent on Rome.

After looking deeper into this edit I can see that removing country information from the place=city/town makes sense. I am unsure about the bilingual name, but would think this should be discussed with the Italian community. The classification change from town to city has so much impact that I reverted it for the moment. Of course, everything is possible, but I would expect a discussion before making a breaking move like this.

forgot it, please also explain the deletion of the ordine di Sant'Agostino: node/8709827024/history

Hi, in this changeset you have removed a lot of tags from the vatican city, namely ISO3166-1:alpha2, contry_code_fips, and all official name tags (15), and introduced a very long bilingual value in the name tag, all with the comment "fixed tags vatican city". You also changed the place label from town to city, with no discussion anywhere, this is a major breaking change for the area and a "city" classification seems questionable for the vatican (it is for the first time in 13 years and 82 object versions that this value is set for place). Can you please explain your motivation and reasoning for this edit?

Thank you,
Martin

Ciao Davide Ser, ho notato che al Lotto57 ci sono tag per indicare l'indirizzo Via Caffaro 24. Vorrei chiederti se fosse giusto, intendo se questo è l'unico ingresso a tutto il lotto, altrimenti dovrebbe essere spostato il civico su un nodo alla sua posizione.
Grazie,
Martin

Ciao FraMauro, per qualche svista (credo) hai aggiunto dei tag address sulla strada way/124087393.
L'ho trovato con il nuovo tool https://nominatim.org/qa/
un saluto,
Martin

Hi, I just wanted to let you know in this changeset you somehow ended up with a broken multipolygon relation (only inner ways). Maybe this is an iD-editor bug?
relation/106272

ich glaube ich stand auf dem Schlauch, mit "Wikipedia/Wikidata-Einträge [angepassen]" war gemeint, die osm tags für WP und WD auf andere Entitäten zeigen zu lassen, nicht die Inhalte bei WP und WD zu ändern. Das kann man natürlich gerne machen, wenn ein anderer Artikel da besser passt, wobei der andere WP Artikel z.B. sagt, Dessau die Stadt sei jetzt Dessau ein Stadtteil geworden.
Es gibt z.B. https://www.wikidata.org/wiki/Q487070 dort wird "Dessau" als "town" und als "Ortsteil" und als "municipality" bezeichnet (was m.E. falsch ist, municipality ist Dessau-Roßlau, nicht Dessau).
https://www.wikidata.org/wiki/Q3828 ist ebenfalls eine municipality laut wikidata, und ersetzt die vorgenannte 487070.
Wikidata ist nach wie vor ein Riesenhaufen Gemischtwaren.
Die englische WP sagt von Dessau-Roßlau, es sei ein "urban district" https://en.wikipedia.org/wiki/Dessau-Ro%C3%9Flau
und aber auch, es sei eine "town".
Von "Dessau" wird ebenfalls gesagt es sei eine "town": https://en.wikipedia.org/wiki/Dessau und "former municipality".

die Wikipedia und Wikidata Artikel muss man nicht ändern, der Anspruch dass das 1:1 passt lässt sich oft nicht erfüllen weil die WP Artikel mehrere Aspekte in einem Artikel behandeln, und Wikidata im wesentlichen 1:1 zu den Artikeln Objekte hat, also auch mischt.

Ich glaube Dessau-Roßlau ist für die admin Relation genau richtig, aber die Stadt heißt immer noch Dessau. Im Rathaus von Dessau sitzt jetzt die Stadtverwaltung von Dessau-Roßlau. Das Bauhaus Dessau ist immer noch das Bauhaus Dessau, Dessau Hauptbahnhof, etc.

Zu dem node/240098523 place=county sagt mir nichts, Grafschaft? Gibt es die OpenGeoDB noch? Wenn noch jemand diese tags auswertet im Zweifel lieber lassen

Multipolygone erst einzusetzen ab 2000 Nodes bedeutet andersrum betrachtet ja, dass man erstmal die Fläche an 2000 Stellen mit anderen Flächen verknüpft hat und das jedesmal wieder auftrennen muss wenn man was ändert. Ich würde die Grenze eher bei 5-10 nodes sehen