User talk:RKDdata

Welcome to Wikidata, RKDdata!

Wikidata is a free knowledge base that you can edit! It can be read and edited by humans and machines alike and you can go to any item page now and add to this ever-growing database!

Need some help getting started? Here are some pages you can familiarize yourself with:

Introduction – An introduction to the project.
Wikidata tours – Interactive tutorials to show you how Wikidata works.
Community portal – The portal for community members.
User options – including the 'Babel' extension, to set your language preferences.
Contents – The main help page for editing and using the site.
Project chat – Discussions about the project.
Tools – A collection of user-developed tools to allow for easier completion of some tasks.

Please remember to sign your messages on talk pages by typing four tildes (~~~~); this will automatically insert your username and the date.

If you have any questions, don't hesitate to ask on Project chat. If you want to try out editing, you can use the sandbox to try. Once again, welcome, and I hope you quickly feel comfortable here, and become an active editor for Wikidata.

Best regards!

Václav Levý

Seems dates of birth and death for https://rkd.nl/en/explore/artists/433734 are a century too late. Cheers, --Marsupium (talk) 15:48, 5 July 2019 (UTC)[reply]

@Marsupium: Thanks for pointing that out! Corrected in our database. Best, --RKDdata (talk) 11:27, 21 November 2019 (UTC)[reply]

Willem van der Lelij

Willem van der Lelie (Q65645828) lijkt twee RKDids te hebben: 96127 en 416948 --Hannolans (talk) 09:56, 26 July 2019 (UTC)[reply]

Dank voor de aanwijzing; ik zal kijken of die verwerkt kan worden. Overigens komen alle QIDs met een dubbele RKDid op deze lijst, die ook (onregelmatig) periodiek door het RKD verwerkt wordt. Groet, --RKDdata (talk) 12:16, 29 July 2019 (UTC)[reply]

Ik zie dat het nog twee verschillende personen zijn. Gaat het inderdaad om twee personen of is het nog niet verwerkt? --Hannolans (talk) 06:02, 14 October 2019 (UTC)[reply]

@Hannolans: Deze moet nog verwerkt worden. Zoals ik eerder al schreef werken we periodiek de lijst van single value violations voor P650 (en P350) bij, dus er kan enige tijd tussen een melding op WD en bijwerken in onze eigen databases zitten. In dit geval is dat wat langer dan we normaal proberen aan te houden, maar hij is dus nog niet vergeten! Groet, --RKDdata (talk) 11:23, 21 November 2019 (UTC)[reply]

Gerrit van Giessen / G. van der Giesen

Zijn topografisch tekenaar Gerrit van Giessen (https://rkd.nl/explore/artists/91335) en 'beeldhouwer' G. van der Giesen (https://rkd.nl/explore/artists/207479) dezelfde persoon? Periode en plaats van werkzaamheid komen aardig overeen. AnnikaHendriksen (talk) 16:27, 26 July 2019 (UTC)[reply]

Daar lijkt het inderdaad op. Dit is inmiddels verwerkt. Dank voor de opmerking! Groet, --RKDdata (talk) 11:31, 21 November 2019 (UTC)[reply]

Pedro Menchón

– https://rkd.nl/en/explore/artists/400854 died before his birth. --Marsupium (talk) 02:37, 28 September 2019 (UTC)[reply]

@Marsupium: Not anymore. Thanks for pointing out. Added rkdartists id to wd item as well. Best, --RKDdata (talk) 11:40, 21 November 2019 (UTC)[reply]

Louise Wilson

RKD Louise Wilson 267946 staat geboren 1962, maar ze was tweeling en haar tweelingzus werd geboren in 1967. Ze is verward met naamgenoot Louise Wilson (Q53530907). --Hannolans (talk) 06:00, 14 October 2019 (UTC)[reply]

@Hannolans: Dat is een onfortuinlijke verwisseling. Nu gecorrigeerd. Dank voor het doorgeven! Groet, --RKDdata (talk) 12:12, 21 November 2019 (UTC)[reply]

Clara von Wille date of birth

Seems the date of birth of https://rkd.nl/explore/artists/234006 should be 1838, not 1938. Best, --Marsupium (talk) 11:17, 25 November 2019 (UTC)[reply]

Corrected. Thanks for pointing out! --RKDdata (talk) 09:51, 11 December 2019 (UTC)[reply]

Le Clerc workshop and RKDartists

458466 seems to mix up a person and his business. Is it possible to split the two? Multichill (talk) 15:37, 11 December 2019 (UTC)[reply]

Sarah Bernhardt

The very first Single value violation concerns two RKD records that need to be merged. Cheers! --Vladimir Alexiev (talk) 08:12, 19 August 2020 (UTC)[reply]

Q4605: 209830, 111223 (Sarah Bernhardt: yes, in addition to actress she was painter and sculptor)

Vladimir Rusakov (Vlady)

Merge the two RKD records of Vlady Kibalchich Russakov (Q369214): https://rkd.nl/en/explore/artists/101949 Vlady and https://rkd.nl/en/explore/artists/338580 Vladimir Rusakov.

Wikipedia says "Vladimir Victorovich Kibalchich Rusakov was a Russian-Mexican painter, known simply as "Vlady" in Mexico".

--Vladimir Alexiev (talk) 14:04, 19 August 2020 (UTC)[reply]

@Vladimir Alexiev: Thanks for letting me know! We actively use the single violations list to check our database, so there's no need to report every single item that's on there. However, there's a slight backlog on our side, and there's a problem as well with the single violations report in that it doesn't distinguish between preferred/deprecated ranks, meaning that the list is longer than it actually is. Best! --RKDdata (talk) 14:24, 19 August 2020 (UTC)[reply]

@RKDdata: It's nice that you monitor the Violations lists, but the records about Rusakov and Bernhardt are still not merged.

Could you please explain your data merging workflows/timelines? Thanks! --Vladimir Alexiev (talk) 10:00, 20 July 2021 (UTC)[reply]

@Vladimir Alexiev: There's no fixed timeline. Apparently this violation still hasn't been taken care of, but many others have and I it will be taken care of eventually. Don't worry. Best, RKDdata (talk) 07:22, 21 July 2021 (UTC)[reply]

Add RKD to VIAF

Adding RKD as a VIAF contributor would be of great value to the art community.

But coreferencing the 260k RKD entries to VIAF is no mean feat. They can be seeded with RKD-ULAN coreferences.

We're now helping the Getty with ingesting RKD into ULAN, and ULAN is a VIAF contributor, so eventually the loop will be closed. I used Beacon to download 49752 corefs RKD-ULAN that will be vetted by Getty editorial staff (saved as RKD-ULAN-20200819.txt).

There are ambiguities, but they are few (202+189):

# total lines
wc -l RKD-ULAN-20200819.txt
49752

# unique ULAN IDs
cut -d "|" -f 1 RKD-ULAN-20200819.txt |sort|uniq|wc -l
49550 (202 non-unique)

# unique RKD IDs
>cut -d "|" -f 3 RKD-ULAN-20200819.txt |sort|uniq|wc -l
49563 (189 non-unique)

# most ambiguous ULAN IDs
cut -d "|" -f 1 RKD-ULAN-20200819.txt |sort|uniq -c|sort -rn|head -3
      2 500486581
      2 500472393
      2 500386980

# most ambiguous RKD IDs
cut -d "|" -f 3 RKD-ULAN-20200819.txt |sort|uniq -c|sort -rn|head -3
      3 54715
      2 99180
      2 98925

# check some of the most ambiguous
grep -E "^500486581|54715$" RKD-ULAN-20200819.txt
500002035|Q1268739|54715
500012005|Q1268739|54715
500145637|Q1268739|54715
500486581|Q369214|101949
500486581|Q369214|338580

Let me know if I can help in any way --Vladimir Alexiev (talk) 09:53, 5 October 2020 (UTC)[reply]

@RKDdata: The complete RKDartists dataset (248,591 records) was obtained by Ontotext on 2020-12, converted to ULAN XML and final version submitted to Getty in 2021-05.

We also submitted RKD-WD links (92.5k) and RKD-ULAN links (50.4k) harvested from WD on 2020-12: Getty judged that the ambiguities are a small enough percentage.

I expect the Getty to ingest this complete dataset to ULAN sometime in late summer or early fall, at which point we could extract a full RKD-ULAN coreference from Getty LOD. But I'm not sure the Getty will add a lot more coreferences...

RKD Thesaurus vs AAT

We've also matched RKD thesauri (nationalities, roles) to AAT and submitted to Reinier van't Zelfde in the spring. Cheers! --Vladimir Alexiev (talk) 09:57, 20 July 2021 (UTC)[reply]

I've seen the list with matches between RKD thesaurus and AAT. It's still on my to do list to look at that in some more detail, but from a glance I could see that it does contain incorrect matches. Best, RKDdata (talk) 07:24, 21 July 2021 (UTC)[reply]
I just sent a refresh to Reinier (the latest versions of Getty's coreferencing to RKD thesauri):
- role-RKD-AAT.tsv : CODE is an internal Getty code, please disregard. The mapping now covers all 683 RKD roles
  - Some matches are broaderMatch not exactMatch, eg 53764 artist's wife -> 300154337 spouse
- Nationalities: two files, not sure which has the better matches:
  - nationality-RKD-AAT-term-descr.tsv has your codes,
  - nationality-AAT_xref.tsv has your names, but is the one I used when translating RKD artists records to ULAN.
@RKDdata: Please send any mismatches you find, Getty will be interested to fix them. Cheers! Vladimir Alexiev (talk) 11:53, 15 August 2021 (UTC)[reply]

Andrea Fisher -> female

Andrea Fisher is mentioned as male (https://rkd.nl/nl/explore/artists/251440) but should be female according to ULAN and Museum Arnhem --Hannolans (talk) 15:42, 21 March 2022 (UTC)[reply]

Fixed RKDdata (talk) 09:52, 22 March 2022 (UTC)[reply]

Duplicaten

Ik zie dat dit account weer actief is geworden. We hebben weer een hoop duplicaten gevonden zoals bijvoorbeeld Q115583238 met 3829 & 30212. De normale constraints lopen een beetje stuk op de triptieken en andere samengestelde werken dus ik heb een query gemaakt waar de meeste ruis eruit is. Zou je deze lijst na willen lopen en waar nodig samenvoegen in RKDimages? Graag de identifier die dan niet meer geldig is op deprecated zetten. Alvast bedankt, Multichill (talk) 10:27, 24 December 2022 (UTC)[reply]

In de RKD databases worden continu records ontdubbeld en in sommige gevallen wordt die informatie ook weer in WD verwerkt. Helaas houden de Wikidata-rapporten over constraints violations bij mijn weten nog steeds geen rekening met deprecated values, waardoor het nogal omslachtig is voor ons om deze rapporten daadwerkelijk te gebruiken. Daarnaast is het, voor zover ik weet, ook nog steeds niet mogelijk om de rank van een statement in Openrefine of Quickstatements aan te passen, wat opnieuw niet bevorderlijk is om dit soort zaken goed bij te kunnen houden. RKDdata (talk) 12:47, 23 January 2023 (UTC)[reply]

Pierre Jouffroy

Op https://rkd.nl/en/explore/artists/329466 staat al meer dan een jaar een opmerking over het feit dat het twee personen zijn die door elkaar zijn gehaald. Kijkt daar niemand meer naar? Multichill (talk) 20:24, 28 September 2023 (UTC)[reply]

Er wordt wel naar gekeken, maar helaas tijdelijk niet zo veel als we zouden willen (waarbij de termijn van tijdelijk natuurlijk ook alweer uitgelopen is...). Groet, RKDdata (talk) 06:23, 2 October 2023 (UTC)[reply]

Pieter Vogelaer

https://rkd.nl/artists/81599 has weird date of death "1710/1708. between 1700-1708". I guess the the "1700" should be a "1700" also like https://ecartico.org/persons/7951: "between 1700 and 1708"!? Best, --Marsupium (talk) 00:27, 11 November 2024 (UTC)[reply]