About this board

ReasonatorQuickStateTABAutoList2AddNameresolverBEACONBEACON2stats+laglagstatsmetricsWE-FrameworkRDAutoEditImageSearchTheGameDistributedGameWikidata listCreatorSPARQLPagePileharvesttemplatesdevrandom

Pre-flow Archives
2014 1 2 3 4 5 6 7 8 9 10 11 12
2015 1 2 3 4 5 6 7 8 9 10 11 12
2016 1 2 3 4
MediaWiki message delivery (talkcontribs)

MediaWiki message delivery 01:13, 28 March 2023 (UTC)

Reply to "Tech News: 2023-13"
MediaWiki message delivery (talkcontribs)
Reply to "Wikidata weekly summary #565"
William Graham (talkcontribs)

I see you're doing a quickstatements run for gender of humans citing to VIAF. Where in VIAF does it list gender?

William Graham (talkcontribs)

Sorry, to clarify. VIAF is an aggregate of data from other sources. So I think you should be cited to the original authority ID and not VIAF.

Lockal (talkcontribs)

Hi, there is a collapsed "About" block in VIAF. Also I import from multiple datasources - this one goes after IdRef. The next one will be BNF, I suppose. So I don't think that VIAF is bad on its own - and gender is actually stated on VIAF pages, which makes the reference valid (technically). If you think that some of my edits require additional references (i. e. VIAF+BNF), I can add them, but for now I'm trying not to overreference statements (otherwise I just have no capabilities to import data).

Reply to "Gender and VIAF"
MediaWiki message delivery (talkcontribs)

MediaWiki message delivery 01:25, 21 March 2023 (UTC)

Reply to "Tech News: 2023-12"
MediaWiki message delivery (talkcontribs)
Reply to "Wikidata weekly summary #564"

Wikidata:Database reports/unmarked supercentenarians

7
Richard Arthur Norton (1958- ) (talkcontribs)

Thanks for fixing this one too, it now runs every day. Can you remind me of how a person needs to be marked as a supercentenarian. See Q12734691, I added event=alleged_supercentenarians, but I don't think that triggers your search. Do they need to be marked as instance_of=alleged_supercentenarians?

Lockal (talkcontribs)

Hi, this is mentioned on top of Wikidata:Database reports/unmarked supercentenarians - subject has role (P2868) = supercentenarian (Q1200828) or alleged supercentenarian (Q106991708).

It is important to mention, that there were at least 2 users - @User:1Veertje (example edits) and @User:Azertus (example edits) who migrated "event" to "subject has role". And while I understand the idea, it was implemented very poorly - they both removed all qualifiers (including sourcing circumstances (P1480)) and references :/

So yes, Azertus and 1Veertje, could you please response here about the dates and types of edits you performed, I can try to build a script to analyze & restore lost data.

Richard Arthur Norton (1958- ) (talkcontribs)

Oops. I should have read that, thanks!

1Veertje (talkcontribs)

It's too long ago, I don't remember. If I were to move a claim like that today I would use the moveClaims gadget

Lockal (talkcontribs)

Ah, I have an idea for script:

For each item that has role = centenarian or supercentenarian or "alleged <...>" or "oldest human"
fetch types of edits from history
if it is a removal of statement with property "event" by any user, then fetch the state before removal, and aggregate all removed qualifiers and sources into patch for current statement and apply it.

Sounds reasonable for me, I'll try to do on weekends.

Richard Arthur Norton (1958- ) (talkcontribs)

I think we should mark all the entries as alleged supercentenarians, then if more evidence is found, they can be switched to "supercentenarians". Most are people from before we had birth records. I use the report to look for typos where people type "1900" instead of "1800" and add in an extra 100 years of life. I also use it to look for vandalism, where an IP will change the birth or death date for giggles.

Azertus (talkcontribs)

Obviously, I don't remember exactly. I think I tried to keep existing qualifiers/info, but apologies if that wasn't the case. In my archives, I found a set of QS dated 2021-06-21, and on 2021-11-11 I recorded the following queries in my notes:

     - hopeful centenarians https://query.wikidata.org/#%23title%3A%20Hopeful%20centenarians%0A%23%20Description%3A%20Calculated%20age%20of%20crrently%20living%20%28no%20death%20date%29%20centenarians.%0ASELECT%20%3Fitem%20%3FpropLabel%20%3Fage%20WHERE%20%7B%0A%20%20%3Fitem%20wdt%3AP31%20wd%3AQ5.%0A%20%20%3Fitem%20%3Fpred%20wd%3AQ2944360.%0A%20%20%3Fitem%20wdt%3AP569%20%3Fborn.%0A%20%20MINUS%20%7B%20%3Fitem%20wdt%3AP570%20%3Fdied.%20%7D%0A%20%20BIND%20%28%28%222021-06-25%22%5E%5Exsd%3Adate%20-%20%3Fborn%29%20%2F%20365%20AS%20%3Fage%29%0A%20%20%23FILTER%20%28%3Fage%20%3C%20100%29%0A%20%20%3Fprop%20wikibase%3AdirectClaim%20%3Fpred.%0A%20%20SERVICE%20wikibase%3Alabel%20%7B%20bd%3AserviceParam%20wikibase%3Alanguage%20%22%5BAUTO_LANGUAGE%5D%2Cen%22.%20%7D%0A%7D%20ORDER%20BY%20%3Fage

     - young centenarians https://query.wikidata.org/#%23title%3A%20Young%20centenarians%0A%23%20Description%3A%20Humans%20linked%20to%20centenarians%2C%20with%20a%20calculated%20age%20less%20than%20100%0ASELECT%20%3Fitem%20%3FpropLabel%20%3Fage%20WHERE%20%7B%0A%20%20%3Fitem%20wdt%3AP31%20wd%3AQ5.%0A%20%20%3Fitem%20%3Fpred%20wd%3AQ2944360.%0A%20%20%3Fitem%20wdt%3AP569%20%3Fborn.%0A%20%20%3Fitem%20wdt%3AP570%20%3Fdied.%0A%20%20BIND%20%28%28%3Fdied%20-%20%3Fborn%29%20%2F%20365%20AS%20%3Fage%29%0A%20%20FILTER%20%28%3Fage%20%3C%20100%29%0A%20%20%3Fprop%20wikibase%3AdirectClaim%20%3Fpred.%0A%20%20SERVICE%20wikibase%3Alabel%20%7B%20bd%3AserviceParam%20wikibase%3Alanguage%20%22%5BAUTO_LANGUAGE%5D%2Cen%22.%20%7D%0A%7D%20ORDER%20BY%20%3Fage


Hope that helps.

Reply to "Wikidata:Database reports/unmarked supercentenarians"
MediaWiki message delivery (talkcontribs)

MediaWiki message delivery 23:20, 13 March 2023 (UTC)

Reply to "Tech News: 2023-11"
MediaWiki message delivery (talkcontribs)
Reply to "Wikidata weekly summary #563"
Melderick (talkcontribs)

Hi Lockal, Thank you for your re-revert. I don't quite understand what happened when I checked yesterday. Must have been some cache somewhere as many months ago Stephen was 108147 and Eodon was 108148.

I did some research on Rodovid and I found this diff : https://en.rodovid.org/wk?title=Person%3A108148&diff=466531&oldid=327836

Looking at the history of the user Dn Gov I see he has changed the content of several other persons, possibly impacting the value we already had in Wikidata. That's quite confusing.

Lockal (talkcontribs)

Yes, and it was followed by a sequence of moves.

And yes, Rodovid indeed suffers from cached pages (but I used raw XML, so I was not affected), interwiki conflicts and lack of centralization - people are trying to synchronize https://en.rodovid.org/history/Person:108148 + https://br.rodovid.org/history/Den:108148 + https://de.rodovid.org/history/Person:108148 + https://fr.rodovid.org/history/Personne:108148 + https://pt.rodovid.org/history/Pessoa:108148 - and each version can contain different data...

Melderick (talkcontribs)

Well, let's hope it doesn't occur too often.

Reply to "Revert on Q2670176"
MediaWiki message delivery (talkcontribs)

MediaWiki message delivery 23:50, 6 March 2023 (UTC)

Reply to "Tech News: 2023-10"