User talk:Multichill/Archives/2017/February

Tate Collection

Hello Multichill. Nice upload with Tate collection. I just wonder if there is not an inversion with height (P2048) and width (P2049). Best regards --Shonagon (talk) 07:44, 1 February 2017 (UTC)

@Shonagon: not that I'm aware. You can lookup the original data at https://github.com/tategallery/collection . For example The Blind Fiddler (Q28539866) at https://github.com/tategallery/collection/blob/master/artworks/n/000/n00099-16172.json . Looking at the image, the dimensions don't add up. This seems to be a problem in the source data. Multichill (talk) 08:36, 1 February 2017 (UTC)

Yes Multichill. After looking in the some records, there is an inversion in the dataset on github. --Shonagon (talk) 09:11, 1 February 2017 (UTC)

Tracked at https://github.com/tategallery/collection/issues/35 . Multichill (talk) 09:18, 2 February 2017 (UTC)

Details of paintings

Hey Multichill, I have a question about Sum Of All Paintings: How to deal with details of paintings like in 'The Music Party' (Q28045233). I ask because the crotos tool shows only one of the linked images, in this case the detail. Best regards, --ChristianSW (talk) 12:00, 1 February 2017 (UTC)

Hello ChristianSW . Crotos uses only one image for artwork. In hat case the image had to be preplaced and it would be changed on Crotos after next update. I made it manually for having now a display of details of the image. The qualifier relative position within image (P2677) on depicts (P180) responds to your ask. The difficulty is to get the data (for example, depicts (P180) Anne, Princess Royal and Princess of Orange (Q239487), qualifier relative position within image (P2677) with the value "pct:10.4,22,27.5,37.7"). So we have a tool for that : IIIF Image Cropper for Wikimedia Commons and Wikidata. Those tools certainly could be improved/replaced, but for now they could help to get data and to show what is done with relative position within image (P2677). Best regards. --Shonagon (talk) 07:46, 2 February 2017 (UTC)

Hello Shonagon! Thank you for answer, your help and your great project! I wrote a blogpost about it. Best regards --ChristianSW (talk) 18:49, 2 February 2017 (UTC)

RKD

Hoi Multichill,

Dank voor je uitgebreide uitleg! Heel leerzaam! Ik ben er heel blij mee. En leuk je te zien.

Ik heb diverse functionaliteiten uitgeprobeerd, met name Mix and Match! Maar bij RKDartists zie ik een probleem.

Hier wordt oude data van RKDartists getoond. Dit wordt duidelijk omdat reeds lang verwijderde records worden getoond. Bovendien worden records met de qualifier "(in bewerking)" getoond. Deze qualifier is echter in het voorjaar van 2016 van records gesloopt en komt niet meer voor in de database.

https://tools.wmflabs.org/mix-n-match/visual_match.html#catalog=13

Kun jij eens kijken? En dit verwijzen naar de juiste persoon?

RKDdata (talk) 17:19, 12 November 2016 (UTC)

@RKDdata: Ik vond het ook erg leuk om bij jullie te gast te zijn. Ik heb je vraag op Topic:T4gewbu82uv1mz86 gezet. Multichill (talk) 16:22, 23 November 2016 (UTC)

Beste Maarten,

Kun je deze items toevoegen aan de RKDimages Singe Violations exceptions? Ik denk dat ik het zelf ook kan, maar misschien goed dat je deze nog even ziet!

En ik ben druk bezig met het verwerken van de rest van de lijst!

Q17323991: 21227, 21228 Q17323992: 21225, 21226 Q17324095: 29131, 29133 Q17324156: 45369, 25204 Q17324634: 216151, 26213 Q17324927: 5341, 5342 Q17325400: 23744, 148784 Q17327432: 29157, 216284 Q17334357: 5877, 167396 Q17340681: 21366, 21365 Q17814869: 129236, 133376 Q19926037: 27505, 27506 Q20631541: 21000, 21001

RKDdata (talk) 10:32, 27 November 2016 (UTC)

@RKDdata:Dat gaat hard! Ik heb de meeste toegevoegd, alleen deze twee snap ik niet:

• Q17325400: 23744, 148784
• Q17334357: 5877, 167396

Zijn dat niet gewoon duplicaten? Multichill (talk) 12:14, 27 November 2016 (UTC)

Hoi,

Je hebt natuurlijk gelijk. Dit is waarschijnlijk een knip-en plakfout. Excuus!

En is dit de juiste wijze van communiceren?

RKDdata (talk) 15:23, 27 November 2016 (UTC)

Goedemorgen!

Deze reports lijken al een paar dagen niet meer te draaien. Kun jij eens kijken? Wikidata:Database reports/Constraint violations/P350 Wikidata:Database reports/Constraint violations/P650

RKDdata (talk) 08:55, 4 December 2016 (UTC)

Hoi,

Kun je dit item toevoegen aan de RKDimages Singe Violations exceptions? Q17331780: 21071, 21069

RKDdata (talk) 13:07, 6 December 2016 (UTC)

Hoi Willem, het lijkt erop dat KrBot weer stuk heeft en dat DeltaBot het heeft overgenomen. Rapport is zo te zien weer bij. Ik kan hier verder helaas vrij weinig aan doen.

Ik heb de uitzondering opgenomen.

Zoals je al gemerkt heb ben ik RKDimages links aan het toevoegen. Als ik hier kijk dan zie dat jullie ongeveer 125.000 schilderijen hebben. Linksonder is te zien wat de grootste collecties zijn (Rijksmuseum als grootste "echte" collectie), maar dat zijn er maar een paar. Is het mogelijk om een wat langer lijstje te maken? Bijvoorbeeld de top 100 collecties met schilderijen? Dat is voor mij een goed startpunt om de boel aan elkaar te linken. Als dat mogelijk is en niet teveel moeite, zou je het me dan per email kunnen sturen? Multichill (talk) 13:52, 6 December 2016 (UTC)

Hoi Willem, bedenkt voor het lijstje! Daar ga ik zeker mee aan de slag. Ben je in de gelegenheid op Topic:T4gewbu82uv1mz86 te reageren? Multichill (talk) 11:38, 8 December 2016 (UTC)

Hoi Maarten,

Kun je dit item toevoegen aan de RKDimages Singe Violations exceptions? Q3428057: 106227, 106229, 106230, 106231

RKDdata (talk) 13:36, 18 December 2016 (UTC)

 Done Multichill (talk) 13:57, 18 December 2016 (UTC)

Hoi Maarten, Kun je dit item toevoegen aan de RKDimages Singe Violations exceptions? Q17336037: 39223, 36401 Q3539569: 37070, 37071, 37069

Het done vinkje is handig!

RKDdata (talk) 10:45, 21 December 2016 (UTC)

 Done Donne Triptych (Q3539569) was een hele puzzel. Die had ik door elkaar gehaald met Two Panels from a Triptych (Q26528063). Multichill (talk) 11:06, 21 December 2016 (UTC)

Kun je deze items toevoegen aan de RKDimages Singe Violations exceptions? Q19960949: 44344, 56741 Q27145532: 50348, 50380 RKDdata (talk) 20:24, 22 December 2016 (UTC)

Ik heb nog eens goed naar Archduke Leopold Wilhelm in his Gallery in Brussels (Q19960949) gekeken. Ik zie nu pas dat het drie vrijwel identieke schilderijen zijn. Voor het werk in Spanje hebben we Archduke Leopold Wilhelm in his Gallery in Brussels (Q27892947), maar zo te zien zijn de andere twee door elkaar gehaald. Ik heb een berichtje op Talk:Q27892947 gezet.

De andere staat erbij. Multichill (talk) 11:41, 23 December 2016 (UTC)

Kun je deze items toevoegen aan de RKDimages Singe Violations exceptions? Q28030290: 116509, 116511, 116514, 116516, 116518 RKDdata (talk) 07:40, 23 December 2016 (UTC)

 Done raar schilderij is dit zeg! Multichill (talk) 11:41, 23 December 2016 (UTC)

Dit betreft één object en had nooit op deze wijze mogen worden ingevoerd. Ik het dit uitgezet naar de betrokken conservator. RKDdata (talk) 12:16, 26 December 2016 (UTC)

Kun je deze items toevoegen aan de RKDimages Singe Violations exceptions? Q20275922: 248887, 248888 RKDdata (talk) 11:55, 26 December 2016 (UTC)

Kun je deze items toevoegen aan de RKDimages Singe Violations exceptions? Q20808359: 23607, 23663, 23665, 23666 RKDdata (talk) 08:48, 28 December 2016 (UTC)

Beiden staan erbij. Multichill (talk) 13:57, 28 December 2016 (UTC)

Kun je deze items toevoegen aan de RKDimages Singe Violations exceptions?
Q28073551: 21886, 21888, 21891, 195189
Q28074305: 60470, 60471
Q28061237: 28890, 28888, 28892
Q28060929: 41267, 46799, 46800, 41268
RKDdata (talk) 09:42, 2 January 2017 (UTC)

 Done staan erbij. De Cultural Heritage Agency of the Netherlands Art Collection (Q18600731) staat er bijna op. Dat zijn meer dan 10.000 werken dus daar zal nog zeker wel wat werk uitkomen. Multichill (talk) 10:06, 2 January 2017 (UTC)

Kun je deze items toevoegen aan de RKDimages Singe Violations exceptions?
Q20808487: 42233, 42698
Q20809408: 42628, 42629
Q24063770: 203647, 203648
RKDdata (talk) 16:19, 3 January 2017 (UTC)

@RKDdata: de eerste niet gedaan de andere twee wel. Die eerste is een instinker:

• Moses and the Burning Bush, with Moses Removing His Shoes (Q20808487) / Cat. 339 / RKDimages 42698
• The Expulsion of Saint Joachim from the Temple (Q20808288) / Inv. 339 / RKDimages 42233

Lekker handig van ze om "Cat." en "Inv." door elkaar te gebruiken! Dat zal nog wel een paar foutjes geven. Zou je het aan jullie kant ook willen aanpassen? Multichill (talk) 08:55, 4 January 2017 (UTC)

Je hebt gelijk. Beide werken hebben dezelfde inventarisnummers. Ik laat dit controleren en bijwerken. RKDdata (talk) 11:27, 4 January 2017 (UTC)

Kun je deze items toevoegen aan de RKDimages Singe Violations exceptions?
Q21617142: 216295, 64552 Q21614174: 50571, 50572 Q24061153: 6791, 6790 Q28113764: 5344, 5347 RKDdata (talk) 15:42, 9 January 2017 (UTC)

 Done behalve Portrait of Susanna Gerarda van Halm (1725-1801) (Q24061153), die heb ik verplaatst naar Portrait of Paul Engelbert Martens (1713-1780) (Q24061245). Zou je op 6791 het inventarisnummer willen aanpassen van 18038 naar 18037? Multichill (talk) 16:29, 10 January 2017 (UTC)

Kun je deze items toevoegen aan de RKDimages Singe Violations exceptions?
Q28094532: 229781, 229782, 229783, 219299 Q20891230: 230163, 219154, 230162 Q28113761: 5345, 5343 RKDdata (talk) 17:40, 20 January 2017 (UTC)

 Done Sorry voor de vertraging. Multichill (talk) 18:13, 16 February 2017 (UTC)

Kun je deze items toevoegen aan de RKDimages Singe Violations exceptions?
Q2516599: 55856, 55855, 55857 Q20891230: 230163, 219154, 230162 Q28094532: 229781, 229782, 229783, 219299 Q28113761: 5345, 5343 Q20536822: 21361, 21362 RKDdata (talk) 12:57, 8 February 2017 (UTC)

•  Done en deze ook, is alles weer bij. Multichill (talk) 18:13, 16 February 2017 (UTC)

Hello, I am trying to compile a list of villages in Lithuania with coordinates pulled from Wikidata. See en:User:Renata3/sandbox/A for the first batch. However, there are many issues with the coordinates and it seems most of them originate from bot import back in 2014. The problem seems to be the precision of coordinates (see, for example, Q9387657 and this diff). Is there a way to mass-fix them? Thanks, Renata3 (talk) 05:00, 6 February 2017 (UTC)

Renata3: Sorry to hear you're running into problems. To explain the chain:

• The coordinates are set in a template, sometimes with a dimension and sometimes not
• Extension:GeoData extracts the data (see that page for the dimension conversion table)
• This extracted data is available in the api (dim 1000 in this case)
• The dim is converted to precision using this function, I hope you like math (1000-> 0.1540768015183 on this longitude)
• Bot adds this

So what are the issues? What is wrong with the precision? Was the starting dimension of 1000 wrong? If I get my finger behind what goes wrong, I might be able to come up with a way to fix it. Multichill (talk) 15:56, 6 February 2017 (UTC)

So the issue is that when I try to pull the coordinates onto en wiki via {{WikidataCoord}}, if the precision is set at 0.15407.... the outcome is an error message "malformed coordinate data" (see en:User:Renata3/sandbox/A). If I fix the precision to, say, ±0.001° - the template works fine and the coordinates are pulled correctly. I really don't know what the proper fix here is, I am not a techie and I don't like math, but this weird precision makes the coordinates unusable on en wiki. Renata3 (talk) 18:19, 7 February 2017 (UTC)

@Renata3: So en:Template:WikidataCoord is busted? That one uses en:Module:WikidataCoord so that one is probably broken. I looked at your listed and notice something:

• Akmeniškiai (Q11033382) - 55°13'12"N, 23°17'17"E works
• Q11169039 - 54°24'N, 25°26'E is broken

Notice how the second one misses the seconds. That's valid output just like 55°N, 23°E is valid. I patched up the module for that case, but that module is just horrible. Asking around if better code is available. The bottom ones are broken because of reaching the maximum of allowed LUA calls on the page. Multichill (talk) 21:07, 7 February 2017 (UTC)

...and I have no idea about Modules or LUA. I am just trying to use Wikidata as intended, and it's not working. Thank you for patching the Module - it seems the rest of the broken coordinates are because no coordinates exist on Wikidata or because the page exceeded the max transclusions. Hopefully it can be further improved. Thanks again. Renata3 (talk) 21:55, 7 February 2017 (UTC)

Quantity on ART UK links

I've replied on my talk page; possibly you haven't seen it, but I hope it answers your comment.

I would though now like to get on, because I would like to use this data, and it doesn't make a lot of sense to have it half-in and half-out of the database.

If you feel strongly about it, and feel creation of some special-purpose subproperty of P1114 specifically for this kind of use would be useful, rather than just the generic P1114, that's fine and you're very welcome to propose it.

But I hope you won't mind me completing the population of the qualifier using P1114 for the time being, so that the data can be in place and complete and usable. If desired, it would seem to be a standard enough bot job at any time in the future to migrate the information from P1114 to any new more specific property that might be created for the purpose.

I hope this is okay; let me know if there is still a problem. Otherwise, if you seem to be around, and I don't hear from you, let me get back to pushing the rest through QuickStatements.

Thanks for taking an interest, and I do look forward to your thoughts as to whether you think a more specific property may be justified and worth proposing. But it does seem quite a trivial series of edits really.

All best, Jheald (talk) 20:09, 6 February 2017 (UTC)

I posted at Project Chat, and on Property talk:P1367, but it doesn't seem to have attracted much interest (the one comment so far seems content).

Would it be all right to finish off the data upload now? I'd also propose to create a new property proposal, "match count", to be a qualifier on identifiers, for the number of items in a database related to this identifer, where discussion can continue. Jheald (talk) 11:24, 8 February 2017 (UTC)

New property now proposed at Wikidata:Property_proposal/Authority_control#match_count. Feel free to help find a better name.

In the meantime, I have restarted Quick Statements with P1114, so that the pattern will consistent, rather than half done, half not. Jheald (talk) 16:07, 9 February 2017 (UTC)

Missing inventory nrs from Prado?

As promised, here are the two paintings that I expected to find in your Prado list but weren't there:

• 
  this one had a Wikidata item, but was misattributed to a whole different collection
• 
  I haven't created an item for this yet

-forgot to sign Jane023 (talk) 14:19, 8 February 2017 (UTC)

OK after realizing that most of the Prado Teniers paintings were incorrectly set to David Teniers the Elder instead of David Teniers the Younger (duh - I think I did this last year at some point) and lining all the numbers up I traced the gaps and the problem seems to be that you only scraped the english website but the spanish version has more paintings. So e.g. the painting on the right has a link that takes you to the spanish page for P01810 but when you click on the english version you get a 404. Can you see if you can re-run your Prado bot to pick up the missing numbers? I am sure the missing Teniers paintings are not the only ones. Jane023 (talk) 10:04, 14 February 2017 (UTC)