Talk:Key:not:brand:wikidata

Is there report not relying on Sophox queries? AFAIK Sophox remains broken with lot of data missing from it Mateusz Konieczny (talk) 09:15, 9 October 2022 (UTC)

@Mateusz Konieczny: Nothing automatic, but the current list looks like:

1. KTM (Q215657) (137)
2. Panda Express (Q1358690) (79)
3. CoCo Fresh Tea & Juice (Q65084369) (73)
4. Casa (Q1991755) (64)

Here's what I did:

1. Get the most prevalent not:brand:wikidata=* values (over 1%) from taginfo. [1]
2. Pass the values into the Wikidata API's wbgetentities action. [2] (sandbox) Alternatively, wrap the QIDs in d:Template:Q and preview the wikitext. [3]

 – Minh Nguyễn ^💬 10:35, 9 October 2022 (UTC)

@Minh Nguyen: I thought about the same value in not:brand:wikidata=* and brand:wikidata=* - your method will catch cases where correct brand:wikidata=* is set and it links brand with the same name as not:brand:wikidata=* but is actually a distinct brand Mateusz Konieczny (talk) 10:55, 9 October 2022 (UTC)

@Mateusz Konieczny: An equivalent Overpass query would look something like this, but it would time out easily. Joining with an extremely popular key like brand:wikidata=* isn't very efficient in Overpass. – Minh Nguyễn ^💬 20:06, 9 October 2022 (UTC)

Thanks for the overpass query. Is there a case where it would be appropriate for an object to have a brand:wikidata=* and not:brand:wikidata=* with the same value? Or should one of those be modified or removed?--Pa 1919 (talk) 10:23, 27 October 2023 (UTC)

"Is there a case where it would be appropriate for an object to have a brand:wikidata=* and not:brand:wikidata=* with the same value?" - no, in such case at least one of them is wrong Mateusz Konieczny (talk) 17:42, 27 October 2023 (UTC)

You may like https://maproulette.org/challenge/40177/ - "Fix conflict involving not: prefixed tags" Mateusz Konieczny (talk) 17:44, 27 October 2023 (UTC)

Thank you!--Pa 1919 (talk) 21:35, 30 October 2023 (UTC)