Frettiebot

Welcome to Wikidata, Frettiebot!

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! Liuxinyu970226 (talk) 03:44, 10 December 2014 (UTC)

en descriptions

Hi Frettie,

At Special:Diff/179524625, it should read "Slovak". --- Jura 20:53, 1 July 2015 (UTC)

P735 additions

Hi Frettie,

Would you make sure that items use for P735 have P31 with "given name".

The following makes no sense:

• Smil II. z Lichtenburka (Q12054987) ‎ (‎Created claim: given name (P735): SMIL (Q417357)) (

Please double check your edits. --- Jura 18:49, 4 October 2015 (UTC)

Hi Jura,

If i have python script, when i see only name to fill empty names, i can place errors. I can control it, thanks.--Frettie (talk) 19:45, 4 October 2015 (UTC)

Some of your contributions ended up on Wikidata:Database_reports/Constraint_violations/P735. Together with Pasleim we fixed most of them. --- Jura 14:48, 6 October 2015 (UTC)

Cosmas of Prague

Hi

Can you explain why this happens:

• Kosmas (biskup praski) (Q1136093) ‎ (‎Created claim: given name (P735): Cosmas of Prague (Q317916))

It's seems really strange. --- Jura 18:54, 4 October 2015 (UTC)

Vincenc

Hi

This

• Vincenc Adam (Q11901526) ‎ (‎Created claim: given name (P735): Vincent (Q4928254))

should link to an item with the label "Vincenc" not "Vincent". --- Jura 18:56, 4 October 2015 (UTC)

I think Vincent is similar given name like Vincenc. But ok.--Frettie (talk) 19:48, 4 October 2015 (UTC)

At WikiProject Names we are trying to get the labels to match the name being used. --- Jura 19:52, 4 October 2015 (UTC)

Wow, thanks, nice work!--Frettie (talk) 19:55, 4 October 2015 (UTC)

Mikroregiony

Ty mikroregiony se nějak nepovedly - Special:Contributions/Frettiebot, prosím o opravu. --Jklamo (talk) 00:50, 30 October 2015 (UTC)

Povedly, na opravě se současně s těmi editacemi pracuje, viz Special:Contributions/Frettiebot.--Frettie (talk) 00:53, 30 October 2015 (UTC)

Reflexive "instance of" link

I undid this change because it made mausoleum an instance of itself. Bovlb (talk) 22:00, 30 October 2015 (UTC)

Přírodní parky

Ahoj, pozor na takovéto editace. Ne vše, co je v kategoriích Přírodních parků, je Přírodní park :-). Dík, --Vojtěch Dostál (talk) 08:55, 30 September 2016 (UTC)

Ahoj, no, to jako jo, ale bohužel to je z toho, že kategorie fungujou tak, jak fungujou. :)--Frettie (talk) 22:12, 30 September 2016 (UTC)

Počet obyvatel 2011

Nazdar, koukám, že jsi před nějakými dvěma roky přidával k obcím (všem?) historické počty obyvatel až do roku 2001. A pak až 2014. Myslím, že by se tam hromadně měl objevit i rok 2011, kdy, stejně jako v předchozích letech, bylo sčítání lidu, které tu nyní chybí. --Harold (talk) 13:48, 5 October 2016 (UTC)

Počty domů

Ahoj, díky za ten import počtů domů v obcích ze sčítání. Měl bych ještě jeden malý námět na vylepšení a to u nejaktuálnějšího údaje (2011) zadan rank preffered. Ono tedy vůbec to zadávání přes has part(s) (P527) a house (Q3947) je divné, ale chápu že aktuálně to jinak nejde. Asi taky hodím návrh na novou property.--Jklamo (talk) 15:39, 21 April 2017 (UTC)

Ahoj, mě to taky přišlo divný, ale User:Vojtěch Dostál to někde diskutoval a tohle bylo výsledkem. Preferred mě nenapadlo, to je pravda - zkusím na to kouknout a u těch chybějícíh cca 70 tisíc to hodit rovnou - u dalších snad půjde to upravit. --Frettie (talk) 15:42, 21 April 2017 (UTC)

Ahoj, no aktuálně opravdu nic kloudnějšího nejde, takže je to OK. A ještě možná jeden námět, bylo by dobré indikovat, k čemu se číslo vztahuje, tedy jestli k obci nebo katastrálnímu území/části obce (třeba pomocí sourcing circumstances (P1480)). Zatím se to moc nerozlišuje, ale časem na to myslím dojde a pak bude fajn vědět, k čemu se čísla vztahují.--Jklamo (talk) 15:47, 21 April 2017 (UTC)

No já nevím, tohle je snad asi ok - tohle se vztahuje právě k té entitě, která tam je. Pokud jsem to dobře pochopil. Víc ví asi Vojta.--Frettie (talk) 15:56, 21 April 2017 (UTC)

U vesnic/katastrálních území/částí obce je to OK, ale problem je u obcí. Tam máme většinou zatím jednu položku/entitu pro článek, který reprezentuje zároveň celou obec a zároveň katastrální území/část obce její sídlo. To jsou dvě různé věci a jsou pro to odlišná čísla jak u obyvatelstva tak u počtu domů. Proto by bylo dobré to již při importu rozlišovat.--Jklamo (talk) 16:43, 21 April 2017 (UTC)

Tak preferred už se mi tam povedly zavést.--Frettie (talk) 16:11, 21 April 2017 (UTC)

Fajn. Mimochodem Wikidata:Property proposal/number of houses.--Jklamo (talk) 16:43, 21 April 2017 (UTC)

Ahoj, upřesňuji že Frettie importuje k částem obcí, nikoliv k obcím. Viz toto query hledající obce s uvedenou informací o počtu domů - prázdné. Tedy odpadá problém obec vs. stejnojmenná část obce. Jinak co se týče samostatné property, nemám vyhraněný názor, na to rozumím struktuře Wikidat příliš málo --Vojtěch Dostál (talk) 18:36, 22 April 2017 (UTC)

Streets in the Czech Republic

Ahoj, měl bych ještě jeden dotaz na ty importované ulice. Českou label dotyčných položek tvoří vždy jen názen ulice, např. "Komenského", zatímco v angličtině je jako label v takovém případě uvedeno "Category: Komenského (Litovel)", viz např. Q43462454. Bylo by možné ty labely nějak sjednotit? Nejlepší by asi bylo změnit tu anglickou podle vzoru té české, protože kdyby ta položka měla být o kategorii, tak by popis byl "category in Wikimedia projects", nikoliv "street in Litovel". Co myslíš? --Jan.Kamenicek (talk) 20:04, 5 December 2017 (UTC)

Ahoj, vim o tom, je to v planu. Frettie (talk) 20:06, 5 December 2017 (UTC)

Chybka

[1]. Nevím kolik daších článků bylo ovlivněno. --Silesianus (talk) 08:58, 19 July 2018 (UTC)

Ahoj, jo, to tak je, když se objeví nějaká podobná blbost - upravil jsem i další podobný věci. Dík.--Frettie (talk) 09:07, 19 July 2018 (UTC)

wrong edits

Please try to avoid them. For example one can never be the child and father at the same time. Example: https://www.wikidata.org/wiki/Q110456 FS100 (talk) 12:28, 30 July 2018 (UTC)

@Frettie: Another example of the same I had to revert: https://www.wikidata.org/w/index.php?title=Q5056379&diff=712733906&oldid=710397613 I hope the bot was fixed since then? Laboramus (talk) 20:27, 19 October 2018 (UTC)

@Laboramus: Yes, this was one time operation. Thanks.--Frettie (talk) 07:50, 22 October 2018 (UTC)

Nonsensical place of birth and death

Hi,
looks like there have been added nonsensical places of birth or death, ossuary (Q255595) which isn't an instance of a location: here and here, no idea how widely spread this problem is, but you should probably check your edits.
Thank you! Cheers, --Marsupium (talk) 16:25, 14 September 2018 (UTC)

Hello, thank you, i correct my script. Thank you very much!--Frettie (talk) 06:45, 16 September 2018 (UTC)

native label (P1705) on family name items

Hi Frettie,

Generally, these items would just have one such statement, with the language code "mul", e.g. at Q1605060#P1705. Please avoid adding Czech in addition.

Where applicable, Czech could be added with language of work or name (P407). --- Jura 08:57, 23 January 2020 (UTC) hi, ok, thanks. --Frettie (talk) 12:30, 23 January 2020 (UTC)

Názvy polských vesnic

Nazdar, myslím že hromadný import názvů polských vesnic z angličtiny nebyl úplně šťastný: jednak se na WD rozlišovače nepoužívají, jednak to natahalo nesmyslné anglické názvy vojvodství do českých štítků (viz [2], [3]). Ideální by bylo teď podle vzoru (to by nemělo být tak těžké) nahradit anglické názvy vojvodství českými. Díky — Draceane ^talk_contrib. 08:28, 24 January 2020 (UTC)

Ahoj, příště to napiš ideálně přímo mně. Chápu to, nicméně je potřeba nějak začít a teď je pro případnou opravu (se kterou počítám, kdo by čekal, že to mají anglických WD tak rozbitý) pro mě lepší, když už to dojede. Dej mi čas, upravím to. --Frettie (talk) 16:40, 24 January 2020 (UTC)

@Frettie: Díky, jasně. (Jinak: viděl jsem diskusi u bota, tak jsem psal sem a teď to nechci tahat po x diskusích. Do příště se poučím.) — Draceane ^talk_contrib. 10:21, 27 January 2020 (UTC)

Same VIAF added more than once

Hi! Firstly thank for the great number of VIAF ID (P214) you added. I want to report a little problem: as you can notice here, maybe sometimes the bot has added the same VIAF many times. Could you check and remove superfluous values? Thanks in advance and good night, --Epìdosis 23:09, 8 July 2020 (UTC)

Error

Please stop your bot edit warring with me on William M. Adams (Q2579787). It keeps adding an incorrect identifier — Martin (MSGJ · talk) 09:12, 9 December 2022 (UTC)

Hi, there is better solution than delete ISNI identifier, you can deprecate this and that will be ignored by bot. --Frettie (talk) 09:19, 9 December 2022 (UTC)

Athlete

Hello, greetings and first of all thanks for your work. However, edits like this only fill the database with inaccurate information. Athletes in almost all cases will indicate the specific sport they practice and saying that it is only an athlete is so generic that it is not necessary to indicate it. Please, in this case I would tell you that it is better not to include this value in this property. Thank you. Vanbasten 23 (talk) 22:29, 17 December 2022 (UTC)

@Vanbasten 23:Hi, thanks for your message, I understand that it may look wrong, but it is basically right. "Athlete" is a profession, a particular sport is also a profession. It's similar to a musician and a violist. It's also not very possible in about a million records to manually check every authorized record from the National Library of the Czech Republic. --Frettie (talk) 23:33, 17 December 2022 (UTC)

It is right? Yes. Inaccurate? Yes. My bot just to remove occurrences of generalities like "athlete", and does your bot put them in? It's not necessary. You should check if it has a previously more specific value, and if so, don't add it. It is not about adding, but rather that the data entered have quality. It´s not manually, you should program it like this. There is no point in working twice. Thank you! Vanbasten 23 (talk) 23:43, 17 December 2022 (UTC)

@Vanbasten 23 Hello. I understand your frustration, but Wikidata works like this. We do not remove sourced (though less precise) information, when it is correct. We keep it and, when available, we also add more precise information. Vojtěch Dostál (talk) 13:36, 18 December 2022 (UTC)

Sorry but no. In Wikidata we enter exact information. You must not enter inaccurate information. That there is nothing about his sports occupation, perfect, add it, but if there is already, do not put some common sense. If he is already a footballer, do not say that he is also an athlete. Thanks. --Vanbasten 23 (talk) 14:46, 18 December 2022 (UTC)

If he is already footballer, but he is ice hockey player too – he is sportsman too. --Frettie (talk) 15:01, 18 December 2022 (UTC)

Sucharit Bhakdi (Q1473437)

Bot is adding a claim which is not true. Bhakdi is NOT an epidemiologist, no wiki article has mentioned it. Please stop the bot from adding this. Thx. --Julius Senegal (talk) 18:13, 4 January 2023 (UTC)

Hello, there is category "Epidemiologists" at enwiki, and in source (Authority database of National library of the Czech Republic) is epidemiologist record too. --Frettie (talk) 18:41, 5 January 2023 (UTC)

First corrected, the whole article does not state it.

2nd irreelvant, all wikipedia entries do not describe him as such. This is an error. --Julius Senegal (talk) 09:37, 29 January 2023 (UTC)

@Frettie:. --Julius Senegal (talk) 07:52, 8 March 2023 (UTC)

Bernard Golse

Bonjour, vous continuez à insérer des informations redondantes (psychiatre alors qu'il est indiqué pédopsychiatre, information plus précise) et professeur d'université, alors qu'il est indiqué praticien-hospitalier professeur d'université, information plus précise), en vous appuyant sur une référence tchèque. Merci d'indiquer vos motivations, cordialement, Pierrette13 (talk) 07:52, 25 January 2023 (UTC)

Hi, if you find these added data redundant, you need to mark them as deprecated, then the bot will not insert them anymore. The source is the National Library of the Czech Republic. I.e. an authority with manual control. Thanks. --Frettie (talk) 10:26, 25 January 2023 (UTC)

Bonjour et merci de votre réponse, je vais regarder comment on fait pour marquer deprecated; bonne fin de journée, Pierrette13 (talk) 13:52, 25 January 2023 (UTC)

@Pierrette13 Generally, redundancy is not a valid reason for deprecation. Vojtěch Dostál (talk) 14:57, 28 January 2023 (UTC)

Hello User:Vojtěch Dostál, l'information tchèque est plus tardive et moins précise que celle qui existait préalablement et qui est tout à fait sourcée (il est pédopsychiatre et praticien-hospitalier professeur d'université). Merci de ne pas remettre cette information tchèque de moindre valeur, cela n'a aucun sens d'indiquer "psychiatre et pédopsychiatre", "professeur d'université et praticien hospitalier professeur d'université", --Pierrette13 (talk) 15:10, 28 January 2023 (UTC)

@Pierrette13 Lesser precision is not a reason for deprecation either. I am sorry but all relevant data belong to Wikidata. We are curating Wikidata by setting the most precise statements as "preferred" and not by deprecating the less precise ones... Vojtěch Dostál (talk) 15:26, 28 January 2023 (UTC)

Hello User:Vojtěch Dostál, merci de votre réponse, eh bien restons sur "préféré". Ce que je ne souhaite pas c'est que l'infobox de la page Wikipédia soit bourrée d'infos de valeurs médiocres. Par contre je ne sais pas faire sur la notice WD, pouvez-vous le faire ? Merci à vous, bonne fin de journée, --Pierrette13 (talk) 16:33, 28 January 2023 (UTC) P.S. l'idée de "deprecated" ne vient pas de moi...

Pierrette, je viens de passer les deux occupations en rang préféré? Si tu veux en savoir plus, tu peux lire Help:Ranking/fr. Pamputt (talk) 06:36, 9 February 2023 (UTC)

Thanks Pamputt! --Frettie (talk) 07:02, 9 February 2023 (UTC)

@Frettie: you should not flood the items with useless data and then expect other people to manually mark the useful data as preferred. Human users will not keep up with the bot. I 100% agree with Pierrette and with @Vanbasten 23: above. You have to avoid redundancies automatically by checking the subclasses. For example Professeur des universités – Praticien hospitalier (Q3406835) is a subclass of university teacher (Q1622272) so adding the latter is useless. --GrandEscogriffe (talk) 17:37, 9 February 2023 (UTC)

FYI the matter is now discussed on the French village's pump. I personnally disagree with GrandEscogriffe. --Jahl de Vautban (talk) 00:39, 11 February 2023 (UTC)

Q112500598

Bonjour. L'ajout de Q1622272 sur la notice Q112500598 est erroné : si la base NKC l'indique, cette personne n'est pas et n'a jamais été professeur d'université. Comment faire pour empêcher l'ajout automatisé de ce descripteur ? Cordialement, CptKeyes (talk) 09:57, 30 January 2023 (UTC)

@CptKeyes:Hello, just set the wrong record as deprecated. --Frettie (talk) 15:28, 1 February 2023 (UTC)

@Frettie:. C'est noté, merci ! --CptKeyes (talk) 20:01, 1 February 2023 (UTC)

Wrong edit

Hello! This one seems wrong to me, the subject was a child and did not have a workplace. Syrio posso aiutare? 21:09, 3 February 2023 (UTC)

@Syrio Feel free to deprecate the statement if you think it's wrong. Vojtěch Dostál (talk) 16:32, 7 February 2023 (UTC)

Wrong edits in Andreas Kalcker

It seems that the bot is adding false information as if they were true qualifications for this person taken from the Czech library entry. The problem seems to be that such affirmations are simply taken from the person's book blurb because it is just a library entry for the author of the book they have (a book which has been banned form amazon for advocating and giving instructions on bleach enemas for "curing" autism in kids...) and not from any reputable source, as this is a known fraudster and scammer who claims to have multiple credentials just see https://es.wikipedia.org/wiki/Andreas_Kalcker . I've removed the 2 wrong entries, would it be possible to avoid such problems in the future? Thanks in advance Ebergerz (talk) 13:52, 16 February 2023 (UTC)

@Ebergerz: Thanks for your message, just set the wrong information as deprecated. But here we should probably inform the National Library Czech Republic that may have incorrect data. What do you think, @Vojtěch Dostál:? I set the information as deprecated. --Frettie (talk) 18:54, 16 February 2023 (UTC)

Impossible to avoid in imports, deprecation is the best thing to do. We can send the Library a link to a query with deprecated statements later, I don't think it would be efficient one by one. Vojtěch Dostál (talk) 20:11, 16 February 2023 (UTC)

Thanks for the quick action. To be honest this goes way over my very basic knowledge about wikidata. You do an amazing work, I just happened to catch this punctual issue because it affected a page I have worked on and thus know quite a bit about the specifics case. Thanks again. Ebergerz (talk) 01:40, 17 February 2023 (UTC)

BTW I just saw that even though it has been deprecated it shows up in the infobox. Should I delete the statement from the wikidata entry? or what is the best way to manage this kind of situation? (I deleted one of them, then thought I probably shouldn't have so reverted it). Ebergerz (talk) 02:12, 17 February 2023 (UTC)

@Ebergerz: You're welcome! The infobox shouldn't display deprecated data, infobox will need be modified to not display the deprecated data. I studied source code of the Infobox and Wikidata Module, but it's in Spanish and that is problem for me (and its in Lua too :) ). So, try contact the module author and tell him that it is only appropriate to display non-deprecated values. I think it shouldn't be a problem by the source code of the module. --Frettie (talk) 08:28, 17 February 2023 (UTC)

I just deprecated wrong statements from the National Library of the Czech Republic in Q69666. I kindly ask you to establish a routine to inform the library about this. As your bot likely re-introduces wrongs statements in case they have removed from the Wikidata entry, I also ask to check if a statement has been removed before. The deprecated status was unknown to me after many years on Wikidata and it is quite labour-intensive to set it. Thanks, --Polarlys (talk) 07:21, 28 September 2023 (UTC)

Ne pas confondre la diplomatique et la diplomatie (cf. Jules Quicherat Q2408040)

Could you tell to Frettiebot that diplomatics (auxiliary science of history) and diplomacy are not the same things ? ThbdGrrd (talk) 10:49, 27 February 2023 (UTC)

Linus Westheuser (Q95223779)

Hi Frettie,

please be so kind not to sync poetry to the item, his main area of work is something different and he only made minor contributions to poetry. Thanks in advance!--Jackie Bensberg (talk) 10:02, 8 May 2023 (UTC)

Hi, i set the poetry to deprecated. --Frettie (talk) 10:07, 8 May 2023 (UTC)

Would it be possible to introduce a blacklist directly in the bot?--Jackie Bensberg (talk) 10:12, 8 May 2023 (UTC)

For one item? No, it is not good option. --Frettie (talk) 10:19, 8 May 2023 (UTC)

For every item where the librarian's classification is not undisputable.--Jackie Bensberg (talk) 10:26, 8 May 2023 (UTC)

There is no possible way to know it. --Frettie (talk) 10:29, 8 May 2023 (UTC)

I don't ask you to be a clairvoyant, just to let your bot read a opt-out Wikipage in a dumb text format.--Jackie Bensberg (talk) 10:39, 8 May 2023 (UTC)

If a property (e.g. poetry) is marked as deprecated, it is skipped. Better option is not possible - i think.--Frettie (talk) 10:45, 8 May 2023 (UTC)

Jesuit (Q12319698)

Please stop adding "jesuit" as an occupation, it is a religious_order, another bot is removing you additions, then your bot readds it. It is wasting valuable CPU time. RAN (talk) 02:12, 15 June 2023 (UTC)

You are still doing it, please address the issue. The person below is having the same problem with you. --RAN (talk) 02:29, 16 June 2023 (UTC)

You are still doing it, please address the issue. --RAN (talk) 19:33, 3 July 2023 (UTC)

Hi, i am sorry, I didn't see your message, I'll try to set it up. --Frettie (talk) 20:14, 3 July 2023 (UTC)

Edit war with KrBot

Hello, Frettiebot is in an edit war with KrBot in Paulo Trigo Pereira (Q21072405). Frettiebot puts occupation (P106): member of parliament (Q486839) and KrBot remove it and move it to position held (P39), see autofix on Property P106. I leave a similar message in User talk:Ivan A. Krestinin. Joao4669 (talk) 06:39, 15 June 2023 (UTC)

Hello, i'll check it, thanks! --Frettie (talk) 12:58, 15 June 2023 (UTC)

I've just spotted this is still going on with eg Dennis Herbert, 1st Baron Hemingford (Q5258507) - it seems to have happened six times over the past week for that one item!

In practice I think we would not normally use occupation (P106) or position held (P39):member of parliament (Q486839) for most of them - most countries use a subclass of Q486839. Herbert has member of the 37th Parliament of the United Kingdom (Q41582593), Pereiera has member of the Assembly of the Republic (Q19953703), and so on.

Would it be possible to suppress member of parliament (Q486839) from the bot's list of occupations? That would solve this problem. Andrew Gray (talk) 11:30, 17 June 2023 (UTC)

An alternative solution might be to have the bot log its edits and not re-do one that it's already done? Presumably there are others where it's just going to keep putting in the same edits after a human or an autofix bot corrects them, and that's not great. Andrew Gray (talk) 11:35, 17 June 2023 (UTC)

Hello! Please remove the addition of the member of parliament to the occupation from the list of tasks of your stick. On the one hand, the member of parliament is more of a position (P39: 4628, P106: 307) and on the other hand, it has already been said that he is waging a edit war w/index.php?title=Q716090&action=history see here or here. Thanks Pallor (talk) 14:46, 17 June 2023 (UTC)

Hi, i set it up, in next run it will be ok. --Frettie (talk) 08:15, 18 June 2023 (UTC)

Hi @Frettie:, same thing is happening with senator (Q15686806) on e.g. Marcus Antistius Labeo (Q436034). Could you also fix it? Thanks! --Jahl de Vautban (talk) 16:13, 27 June 2023 (UTC)

Hi, thanks, ill check it in few hours or days. --Frettie (talk) 08:29, 28 June 2023 (UTC)

Corrected. --Frettie (talk) 11:23, 29 June 2023 (UTC)

Please also see

• Wikidata:Forum#Editwar_zwischen_zwei_Bots
• https://www.wikidata.org/w/index.php?title=Q1751004&action=history
• User talk:Ivan A. Krestinin#Edit war with Frettiebot

M2k~dewiki (talk) 17:29, 7 July 2023 (UTC)

@Frettie: another one here: https://www.wikidata.org/w/index.php?title=Q1751004&action=history (please convert occupation (P106)freemason (Q23305046) into member of (P463)freemasonry (Q41726)). Thanks! --Epìdosis 07:31, 10 August 2023 (UTC)

Hi, it will be ok in few minutes. --Frettie (talk) 07:33, 10 August 2023 (UTC)

František Vrhel

Ahoj, nevím, jestli bot opět přidá Klatovy, nicméně v záznamu je už opraveno místo úmrtí na Prahu. RiniX (talk) 12:27, 5 July 2023 (UTC)

Teď nepřidá, je na pár měsíců zablokovanej. --Frettie (talk) 04:42, 6 July 2023 (UTC)

Joseph Daniel Huber (Q106258761)

Ahoj, Andre Engels (talk • contribs • logs) has separated Daniel Huber (Q18945169) from the above item, but Frettiebot reinstates the wrong link.-- U. M. Owen (talk) 15:48, 10 September 2023 (UTC)

Multiple places of birth

Hello, hoping you can have a look at Raihan Ismail (Q55190158) – the subject already has a place of birth (P19) = Egypt (Q79), but three times Frettiebot has added another place of birth (P19) = Australia (Q408). Firstly this appears to be incorrect in NA-CR, as the current reference says "The daughter of an Islamic scholar, Raihan was born in Egypt and raised in Malaysia. She migrated to Australia in 2007 and became an Australian citizen.", and secondly, why is the bot adding another place of birth from an authority control source when there is already a (referenced) P19 claim in the item? --Canley (talk) 03:17, 18 September 2023 (UTC)

Bot inserts the second place of birth because the NK CR is an authority for us and we honestly did not think that this could happen. The bad thing is that we don't know what is right without human check. I'll look into it. --Frettie (talk) 15:07, 18 September 2023 (UTC)

Rozmnožení vandalismu

Zjevně pokud kdokoli založí či změní jakoukoli položku a dá do ní NKČR ID nějakého povolání, pak Czech-Authorities-Sync okamžitě nakluše a namnoží odkaz na tuto položku jakožto povolání do všech položek, u kterých NKČR toto povolání eviduje. To fakt není dobré. Teď by někdo měl opravit všech těhle 1387 položek (a bůhví kolik dalších, kde už to nastalo dříve a nikdo si toho ještě nevšiml), a to tohle ještě není tak hrozné. Až třeba někdo přidá/změní NKČR v Q488111 tak, aby odkazovalo na spisovatele či matematiky, bude to větší ostuda.

Nemůžu úplně tvrdit, že vím, co s tím… Pro začátek možná ignorovat položky s duplicitním NKČR ID? Anebo spíš vůbec: Stáhnout aktuální mapování a neaktualizovat ho automaticky. Leda po ruční kontrole (beztak: jak často/jak moc se to reálně správně mění?). Mormegil (talk) 07:39, 6 October 2023 (UTC)

Ahoj, jen velmi rychle, teď se na to tak týden dva nedostanu, nebudu úplně u počítače. Ale zrovna tohle je takovej vtipnej úlet KrBota, kdy sem tam nám někdo hlásí, že zas se rozhodl přeunout nějakou další náhodnout položku podle nějakých pravidel, který asi nikde nejsou definovaný (snad ve zdrojáku KrBota). Postupně to čistíme a v předběžným plánu je získávat stejná zdrojová data o tom "co přesouvá" a mezi čím. Pač to, že ta data bereme automaticky (a že přibývají poměrně často) je v podstatě základ té práce bota. Jinak tam kontroly máme, jsou tam kontroly opakovaného zápisu, ... Ještě tam kontrolujeme více drobností, ale teď úplně nemám čas a ani prostor to hledat. Ale víme o tom. Jen nejsnazší řešení by bylo asi zakázat pseudosprávný opravy KrBota ... a možná by bylo i nejsprávnější. --Frettie (talk) 07:48, 6 October 2023 (UTC)

Ne, to fakt není, zjevně to bylo jen velmi rychle. To, co se stalo, bylo:

1. Někdo (v tomto případě nejspíše mylným kliknutím, ale klidně by to mohl být úmyslný vandalismus, to není důležité) založil položku Q122839955, ve které bylo mimo jiné i tvrzení NL CR AUT ID (P691)ph122664.
2. Czech-Authorities-Sync si povšiml, že 1387 položek (zcela namátkou Jan Slouka (Q120583391), který má Jan Slouka (Q120583391)NL CR AUT ID (P691)xx0085831) má v NK ČR uvedenu profesi matematici (ph122664), toto ph122664 je evidováno v Q122839955 (a taky v Q170790, ale co je mu po tom…), ale přitom není uvedeno Jan Slouka (Q120583391)occupation (P106)mathematician (Q122839955)! (A ono sice je uvedeno Jan Slouka (Q120583391)occupation (P106)mathematician (Q170790), ale co je mu po tom…) Proto toto tvrzení obratem přidal (do všech 1387 položek). Tohle byla chyba.
3. Jediné, kde do toho vstupuje KrBot, je to, že si o něco později Silesianus povšimnul nesmyslně zduplikované položky Q122839955 a sloučil ji (správně) do Q170790. Načež tedy KrBot (v zásadě zcela správně) všechna tvrzení occupation (P106)mathematician (Q122839955) aktualizoval na occupation (P106)mathematician (Q170790). Tohle je ta výše odkazovaná dávka 1387 editací, ale ta je sama o sobě úplně v pohodě. Problém je v tom, že existuje přes tisíc položek, do kterých FrettieBot přidal duplicitní povolání. (Viz aktuální stav toho Q120583391#P106.)

Jak jsem psal, tohle byl zatím víceméně neškodný binec. Ale až se stane, že někdo přidá pornographic actor (Q488111)NL CR AUT ID (P691)ph122664, přiběhne FrettieBot a okamžitě přidá do více než tisíce položek tvrzení, že daný matematik působil jako pornoherec. Třeba.

--Mormegil (talk) 08:26, 6 October 2023 (UTC)

Jo takhle… no to je pravda, tohle je blbá situace. Tuším, že Vojta Dostál i proto navrhoval ty hashtagy, aby se to dalo líp revertovat (ale je možný, že si to pamatuju špatně). Ale jak takovou situaci řešit, to fakt nevím, asi leda zkusit rovnat jednou za čas stav těch ph12345 položek s nkčr stavem a podezřelý výskyty (třeba dle textu či synonym) někam házet ke kontrole. Ale ani to není všespásný. Protože rychlost. Zajímavej námět k rozvoji.—Frettie (talk) 10:15, 6 October 2023 (UTC)

Ano, nějaké snadné revertování by jistě bylo dobré (a kdyby v tomhle konkrétním případě všechna přidání toho Q122839955 byla v jedné editační skupině, tak by to určitě pomohlo), ale popravdě mě úplně nenapadá, jak přesně to seskupovat (maximálně časově, všechny editace provedené daný den nebo tak něco? Nebo bot jede nějakou dostatečně identifikovatelnou logikou, podle které by to šlo?)… Podezřelé změny házet ke kontrole ex post je IMHO víceméně k ničemu. Podezřelé změny házet ke kontrole před tím, než se podle nich bot rozjede, je to, co navrhuji výše. (Resp. pořád se mi nechce věřit, že by těch změn bylo nějak hodně, takže nevidím potřebu je nějak předem filtrovat na podezřelé; prostě každou změnu někdo musí odsouhlasit. Ale možná se s tím počtem změn pletu.) A „bot může sloužit k množení vandalismu“ je IMHO spíš „chyba, kterou je potřeba docela rychle opravit“ než „námět k rozvoji“. --Mormegil (talk) 06:51, 9 October 2023 (UTC)

Konečně budu mít snad čas a sílu se tomu věnovat, vrátil jsem se z dovolený. :) Já jen přemýšlím, jak v praxi provést to "podezřelé změny hlásit dopředu", ono to moc rozumně nejde, pač to jede dost sekvenčně a z velký báze dat. Tj. ta předkokntrola je taková blbá - a jak vlastně poznat podezřelou změnu. Mám tam simulační mód, kterej provede ty změny a z toho by to šlo poznat. Ale trvalo by to. Zkusím to nějak domyslet. --Frettie (talk) 05:52, 14 October 2023 (UTC)

Bot edit war

@Frettie Hi, could you have a look at B. C. Schiller (Q15784916)? A bot edit war has been going on since February and I tend to agree with KrBot. --Emu (talk) 14:18, 8 October 2023 (UTC)

Hm, i try to deprecated lang information. It's complicated situation, because its true, that B. C. Schiller (Q15784916) speaks german ... but it is duo. hm. --Frettie (talk) 17:26, 13 October 2023 (UTC)

That doesn't work. Mormegil (talk) 18:17, 13 October 2023 (UTC)

Hm, for Frettiebot sure. For KRBot not ... --Frettie (talk) 05:53, 14 October 2023 (UTC)

@Frettie Please fix this. Thank you. --Emu (talk) 11:37, 8 December 2023 (UTC)

Hi, hm. Podíval jsem se na to, ale tady je to složitý. I looked at it, but it's complicated here. If KRBot hadn't removed the languages used, and it remained deprecated, then Frettiebot wouldn't be add this further. But since it's not obvious that it's wrong (because it disappears), it's re-imported. I will try to do some control schemes for similar situations in few days. --Frettie (talk) 11:45, 8 December 2023 (UTC)

I have no idea (and probably nobody does, only owner) how KrBot works, but I guessed from various clues that addi group of humans (Q16334295) to P31 could help. We'll have to wait for the next KrBot run and we will see if it helps. Alternatively, I've found a way to allow languages to be added only to humans and other related instances, for example (but it will be very complicated).--Frettie (talk) 12:34, 8 December 2023 (UTC)

I see your point but I think it would be best to implement an exception for your bot. --Emu (talk) 12:35, 8 December 2023 (UTC)

@Emu: Hi, have a nice weekend, it seems this edit is correct in principle. The edit war is no longer.--Frettie (talk) 08:27, 10 December 2023 (UTC)

It did not. I have blocked Frettiebot from editing this item. --Emu (talk) 21:21, 11 December 2023 (UTC)

Ok, KrBot is totally mystic. I'll try to simulate specific behavior for BC Schiller. --Frettie (talk) 15:05, 12 December 2023 (UTC)

I try to discuss it with KrBot owner. --Frettie (talk) 15:16, 12 December 2023 (UTC)

I have blocked Frettiebot from editing this item again. --Emu (talk) 14:14, 8 February 2024 (UTC)

Hi @Emu:, thanks. Thanks for reminding me of this situation, which I also need to resolve. Frettie (talk) 14:16, 8 February 2024 (UTC)

Bot edit war

Hello! My request is that you stop the bot so that it does not update the deleted occupation again. If something is deleted, there must be a reason, if the bot enters it again and again in the elements, it is only good for an editing war to develop, and when we set the data to be obsolete, we have to write in the justification that it is source known to be unreliable (Q22979588). Pallor (talk) 14:46, 13 October 2023 (UTC)

Deprecate it, not delete it. Its good way to solve it. —Frettie (talk) 14:48, 13 October 2023 (UTC)

If it is the best solution (which I and many others do not think, but it is still better than having a useless statement at normal rank) then program the bot to do it. You are using your bot to force people to do the work the way you like. It is disrespectful. GrandEscogriffe (talk) 16:31, 20 October 2023 (UTC)

Alias stejný jako štítek

Zdravím. Narazil jsem už na několikátý potok, kde jsi jako "alias" přidal název naprosto stejný s labelem. Např. tady. To je asi principiální chyba. ŠJů (talk) 19:09, 18 October 2023 (UTC)

Je to možný, až ty potoky dokončím, tak tohle můžu případně upravit. Ale v praxi to je úplně jedno. --Frettie (talk) 06:10, 19 October 2023 (UTC)

Úplně jedno to není, jestli si to tu zbytečně znepřehledňujeme například bezdůvodnými duplicitami. Zrovna tohle by asi měl hlídat a okamžitě odstraňovat automaticky nějaký celoprojektový skript, pokud možno by takovou duplicitu vůbec neměl ukládat. --ŠJů (talk) 01:45, 21 October 2023 (UTC)

Jo, bylo by to lepší, to máš pravdu. Určitě to můžeš někde navrhnout. Já ti nabízím, že to pak tady vyčistím, pokud to je problém. --Frettie (talk) 06:56, 21 October 2023 (UTC)

Čištění probíhá. --Frettie (talk) 17:58, 23 October 2023 (UTC)

Není Vídeň (Q94528) jako Vienna (Q1741)

Můžeš prosím zkontrolovat, jak program interpretuje "Vídeň" z autoritních dat? Je trochu zvláštní, že v jihomoravské vesnici se 464 obyvateli zemřelo tolik rakouských osobností - viz https://w.wiki/7qEY . Všiml jsem si toho po této editaci: https://www.wikidata.org/w/index.php?title=Q28076155&diff=1993080520&oldid=1993080515 . Díky! Sapfan (talk) 10:14, 19 October 2023 (UTC)

Ahoj, dobrej tip a dobrý oko. Koukal jsem na ten případ, co uvádíš a blbž je, že my to máme zřejmě správně. Už v NK v Alephu mají "Vídeň, Česko", to je pak blbý, když to má zdroj blbě. Když se podíváš přímo na záznam v NK ČR a zobrazíš si MARC zobrazení, tak je to tam jasně vidět. @Vojtěch Dostál tuším sem tam posílal nějaký hromadnější zprávy o chybách do NK ČR – Vojto, jaká je ideální cesta? Sepsat ty lidi, kde to je špatně nebo bude stačit poslat link na to query, kdy všichni jsou podezřelí? --Frettie (talk) 13:35, 19 October 2023 (UTC)

Hromadné věci asi můžete reportovat Petře Šťastné. Teď určitě informaci ponechat a zavrhnout rank :) Vojtěch Dostál (talk) 15:14, 19 October 2023 (UTC)

Díky oběma za odpověď. Místo úmrtí Vídeň (Q94528) jsem označil jako zavržené z důvodu error in referenced source or sources (Q29998666) všude kde jsem ho našel. S výjimkou několika, bohužel ne všech, zápisů s odkazem na Encyclopedia of Ostrava History (Q105721987) (příklad: Rudolf Prisching (Q95383590)), kde je ve zdroji jasně "Vídeň (Rakousko)" a chyba vznikla při opisu. Tím je snad záležitost vyřešena. --Sapfan (talk) 19:01, 19 October 2023 (UTC)

Confused Biała (Q855902) with Biała Głuchołaska (Q855904)

Ahoj. Vidím, že spoustu českých potoků, které ústí do Jesenické Bělé, jsi poslal do úplně jiné Bělé, která se nachází poměrně hluboko v polském vnitrozemí (nejblíže cca 20 km od českých hranic) a ještě jsi k této druhé Bělé přidal tvrzení, že se nachází také v Česku, a to vše s odvoláním na DIBAVOD. Opravdu mají v DIBAVOD takový bordel, anebo jsi to jen chybně přiřadil a netrklo tě, že u spousty potoků opravuješ správný údaj na chybný? ŠJů (talk) 18:35, 28 October 2023 (UTC)

Těch editací není tak málo, abych všechno zkontroloval, mám i svůj život. Pokud je tam něco špatně, tak to oprav, nebo mi řekni, abych to opravil. To tak prostě je. A jinak, v praxi to byla jedna chyba v bázi (ne v dibavod, předpokládám, že spíš v přiřazení QID – ručním) předpokládám, nekontroluju každou editaci, to opravdu nejde. Ještě tam pár věcí asi budu dohledávat (třeba jednotlivý úseky řek, který jsou v dibavod rozsekány na části, ale musím vymyslet jak na to). Klidně to můžeš vše projít, mně to vadit nebude. :) --Frettie (talk) 19:16, 28 October 2023 (UTC)

Odstraňovat správné souřadnice a nahrazovat je chybnými je tedy věc, která by se opravdu neměla stávat ani omylem. --ŠJů (talk) 18:47, 28 October 2023 (UTC)

To máš tak, všech těch 7 tisíc potoků jsem procházel ručně, je možný, že sem tam jsem si spletl potok. To klidně oprav, nemám s tím problém. --Frettie (talk) 19:16, 28 October 2023 (UTC)

Nesprávné přítoky z Biała (Q855902) jsem odstranil (do Biała Głuchołaska (Q855904) jsem je nepřidal). U jednotlivých potoků jsem ručně opravil údaj, kam ústí (bez reference, protože DIBAVOD nemá přímé odkazy do on-line databáze, takže nelze ověřit, co skutečně obsahuje). V položce Biała (Q855902) jsem odstranil chybné ID DIBAVOD a chybný údaj o státu a okresu (do Biała Głuchołaska (Q855904) jsem je nepřidal).

S drobnými chybami je nutné počítat, ale taky je třeba sledovat všechny signály, které by měly před takovou chybou varovat. Jen tak odstranit správné souřadnice a nahradit je chybnými bez ruční kontroly? To fakt ne. A jakmile se bot pokouší přidat jakýkoliv údaj, které je v rozporu s nějakým údajem již v položce uvedeným, okamžitě by se měla rozsvítit pomyslná červená kontrolka. A u 55kilometrové řeky by se takové omyly fakt stávat neměly. Jo chápu, stane se i něco, co se stávat nemá, ale pak je teda potřeba se pořádně chytit za hlavu a poučit se z toho, na co si dávat pozor. (A ovšem samozřejmě je třeba pečlivost i v maličkostech a nemnožit ani úplně zbytečné drobné chyby, jako Tributary s bezdůvodně velkým "T", aliasy duplikující štítek atd.). --ŠJů (talk) 19:32, 28 October 2023 (UTC)

Jinak samozřejmě dík za tu práci, docela se to hodí mít ty potoky naimportované a doteď toho většina, na co jsem narazil, byla správně. Jen tam občas jsou nějaké záhadnosti, že DIBAVOD nějaký potok rozdělí na dva úseky se zcela stejným názvem, nebo má nějaké bezvýznamné rameno pod stejným názvem jako celý potok, a to pak pochopitelně může haprovat, když se ty údaje přidají do již existující položky, která původně byla určena pro celý potok. --ŠJů (talk) 19:41, 28 October 2023 (UTC)

Jasný, to Tributary jsem psal ručně, až pozdě mi došlo, že to je hloupost. Chyby tam budou. Ono ten Dibavod (resp. ten ústav) tam má místama opravdu divný věci, navíc jim poměrně dost potoků/toků chybí (proč, nemám tušení, jaký je klíč, nemám tušení), pak ještě existuje pár další evidencí, které nejsou mezi sebou kompatibilní a tak se třeba u většiny toků nedá přiřadit správný správce. Důvodem je to, že každý ministerstvo používá vlastní evidenci a teď to ručně lepí dohromady a sami to vidí na mnoho let dopředu. No, mohli si říct, že to není třeba dělat dvakrát ... A takovejch perliček tady máme dost. --Frettie (talk) 20:19, 28 October 2023 (UTC)

Už dřív jsem si všiml, že u rybníků měl DIBAVOD často úplně chybné přiřazení k obcím. Spíš zpravidla než výjimečně.

Teď jsem narazil ještě na jeden problém u jednoho z Račích potoků. U přeshraničních toků tam jako "ústí" přebíráš souřadnice, kde potok opouští Českou republiku, a tím pádem nesedí ani celková délka toku. (Krom toho jsem taky zmatkoval kvůli tomu, že jedno rameno Račího potoka je uvedeno jako další potok se stejným názvem, svoje chybné opravy už jsem revertoval.) --ŠJů (talk) 21:02, 28 October 2023 (UTC)--ŠJů (talk) 21:02, 28 October 2023 (UTC)

Bjarne Stroustrup

About Bjarne Stroustrup (Q92620), I don't think that Bjarne's field of work is informatics; it would be rather computer science. The bot says "Czech-Authorities-Sync", but NL CR AUT[4] says "počítačová věta" and "počítačový vědec", which does not lead to "informatics" but rather to "computer science". Thus, I am not sure what is going on here. --Dan Polansky (talk) 11:40, 11 November 2023 (UTC)

Hi, i tride to correct them, but at moment the bot don't work. I correct Bjarne Stroustrup (Q92620) by "hand". And i corrected connection of NK ČR "počítačová věda" to WD computer science (Q21198). --Frettie (talk) 11:48, 11 November 2023 (UTC)

Conflating articles with people

I reverted your changes to Oskar Baudisch (Q59509334) and Constructing Alien Space in South Slavic Oral Lyric (Q114907080) because you apparently conflated a scholarly article with a person. Bovlb (talk) 19:25, 18 January 2024 (UTC)

Also Professor David Poswillo CBE (1927-2003): Skilled oral and maxillofacial surgeon, influential scientist, teacher and adviser. (Q47904911), where you appear to have been edit warring. Bovlb (talk) 19:34, 18 January 2024 (UTC)

...and Capitalizing on Language Learners’ Individuality: From Premise to Practice Tammy Gregersen and Peter D. MacIntyre. Bristol, England: Multilingual Matters, 2014. Pp. xxvii + 259 (Q58375983), Russell, Henry Norris (1877-1957), astronomer (Q103878043), and Astaire, Adele (1898-1981), musical theater dancer (Q103845366) which I fixed but the bot rushed to add the identifiers back immediately (together with Professor David Poswillo CBE (1927-2003): Skilled oral and maxillofacial surgeon, influential scientist, teacher and adviser. (Q47904911) again)! Stop that, please! --Mormegil (talk) 06:18, 19 January 2024 (UTC)

Hi @Mormegil, Bovlb: I'm stopping the bot, this needs to be resolved, it didn't happen that often before i think, now it looks like a possible change in methodology in the NKP ČR. Thanks! --Frettie (talk) 08:00, 19 January 2024 (UTC)

Bovlb, after your revert, conflation of human with article was done again, adding ISNI: https://www.wikidata.org/w/index.php?title=Q59509334&action=history CV213 (talk) 16:22, 26 January 2024 (UTC)

@Bovlb: This is an edit from before the script was edited, i.e. January 19, afterwards this behavior was taken care of.-Frettie (talk) 09:41, 27 January 2024 (UTC)

Czech-Authorities-Sync - please stop re-adding false claims after being reverted

Czech-Authorities-Sync is faulty. You create unique value violations

1. https://www.wikidata.org/w/index.php?title=Q29865014&diff=prev&oldid=2062426365
2. https://www.wikidata.org/w/index.php?title=Q5531495&diff=prev&oldid=2062421116

Please, never add ISNI via bot if it is already in Wikidata on another item.

@Mormegil, Bovlb, Emu, Epìdosis: fyi. CV213 (talk) 16:14, 26 January 2024 (UTC)

@CV213: It's better to set the mistakes (which is from external database) as deprecated. I set it up for those two cases. --Frettie (talk) 16:38, 26 January 2024 (UTC)

No it isn't. It pollutes the Wikidata database. CV213 (talk) 16:43, 26 January 2024 (UTC)

Please don't put false claims on items and then just deprecate them. Fix your tools or stop using the source or fix the source. CV213 (talk) 16:45, 26 January 2024 (UTC)

No, but this discussion is already resolved. --Frettie (talk) 16:54, 26 January 2024 (UTC)

Not sure what you mean by no. User:Pallor, what was the result of the discussion Frettie pointed to? CV213 (talk) 17:02, 26 January 2024 (UTC)

CV213: The previous discussion was not about unique identifiers, but about objections related to other data (occupation, place of birth/death, etc.).

Frettie was not a factor in the discussion, Vojtěch Dostál, on the other hand, correctly agreed to create a system that would filter the entry of unnecessary/incorrect data. Finally, however, Jahl de Vautban opened a new topic here. , but it seems that the ideas are not coming Pallor (talk) 00:10, 27 January 2024 (UTC)

User:Pallor, thank you. On external Czech-Authorities-Items I saw a CZ-ID, a WD-ID, and an ISNI. So, maybe the matching is done on WD-ID, ISNI or both. If there is one mistake in the matching, all other claims on the external item can end up at the WD-item. Elephants can be presidents of the United States, Cars can have an occupation. The toolchain used by Frettie and Vojtěch Dostál doesn't seem to take into account the type of items. And it doesn't seem to look if a given ISNI is already contained in WD, which is a hint towards a potential existence of some inconsistency either in WD, ISNI, or the CZ-Dataset. CV213 (talk) 08:29, 27 January 2024 (UTC)

"No" is for "pollution of database". --Frettie (talk) 17:26, 26 January 2024 (UTC)

Another item with two false IDs, removed, re-inserted, now removed again https://www.wikidata.org/w/index.php?title=Q25442482&action=history CV213 (talk) 17:23, 26 January 2024 (UTC)

I revereted your edit and deprecate this. It's only solution to correct data and not reinserted data. --Frettie (talk) 17:26, 26 January 2024 (UTC)

Despite you claiming so, it is not true. CV213 (talk) 17:53, 27 January 2024 (UTC)

Mass creation of duplicates

Why create https://www.wikidata.org/w/index.php?title=Q122873108&action=history if the ISNI is already on Q109855888? ISNI is *the* international standard identifier exactly meant to identify identities. Between a quarter of a half of the duplicates I fixed from the list at Property talk:P213/Duplicates/humans today originate from Frettie or Vojtech Dostal. CV213 (talk) 16:29, 26 January 2024 (UTC)

Ok, thanks for your work, I appreciate it. --Frettie (talk) 16:40, 26 January 2024 (UTC)

Property talk:P213/Duplicates/humans

Property talk:P213/Duplicates/humans - I give up on cleaning this for now. @Mormegil, Bovlb, Epìdosis, Pallor: it is no fun to see that what I fix now had been fixed before but was wrongly readded by a bot. Humans spent time on the items, only to be reverted. And no, deprecating isn't a solution. If the CZ-database has only one claim wrong, than from that set all other claims may be imported to the wrong item. No, thank you. And not to speak of Wikidata:Database reports/Constraint violations/P213 which could be very usefull, but not when false claims stay in deprecated mode. I agree that there are cases where deprecation is useful, but not, when any random source contains one ID mistake in an item resulting in an unlimited quantity of false claims ending up in the wrong Wikidata item - forever. CV213 (talk) 17:31, 26 January 2024 (UTC)

This has nothing to do with Frettiebot or me. Usage of deprecation is the standard Wikidata policy and was recently confirmed by a community-wide discussion. Please accept that, or challenge that rule elsewhere Vojtěch Dostál (talk) 20:49, 26 January 2024 (UTC)

Of course it has to do with you both. You are inserting CZ-Data based on wrong matches and you create duplicates based on CZ-Data which could have been avoided by checking if a specific ISNI already exists in WD. CV213 (talk) 08:36, 27 January 2024 (UTC)

Well, we thought about matching based on ORCID and ISNI present in NKCR, but the disadvantage of this approach is that these identifiers may be added incorrectly to NKCR entries. I estimate there's a 0.1-0.5% error rate in the librarian matching, which could lead to significant conflation in Wikidata. Therefore, so far we reasoned that the one additional control step (merging by Wikidata users) is preferable to automated editing. Vojtěch Dostál (talk) 10:51, 27 January 2024 (UTC)

@Epìdosis What do you think about automated matching to QIDs based on ISNI or ORCID present in NKCR database? Should this be automated or is the potential disruption not worth it? Vojtěch Dostál (talk) 10:52, 27 January 2024 (UTC)

Does NKCR contain type information (P31=human/organization[in VIAF=Corporate])? CV213 (talk) 10:58, 27 January 2024 (UTC)

Well, there is "JA" in the FMT field of all personal authorities (people, families, pseudonyms, characters). Unfortunately, this piece of information is not available in the open dataset AFAIK. Vojtěch Dostál (talk) 12:14, 27 January 2024 (UTC)

AFAIK ISNI is only defined for what in VIAF is called Personal (~NKCR "JA") or Corporate. Two situations for type errors:

1. ISNI exists on NKCR item that is not Personal / Corporate
2. NKCR item has WD ID for a WD item that is not Personal / Corporate

Can you do checks to avoid importing data that causes errors in WD related to the type of an item? If you find such errors, maybe report back to NKCR?

At https://www.nkp.cz/o-knihovne/odborne-cinnosti/otevrena-data there is:

1. Personální autority https://aleph.nkp.cz/data/aut_ja.xml.gz
   1. do you know how many items are contained?
   2. how many of these are in WD already?
   3. how many of these have an ISNI?
   4. if a JA-item has an ISNI and the ISNI is not yet in WD, it would be helpful if an import could be made. Maybe duplicates are created, but one can later merge based on ISNI. ISNI is the international standard ID for such items, so it is maybe the most effective to avoid duplicates or to merge them.
2. Korporativní autority https://aleph.nkp.cz/data/aut_ka.xml.gz

Only items from Personální and Korporativní should probably have an ISNI. The most used type in WD for ISNI is human which is a subclass of Personal. CV213 (talk) 12:54, 27 January 2024 (UTC)

As I explained, the respective field is not available so we likely cannot make that check. But we do regularly report back to the library for errors - most recently 100 errors in the QID subfield. For your other questions (statistics), I don't have those at hand but these are publicly available data so you can do the research yourself. Vojtěch Dostál (talk) 15:44, 27 January 2024 (UTC)

I demonstrated it is available. https://aleph.nkp.cz/data/aut_ja.xml.gz is the file that containts the items that contain "JA" items. CV213 (talk) 15:47, 27 January 2024 (UTC)

We will give it a thought but parsing one huge file (with all the entries) is complicated enough, without having to parse another huge file. Vojtěch Dostál (talk) 15:50, 27 January 2024 (UTC)

@Vojtěch Dostál: I think automating the matching to QIDs based on ISNI or ORCID present in NKCR database is probably worth (BTW, I thought it was already being done); for both these cases I unfortunately know well that they usually have some mistakes (I'm thinking of ORCID-match for GND and SUDOC and of ISNI-match for SBN), but they are a small minority that can be probably be managed manually afterwards. Of the two I empirically hypothetise that ORCID-match should have fewer mistakes than ISNI-match, so I would suggest to start with ORCID-match, have a look at the constraint violations and then, if the number of spotted mistakes is very low, start also ISNI-match. --Epìdosis 13:05, 27 January 2024 (UTC)

Hi, QID assignment based on ISNI and ORCID in NK ČR authorities is done in tables where these occurrences remain and are ready for manual control and manual creation (with QuickStatements). This is mostly done by me sometimes. The value of automatic matching would be relatively small of my opinion, with most a few tens of records each week. --Frettie (talk) 13:16, 27 January 2024 (UTC)

"start with ORCID-match" - that's calling for problems, if ISNI isn't looked at at the same time. Best is to start with ORCID and ISNI, add items that have both and where ORCID and ISNI are not yet in WD. CV213 (talk) 13:29, 27 January 2024 (UTC)

Block

You have been blocked indefinitely for continued making of indefensible changes like Special:Diff/2061318897. If you believe this block is unjustified, you may contest it by editing this page and adding the following template with a suitable reason: {{unblock|1=<the reason for your unblock request>}}. If you are logged in, and the option has not been disabled, you may also email the blocking administrator (or any administrator from this list) by using this form. See Wikidata:Guide to appealing blocks for more information.

---

⁠

The bot is continuing to make edits that defy any justification despite vocal feedback, so I'm giving it an enforced pause until we can sort things out. Bovlb (talk) 18:01, 26 January 2024 (UTC)

Ok. I'll check to see why this happened despite the mub20231194521 record being depreked. It was mistake. What a coincidence that you just noticed this while we discussing a similar topic with another user. --Frettie (talk) 18:10, 26 January 2024 (UTC)

Hmm. You say coincidence; I see a pattern. See also Wikidata:Administrators'_noticeboard#Block_of_Frettiebot. Bovlb (talk) 18:16, 26 January 2024 (UTC)

I fixed this issue, i added a check if the actual NK ČR field is deprecated. If it is, bot will not add the new claim. I tried it on the Lorraine (Q1137) and the ISNI was not added. Commit is here. Bot is stopped. --Frettie (talk) 19:22, 26 January 2024 (UTC)

What do you mean by "the new claim"? It shouldn't add any claim to that item coming from the deprecated external item. CV213 (talk) 08:33, 27 January 2024 (UTC)

Potential new claim. --Frettie (talk) 09:02, 27 January 2024 (UTC)

User:Bovlb Now his friend re-inserts false ISNI claims, based on bugs in their favorite source dataset, that then re-populate constraint violations reports. [5] CV213 (talk) 14:04, 27 January 2024 (UTC)

comedy

You applied the wrong Q-value in this edit, which I have corrected. Terence wrote comedy plays, not general humor like jokes and funny faces. --EncycloPetey (talk) 00:58, 8 February 2024 (UTC)

@EncycloPetey: Hi, i fix it in correct way. Thanks! --Frettie (talk) 07:33, 8 February 2024 (UTC)

ummm, how is that "correct"? The problem was that you added the wrong "comedy". The database does say "komedie", but you selected the wrong wikidata label. --EncycloPetey (talk) 15:37, 8 February 2024 (UTC)

Database's "comedy" was deprecated now, your new "comedy" is set. But I recognize that here it's really better to change the target of the comedy and set it up properly with a link to the NK ČR. So it's fixed and you are right, your "comedy" is better of komedie. Thanks! --Frettie (talk) 16:38, 8 February 2024 (UTC)

štítek

ahoj, nešlo by, aby v případě, že daná osoba nemá český štítek a přidáváš NKČR [6], aby bot doplnil i český štítek? JAn Dudík (talk) 12:04, 27 February 2024 (UTC)

Ahoj,

popřemýšlet o tom můžu, taky jsem na to narazil. Dobrý nápad. --Frettie (talk) 10:31, 1 March 2024 (UTC)

@JAn Dudík: Tak by to mělo být, dneska by se to mělo snad importovat, bude toho asi víc. --Frettie (talk) 09:11, 2 March 2024 (UTC)

Place of birth of Kinga Gál (Q721647)

Hello, your bot keeps re-adding Hungary as place of birth (P19) to Kinga Gál (Q721647), even though she already has a more precise place of birth. I'm currently using the lexeme for a project and this keeps breaking it. I would be grateful if you could stop your bot from doing this. Thank you! Zarasophos (talk) 10:29, 1 March 2024 (UTC)

Hi, correct way is to deprecate non valid value (look at Kinga Gál now). Do you have problem with this way in your project? --Frettie (talk) 10:33, 1 March 2024 (UTC)

Edit warring

I have twice reverted the bot on William M. Adams (Q2579787) as it is trying to add an incorrect identifier — Martin (MSGJ · talk) 08:17, 18 March 2024 (UTC)

William M. Adams (Q2579787) Hi, thanks, i deprecated bad ISNI, because this is correct way. And i contacted NK ČR, because there is mistake and its good to correct this. --Frettie (talk) 08:58, 18 March 2024 (UTC)

Edit warring - Undomesticated Ground: Recasting Nature as Feminist Space

https://www.wikidata.org/w/index.php?title=Q96624830&action=history - please don't add this data ever again. The item is about a literary work. The data you added belonged to a human (the author).

Please perform at least a simple type check before you edit! APW1 (talk) 14:28, 31 March 2024 (UTC)

@APW1: Hi, thanks, i deprecated NKČR property, because this is correct way (if bot doesnt have information, that is bad information, there is no information). And i contacted NK ČR, because there is mistake in their DB and its good to correct this. Thanks again. --Frettie (talk) 22:04, 31 March 2024 (UTC)

Thank you for the explanation! APW1 (talk) 06:42, 1 April 2024 (UTC)

My pleasure! --Frettie (talk) 06:43, 1 April 2024 (UTC)