User talk:Reinheitsgebot: Difference between revisions
→Dubletten: Reply |
→Error detection: reply (CD) |
||
Line 1,199: | Line 1,199: | ||
Reinheitsgebot added in incorrect VIAF identifiers. See for example: {{Q|Q26265809}}. See: [[Wikidata:Database reports/items with P569 greater than P570]] for one of the error detection searches I monitor. I just corrected eight errors. The problem cascaded as a User added in incorrect birth and death dates based on your incorrect VIAF entries. [[User:Richard Arthur Norton (1958- )|RAN]] ([[User talk:Richard Arthur Norton (1958- )|<span class="signature-talk">{{int:Talkpagelinktext}}</span>]]) 16:07, 14 June 2023 (UTC) |
Reinheitsgebot added in incorrect VIAF identifiers. See for example: {{Q|Q26265809}}. See: [[Wikidata:Database reports/items with P569 greater than P570]] for one of the error detection searches I monitor. I just corrected eight errors. The problem cascaded as a User added in incorrect birth and death dates based on your incorrect VIAF entries. [[User:Richard Arthur Norton (1958- )|RAN]] ([[User talk:Richard Arthur Norton (1958- )|<span class="signature-talk">{{int:Talkpagelinktext}}</span>]]) 16:07, 14 June 2023 (UTC) |
||
: It's quite interesting, that no one has deleted wrong matches: https://mix-n-match.toolforge.org/#/entry/98693218 https://mix-n-match.toolforge.org/#/entry/63534239 https://mix-n-match.toolforge.org/#/entry/53265261 https://mix-n-match.toolforge.org/#/entry/111294683 https://mix-n-match.toolforge.org/#/entry/141850181 https://mix-n-match.toolforge.org/#/entry/5746723 [[User:Matlin|Matlin]] ([[User talk:Matlin|<span class="signature-talk">{{int:Talkpagelinktext}}</span>]]) 18:28, 17 June 2023 (UTC) |
|||
== Dubletten == |
== Dubletten == |
Revision as of 18:28, 17 June 2023
This is a bot run by Magnus Manske.
Sourcing
@Magnus Manske: Please stop mass-adding unreliable sources to statements that already have better ones. Nikkimaria (talk) 14:35, 9 January 2018 (UTC)
- Please be more specific, and please talk to me on my talk page. --Magnus Manske (talk) 15:20, 9 January 2018 (UTC)
- @Magnus Manske: I'm referring to the contributions of this bot, which should be stopped. Nikkimaria (talk) 15:42, 9 January 2018 (UTC)
- Apologies, I must have been unclear. When I say on my talk page, I mean on my talk page. And when I say "more specific", I don't mean "state the obvious". This bot does several things. --Magnus Manske (talk) 15:46, 9 January 2018 (UTC)
- This bot's last several hundred contributions have all been edits like this one. Nikkimaria (talk) 15:57, 9 January 2018 (UTC)
- OK, I'll be friendly, even though you ignored my request. I see the edit. May I take it that your issue with it is the use of Find-a-grave as a source? Find a Grave (Q63056) does not have a talk page, and Property talk:P535 does not say anything about it not to be used as a reference. For the example you gave, the entry does have the birth date that that it was used as a reference for. It even has a picture of the grave (albeit only with the birth and death years, not days). So, what exactly is the issue? --Magnus Manske (talk) 16:09, 9 January 2018 (UTC)
- The statement is already reliably sourced; there is no value in adding an unreliable source. Nikkimaria (talk) 16:15, 9 January 2018 (UTC)
- Please don't lie, or at least, be less blatant about it. The statement you linked to has, as its other source, "imported from English Wikipedia". That does not qualify as an actual source, it is merely a tracker to see when it was copied from. It had no source; now it has one. And you still have not established that Find-a-grave should not be used as a reference. --Magnus Manske (talk) 16:23, 9 January 2018 (UTC)
- You are correct that I was mistaken about that particular diff - most but not all of the bot's additions are in cases where there are already sources. So I will use the bot's edits as a tracker for statements requiring reliable referencing, and I would ask that you consider limiting the bot's edits to cases where there are not already multiple reliable references provided. Nikkimaria (talk) 17:03, 9 January 2018 (UTC)
- ...which apparently won't work because the bot will just repeat its edits, so please also limit the bot to only doing one pass per item. Nikkimaria (talk) 19:12, 9 January 2018 (UTC)
- Please don't lie, or at least, be less blatant about it. The statement you linked to has, as its other source, "imported from English Wikipedia". That does not qualify as an actual source, it is merely a tracker to see when it was copied from. It had no source; now it has one. And you still have not established that Find-a-grave should not be used as a reference. --Magnus Manske (talk) 16:23, 9 January 2018 (UTC)
- The statement is already reliably sourced; there is no value in adding an unreliable source. Nikkimaria (talk) 16:15, 9 January 2018 (UTC)
- OK, I'll be friendly, even though you ignored my request. I see the edit. May I take it that your issue with it is the use of Find-a-grave as a source? Find a Grave (Q63056) does not have a talk page, and Property talk:P535 does not say anything about it not to be used as a reference. For the example you gave, the entry does have the birth date that that it was used as a reference for. It even has a picture of the grave (albeit only with the birth and death years, not days). So, what exactly is the issue? --Magnus Manske (talk) 16:09, 9 January 2018 (UTC)
- This bot's last several hundred contributions have all been edits like this one. Nikkimaria (talk) 15:57, 9 January 2018 (UTC)
- Apologies, I must have been unclear. When I say on my talk page, I mean on my talk page. And when I say "more specific", I don't mean "state the obvious". This bot does several things. --Magnus Manske (talk) 15:46, 9 January 2018 (UTC)
- @Magnus Manske: I'm referring to the contributions of this bot, which should be stopped. Nikkimaria (talk) 15:42, 9 January 2018 (UTC)
- I think there was an rfc about this source where people found adding it was ok.
--- Jura 19:20, 9 January 2018 (UTC)- Replacing it with better sources is not something the bot should work against. Nikkimaria (talk) 19:25, 9 January 2018 (UTC)
- I'm not aware of a process called "replacing references" at Wikidata. Contributors add references and rank statements. If there is a mere Wikipedia import source you might consider that, but in general, content isn't deleted, but ranked.
--- Jura 19:29, 9 January 2018 (UTC)- I'm not here objecting to the bot adding this source generally, but spamming it provides no benefit where there are already multiple reliable sources provided for a statement. Nikkimaria (talk) 19:31, 9 January 2018 (UTC)
- You can filter out these references in your application if you don't deem them suitable for your audience.
--- Jura 19:35, 9 January 2018 (UTC)- Since I'm not creating an application this would not address my concern. Nikkimaria (talk) 19:54, 9 January 2018 (UTC)
- Nikkimaria, please define "reliable sources". Is there consensus that Find-a-grave is not suitable? If so, please point me to it, and I can deactivate it as a source. And define "spamming". Would you still call it "spamming" if the source were more to your liking? If yes, you'd basically forbid all bot-adding of sources; if not, you're a hypocrite. --Magnus Manske (talk) 23:05, 9 January 2018 (UTC)
- That's not accurate. I would not refer to all bot-adding of any source as spamming. However, repeatedly adding low-quality sources to statements that are already well-supported by higher-quality ones is IMO spamming, of no benefit. Nikkimaria (talk) 23:53, 9 January 2018 (UTC)
- I've already asked you not to refer to good-faith edits as spamming; now I'm telling you in clear terms for the avoidance of doubt: it is completely unacceptable to do so; your "opinion" on the matter notwithstanding. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:54, 10 January 2018 (UTC)
- To be clear: I have no doubt that the bot op is acting in good faith, and am not interested in seeing him or the bot blocked - it simply needs some adjustment to avoid adding low-value links to well-sourced statements, and (most importantly) repeating edits. Nikkimaria (talk) 13:42, 10 January 2018 (UTC)
- I've already asked you not to refer to good-faith edits as spamming; now I'm telling you in clear terms for the avoidance of doubt: it is completely unacceptable to do so; your "opinion" on the matter notwithstanding. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:54, 10 January 2018 (UTC)
- That's not accurate. I would not refer to all bot-adding of any source as spamming. However, repeatedly adding low-quality sources to statements that are already well-supported by higher-quality ones is IMO spamming, of no benefit. Nikkimaria (talk) 23:53, 9 January 2018 (UTC)
- Nikkimaria, please define "reliable sources". Is there consensus that Find-a-grave is not suitable? If so, please point me to it, and I can deactivate it as a source. And define "spamming". Would you still call it "spamming" if the source were more to your liking? If yes, you'd basically forbid all bot-adding of sources; if not, you're a hypocrite. --Magnus Manske (talk) 23:05, 9 January 2018 (UTC)
- Since I'm not creating an application this would not address my concern. Nikkimaria (talk) 19:54, 9 January 2018 (UTC)
- Please don't make false accusation such as describing a fellow editor's good faith contributions as "Spamming". Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 23:39, 9 January 2018 (UTC)
- You can filter out these references in your application if you don't deem them suitable for your audience.
- I'm not here objecting to the bot adding this source generally, but spamming it provides no benefit where there are already multiple reliable sources provided for a statement. Nikkimaria (talk) 19:31, 9 January 2018 (UTC)
- I'm not aware of a process called "replacing references" at Wikidata. Contributors add references and rank statements. If there is a mere Wikipedia import source you might consider that, but in general, content isn't deleted, but ranked.
- Replacing it with better sources is not something the bot should work against. Nikkimaria (talk) 19:25, 9 January 2018 (UTC)
- The RfC is at Wikidata:Requests for comment/Findagrave removed as a source for information.
--- Jura 07:27, 10 January 2018 (UTC)
Nikkimaria, to bring this to a close:
- you have failed to show that Findagrave is a "low quality source" that should not be used
- you have also failed to show that at the RfC
- you have even failed to show a single example for your claim where an additional Findagrave reference makes things worse somehow
I know what you want, you keep repeating it ad nauseam. But stomping your foot and saying "But I wanna!" is not constructive. --Magnus Manske (talk) 14:27, 10 January 2018 (UTC)
- @Magnus Manske: Will you at least ensure that your bot does not repeat edits? Given that we will need to agree to disagree on the other points, this would allow for a productive route forward. Nikkimaria (talk) 16:17, 10 January 2018 (UTC)
- For other contributors to be able to review, it's preferable that you report bot errors here rather than delete any statemements.
--- Jura 17:09, 10 January 2018 (UTC)- Don't expect so, no. However, I recognize that my feelings about the issues at play here led to me not expressing myself as well as I might have done, and I will apologize to Magnus for that and for any offense I have caused him as a result of that. I felt a quick intervention was needed, and as a result was more brusque in my approach than warranted. I will reiterate that I believe he is acting in good faith, although I disagree with the approach taken by the bot. Nikkimaria (talk) 03:56, 11 January 2018 (UTC)
- Then let's agree to disagree. I have no special attachment to Findagrave, but I fail to see the harm, and I do see the benefit. Could you point me to an instance where the same edit was repeated? That should not have happened, and an example will help to diagnose a potential issue. --Magnus Manske (talk) 09:30, 11 January 2018 (UTC)
- [1]. Nikkimaria (talk) 13:35, 11 January 2018 (UTC)
- I believe I found the issue. It should be fixed for future edits (as in, the bot will remember doing the edit), but it doesn't remember some previous ones, and it's hard to find them (as in, adding a specific source to a specific statement, where the source has been removed now) in the Wikidata item history. So, it might add it again in some instances, but no more than once. --Magnus Manske (talk) 13:53, 11 January 2018 (UTC)
- Thanks, appreciate that. Nikkimaria (talk) 15:00, 11 January 2018 (UTC)
- I believe I found the issue. It should be fixed for future edits (as in, the bot will remember doing the edit), but it doesn't remember some previous ones, and it's hard to find them (as in, adding a specific source to a specific statement, where the source has been removed now) in the Wikidata item history. So, it might add it again in some instances, but no more than once. --Magnus Manske (talk) 13:53, 11 January 2018 (UTC)
- [1]. Nikkimaria (talk) 13:35, 11 January 2018 (UTC)
- Then let's agree to disagree. I have no special attachment to Findagrave, but I fail to see the harm, and I do see the benefit. Could you point me to an instance where the same edit was repeated? That should not have happened, and an example will help to diagnose a potential issue. --Magnus Manske (talk) 09:30, 11 January 2018 (UTC)
- Don't expect so, no. However, I recognize that my feelings about the issues at play here led to me not expressing myself as well as I might have done, and I will apologize to Magnus for that and for any offense I have caused him as a result of that. I felt a quick intervention was needed, and as a result was more brusque in my approach than warranted. I will reiterate that I believe he is acting in good faith, although I disagree with the approach taken by the bot. Nikkimaria (talk) 03:56, 11 January 2018 (UTC)
- For other contributors to be able to review, it's preferable that you report bot errors here rather than delete any statemements.
Clearly false edit
Hi Magnus. This edit [2] was clearly wrong. At this point, English Wikipedia did not list the guy as dead, and he only died a week ago. Could you please have a look and see what happened. If this is a major problem, may be many edits around the time must be reverted.--Ymblanter (talk) 08:29, 16 January 2018 (UTC)
- I do not recall what, exactly, the bot did four years ago. It was most likely some one-off code. --Magnus Manske (talk) 08:37, 16 January 2018 (UTC)
- Persondata had DATE OF DEATH = 27 august 2011.
--- Jura 08:39, 16 January 2018 (UTC)- I see, than this must be on the Wikipedia side.--Ymblanter (talk) 09:05, 16 January 2018 (UTC)
- Persondata had DATE OF DEATH = 27 august 2011.
Duplicates
Hi Magnus, I'm writing you regarding a Wikidata entry for Samuel Blatchley Webb created by your hard working bot way back in March 2015, i.e., Q19423927. Actually, the person described there is Samuel Blatchley Webb's son, James Watson Webb who has his Wikidata entry at Q3161477. Also, the correct Wikidata entry for Samuel Blatchley Webb is Q48257893. Additionally, please note that The Appletons' Cyclopædia of American Biography does not have a separate entry for James Watson Webb, only for his father: [3]. However, James described quite vividly there: [4], which probably confused your bot. Best regards to both, --Taterian (talk) 07:23, 15 February 2018 (UTC) P.S. Done --Taterian (talk) 07:21, 20 February 2018 (UTC)
Revert
Hi, FYI I have reverted your bot: [5]. — Ayack (talk) 12:43, 19 February 2018 (UTC)
Corporate authors from OCLC_Authors are not human!
Hi - you created this entry and several others like it from the Mix n Match OCLC_Authors list, but these are corporate "authors", not humans. KRbot seems to be able to tell that the VIAF entry is corporate and not human, so you might want to check that before doing more like this (if these are still happening). ArthurPSmith (talk) 21:18, 26 February 2018 (UTC)
Year parsing
Deleted this.
--- Jura 13:36, 5 March 2018 (UTC)
Repeated bad edits
Magnus, look at the history of
- https://www.wikidata.org/w/index.php?title=Q49477074&action=history
- https://www.wikidata.org/w/index.php?title=Q49473434&action=history
Reinheitsgebot adds statements like Commons category (P373)}} XXXXX HIER FEHLT ALLES!-->, volunteers remove them, and bots add them over and over again. How do we break the cycle? --Jarekt (talk) 16:03, 21 March 2018 (UTC)
- @Jarekt: Probably by blocking the bot? Nothing else seems to help. —Tacsipacsi (talk) 19:16, 25 March 2018 (UTC)
- I have turned off that function of the bot. @Jarekt:: See here. --Magnus Manske (talk) 09:11, 26 March 2018 (UTC)
- Thank you Magnus, and no blame. I know how much time I spend in Wikiverse on the few things I am involved in and you seem involved in much more tool development and maintenance. But there should be some mechanism for you to be able to pass some of the tasks to others. Maybe WMF should get you an intern assistant, whose job would be to monitor communication channels and run daily chores.
--Jarekt (talk) 12:04, 26 March 2018 (UTC)
- Thanks! If you would link to the wiki page from where the bot works (it’s obvious from the erroneous data that the source is some wikitext), we could probably fix it – or just remove the bad entries –, and backing everything with sources is an important goal on Wikidata (although Commons links don’t require sources most of the time). Anyways, thanks for your many tools! —Tacsipacsi (talk) 22:31, 26 March 2018 (UTC)
- Thank you Magnus, and no blame. I know how much time I spend in Wikiverse on the few things I am involved in and you seem involved in much more tool development and maintenance. But there should be some mechanism for you to be able to pass some of the tasks to others. Maybe WMF should get you an intern assistant, whose job would be to monitor communication channels and run daily chores.
- I have turned off that function of the bot. @Jarekt:: See here. --Magnus Manske (talk) 09:11, 26 March 2018 (UTC)
THANK YOU <3
Thank you very very much for your action via LfDS object ID (P1708). That is much structured fun for many people :) . Regards, Conny (talk) 15:12, 30 April 2018 (UTC).
- My pleasure! Don't forget to check for images! :-) --Magnus Manske (talk) 20:55, 30 April 2018 (UTC)
Peter Marcus vs. Philipp Otto Runge
Hallo Magnus, hier scheint was schiefgelaufen zu sein: Peter Marcus (Q52150818) mit den Normdaten von Philipp Otto Runge (Q52150818). Kannst du mal drüberschauen? Gruß --Kolja21 (talk) 00:34, 14 May 2018 (UTC)
- Habe mal aufgeräumt, aber da scheint es ein Problem in VIAF zu geben :-( --Magnus Manske (talk) 07:30, 14 May 2018 (UTC)
- Danke für die schnelle Rückmeldung. VIAF vermischt häufig Personen, aber in dem betreffenden Datensatz scheint alles in Ordnung zu sein. "Runge, Philipp Otto, 1777-1810" enhält alle drei Suchbegriffe: PETER, MARCUS und 1889.
- Person als Relation: Betthausen, Peter ‡d 1941-...
- Person als Relation: Behmer, Marcus ‡d 1879-1958
- Person als Relation: Marcks, Gerhard ‡d 1889-1981
- Fröhliches Schaffen --Kolja21 (talk) 23:02, 14 May 2018 (UTC)
- Argh, danke. Habe das VIAF-Suche-Automatching mal rausgenommen... --Magnus Manske (talk) 08:51, 15 May 2018 (UTC)
- Danke für die schnelle Rückmeldung. VIAF vermischt häufig Personen, aber in dem betreffenden Datensatz scheint alles in Ordnung zu sein. "Runge, Philipp Otto, 1777-1810" enhält alle drei Suchbegriffe: PETER, MARCUS und 1889.
Dublette
Der neue Eintrag Jean Baptiste Boisduval (Q52148297) scheint eine Dublette von Jean Baptiste Boisduval (Q722712) zu sein. --Kolja21 (talk) 00:38, 14 May 2018 (UTC)
- Danke, merged. Der Bot-Run hat neue Einträge basierend auf Namen und Geburts-/Todesjahr (mit mehreren Mix'n'match-Quellen) angelegt. Wegen unterschiedlichen Geburtsjahren hat er den existierenden Eintrag nicht gefunden. Das ist durchaus so geplant; ein paar gut referenzierte Dubletten können meist gut gefunden werden (z.B. VIAF-Duplikation), und ich finde, Datum-"Streitigkeiten" sollten durchaus dokumentiert werden. Die meisten neuen Einträge sollten aber "unique" sein. --Magnus Manske (talk) 07:24, 14 May 2018 (UTC)
- Das klingt einleuchtend. Herauszufinden, ob Edward Fisher (Q5342908), "Irish mezzotint engraver" und Edward Fisher (Q53045850), "artist born on 1722 at ? and deceased on 1785 in ?" wirklich die gleiche Person sind, ist aber trotz identischer Normdaten gar nicht so leicht. Apropo Normdaten: Alleine bei der GND liegen mittlerweile über 3.000 "Single value" violations vor. Ich habe keine Ahnung, wie die Wartung langfristig bewältigt werden soll. Gruß --Kolja21 (talk) 23:52, 14 May 2018 (UTC)
- Ich hab' mal was gebaut, das die GND-Redirects entfernt. Kann ich täglich laufen lassen. --Magnus Manske (talk) 08:40, 15 May 2018 (UTC)
- Super: Mit einem Botlauf 437 Fälle erledigt. Danke! --Kolja21 (talk) 12:44, 15 May 2018 (UTC)
- Very bad if redirects are deleted! That will break incoming links via the property value. 2.247.26.14 16:35, 16 June 2018 (UTC)
- Ich hab' mal was gebaut, das die GND-Redirects entfernt. Kann ich täglich laufen lassen. --Magnus Manske (talk) 08:40, 15 May 2018 (UTC)
- Das klingt einleuchtend. Herauszufinden, ob Edward Fisher (Q5342908), "Irish mezzotint engraver" und Edward Fisher (Q53045850), "artist born on 1722 at ? and deceased on 1785 in ?" wirklich die gleiche Person sind, ist aber trotz identischer Normdaten gar nicht so leicht. Apropo Normdaten: Alleine bei der GND liegen mittlerweile über 3.000 "Single value" violations vor. Ich habe keine Ahnung, wie die Wartung langfristig bewältigt werden soll. Gruß --Kolja21 (talk) 23:52, 14 May 2018 (UTC)
Im VIAF-Eintrag 288051 kann ich das Geburtsdatum nicht finden. Woher kommen die Daten? --Färber (talk) 12:03, 25 May 2018 (UTC)
- Aus den RDF-Daten. Keine Ahnung, warum die nicht im normalen interface auftauchen. --Magnus Manske (talk) 12:32, 25 May 2018 (UTC)
Nadia Magnenat Thalmann
Hallo. Geht um dies. Es gibt auch andere Quellen mit einem anderen Jahrgang: Auteurs (form. 1982) Née 1942. Canadienne. Was tun? Danke --KurtR (talk) 15:33, 25 May 2018 (UTC)
False edits
Hi. This bot makes recent repeated false edits like this one, this or also this one. The source doesn't indicate any of these informations, which are aberrants (an anonymous master can't have a birth and death date, only a period of activity). I revert some of them but I don't know if there's other ones. Thanks. Mel22 (talk) 18:50, 25 May 2018 (UTC)
- same here and here (reverted) Eru (talk) 20:23, 25 May 2018 (UTC)
- I just checked the marc21 on viaf, there is a date with 1950, but it's say "flourished", what ever it's mean
- <mx:datafield tag="997" ind1=" " ind2=" "> <mx:subfield code="a">1950</mx:subfield> <mx:subfield code="b">0</mx:subfield> <mx:subfield code="c">flourished</mx:subfield> </mx:datafield> Eru (talk) 21:12, 25 May 2018 (UTC)
What are you doing? —Tom.Reding (talk) 01:49, 26 May 2018 (UTC)
Bonjour, j'ajoute ma plainte aux plaintes précédentes. C'est une catastrophe : les ajouts sont tous identiques (année de naissance 1950), complètement fantaisistes et sourcés par une source imaginaire du VIAF !!! --Pierrette13 (talk) 06:13, 26 May 2018 (UTC)
- Bonjour. Ah, je vois que je ne suis pas le seul à être interloqué par ces naissances en 1950. Par exemple sur Cécile Prieur (que j'ai annulé) et Stephen Schultz, sourcés de manière tout aussi imaginaire sur le VIAF. Je suis surpris que depuis mai il n'y ait pas eu d'annulation en bloc de ces modifs. Cordialement, Xavier 90.40.187.195 15:59, 19 December 2018 (UTC)
Your bot incorrectly read birth and death dates from VIAF. Please check to see what VIAF records actually say, and modify your bot's script accordingly. It must be able to recognize dates that are approximate (circa) and import them accordingly, and should also identify which records on VIAF were used when VIAF import records vary. --EncycloPetey (talk) 14:05, 26 May 2018 (UTC)
Your bot also added birth and death dates to Dhanañjaya (Q51137951), but this time VIAF did not have those dates. Your bot invented its own dates. --EncycloPetey (talk) 14:10, 26 May 2018 (UTC)
- @Magnus Manske:, if you don't automatically get notifications whenever this page is updated. Mahir256 (talk) 20:54, 26 May 2018 (UTC)
- @Mahir256, Magnus Manske, EncycloPetey, User:Pierrette13: in fact, I see this date very clearly in xml format of the viaf record https://viaf.org/viaf/263357483/viaf.xml for example, except, it's not a year date, it's a century date : probably a wrong reading of the date encoding (like when 20th century is displayed 2000-01-01 on wd).
- That VIAF XML record says <ns1:dateType>flourished</ns1:dateType> so should not be used as a lifespan (but could possibly be used for the floruit (P1317) property). --Canley (talk) 23:53, 27 May 2018 (UTC)
- please Magnus, could you re-check the encoding of dates on viaf. Many century dates replaced by a date in x50... which is quite wrong :/ --Hsarrazin (talk) 18:21, 27 May 2018 (UTC)
- @Mahir256, Magnus Manske, EncycloPetey, User:Pierrette13: in fact, I see this date very clearly in xml format of the viaf record https://viaf.org/viaf/263357483/viaf.xml for example, except, it's not a year date, it's a century date : probably a wrong reading of the date encoding (like when 20th century is displayed 2000-01-01 on wd).
Invaluable dates
Please can you stop adding birth/death dates from M'n'M catalogue 1056 - it seems they are not always correct. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 09:42, 27 May 2018 (UTC)
- Nr. 1264 as well https://www.wikidata.org/w/index.php?title=Q325097&type=revision&diff=688317704&oldid=688306557 --Masegand (talk) 13:07, 1 June 2018 (UTC)
Wrong year of birth and year of death
Hi,
I would like to inform you that, almost all of your edits from Cheng Zhiyun (Q45643012) 17:13, 16 December 2017 to Liu Tingchen (Q45666339) 18:35, 17 December 2017 are wrong. The years of birth and years of death do not match the China Biographical Database.--QBear (talk) 08:48, 12 June 2018 (UTC)
2018-01-08 P31 human
Don't know the source and if it is fixed. Not human: https://www.wikidata.org/w/index.php?title=Q47115982&type=revision&diff=617757960&oldid=617757919 2.247.48.113 15:37, 26 June 2018 (UTC)
Nehemiah
Reporting some erroneous edits. Cheers, Bovlb (talk) 17:49, 27 June 2018 (UTC)
Import von Platzhaltern
Hi Magnus, danke für den Import aus kaiserhof.geschichte.lmu.de. Nur eine Bitte: Kannst du den Bot so einstellen, dass er die Tns (= Platzhalter) nicht als Dubletten importiert? Beispiel: Johann Rudolf Sprinzenstein (Q55238177) mit gültiger GND (Tp = Person) 139320393 und dem Platzhalter (Tn = Name = BKL in Wikipedia): 102979030. Gruß --Kolja21 (talk) 02:03, 10 July 2018 (UTC)
- Ich werde mein Bestes tun, aber leider sieht man das der ID ja nicht an... --Magnus Manske (talk) 08:18, 10 July 2018 (UTC)
- Danke für die Rückmeldung. Es gibt leider in der Tat keinen getrennten Nummernblock für Platzhalter, aber die Wartungslisten in deWP (User:Wurgl kennt sich mit den technischen Details aus) kriegen das hin. Gruß --Kolja21 (talk) 01:54, 11 July 2018 (UTC)
- Moin!
- Möglichkeiten der Prüfung auf nicht individualisierten Datensatz (Tn):
- Du gehst auf die Webseite
https://portal.dnb.de/opac.htm?method=simpleSearch&cqlMode=true&query=nid%3D<hier Tn einfügen>
und suchst nach der Zeichenfolge<strong>Name</strong>
Bei gefunden --> ist ein Tn, wenn nicht gefunden --> ist ein individualisierter Datensatz (Person oder Körperschaft oder Sachbegriff oder Geographikum oder Veranstaltung). Ob eine GND-Id (egal welcher Subtyp) überhaupt gültig ist, kannst du mit einem HEAD-Request aufhttp://d-nb.info/gnd/<hier Tn einfügen>/about/html
feststellen. Bei Status 404 ungültig/nicht existent, sonst via Status 302 eine Umleitung auf eine URL wie oben. - Du lädst dir von https://data.dnb.de/opendata/ die Datei Tngesamt1806gnd.<Endung> und suchst die Id. Wie der Name schon sagt, sind in der Datei nur diese nicht-individualisierten Ids drinnen. Allerdings gibt das ein kleines Problem: Eine Tn kann individualisiert werden und zu einer Tp werden (ich hatte gestern zwei solche Fälle) und diese Dump-Files werden 3 mal im Jahr (Februar, Juni und Oktober) neu erzeugt. Bei ganz frisch angelegten oder bei solchen die gerade erst dieses "Upgrade" von Tn nach Tp erfuhren, läuft die Prüfung falsch.
- Du gehst auf die Webseite
- Ach ja, der User-Agent bei einem HTTP-Zugriff sollte etwas darstellen, das von der Webseite als Browser identifiziert werden kann. Wenn du als User-Agent sowas wie
"Super-Duper-Bot (siehe https://www.wikidata.org/wiki/User:Reinheitsgebot)"
hast, dann mag dich die Webseite nicht gar so sehr und murmelt oft was in der Art "Session ist abgelaufen". Ich hab rotzfrech den User-Agent meines Browsers reingemacht und gut war es. - Hoffe das hilft. --Wurgl (talk) 05:27, 11 July 2018 (UTC)
- Danke für die Rückmeldung. Es gibt leider in der Tat keinen getrennten Nummernblock für Platzhalter, aber die Wartungslisten in deWP (User:Wurgl kennt sich mit den technischen Details aus) kriegen das hin. Gruß --Kolja21 (talk) 01:54, 11 July 2018 (UTC)
Duplicates from pseudonyms
At Wikidata:Database_reports/identical_birth_and_death_dates/1, items like Q55685004 duplicating Q1000323 show up. Maybe there is a way to match them with the real name before upload.
--- Jura 07:58, 28 July 2018 (UTC)
- They are checked against external IDs (VIAF and GND, for Heinrich Wilhelm Lawätz (Q55685004)), and against Wikidata search, before creation. That is, if any item comes up in the name search, or already has one of the external IDs, it is not created. Not sure what else I can do. Also, not sure what a "real name" is, in this context. --Magnus Manske (talk) 08:11, 1 August 2018 (UTC)
- Die Frage in diesem Fall ist: Soll das Objekt für das Pseudonym erhalten bleiben? ("real name" = Heinrich Wilhelm Lawätz (Q1000323) = GND 123529751.) Auch die GND hat in diesem Fall ja bewusst zwei Datensätze. BTW: Nur aus Neugierde: Wie wählst du aus den über 4 Millionen Tps die Personen aus, für die du Objekte anlegst? --Kolja21 (talk) 11:19, 1 August 2018 (UTC)
- Viaf mentions the other name (sourced from GND, I think). If items for pseudonyms are created, they shouldn't have "P31=Q5", but "P31=pseudonym (Q61002)".
Given that the items are quite complete, I think merging them is fairly straightforward. So the occasional duplicate isn't much of an issue. I would be more worried if the duplicates were nearly empty and almost identical to existing ones.
--- Jura 11:25, 1 August 2018 (UTC)
Deutsche Biographie descriptions
Any reason why clearly German descriptions for the Deutsche Biographie entries your bot is creating are ending up as English descriptions here? Mahir256 (talk) 20:37, 31 July 2018 (UTC)
- Wrong setting, changed now. --Magnus Manske (talk) 08:07, 1 August 2018 (UTC)
Sigismund Päminger
Hallo Magnus, bei Sigismund Päminger (Q55875077) nimmt dein Bot an, dass der "Komponist protestant. Motetten (1495-1567)" mit dem "Beitr. im VD-16" identisch ist. Ich würde die beiden eher getrennt lassen, da der VD-16 Päminger unter anderem Leichenpredigten verfasst hat, also nicht zwangsläufig mit dem Komponist identisch ist. --Kolja21 (talk) 21:42, 3 August 2018 (UTC)
Q1010102
Der revert meines Eintrags in Q1010102 ist fachlich nicht richtig.
Bunyaviridae wurden 2016 nach Peribunyaviridae umbenannt.
Phenuiviridae sind eine neue Virusfamilie seit 2016.
--Murma174 (talk) 11:02, 5 August 2018 (UTC)
P.S. Auch der Eintrag von enwiki in Q1010102 ist nicht richtig. --Murma174 (talk) 17:10, 7 August 2018 (UTC)
- @Magnus Manske: Ping. Tommy Kronkvist (talk), 17:35, 7 August 2018 (UTC).
2030er
Guten Tag Reinheitsgebot, Du hast zu de:2030er hier einen neuen Datensatz angelegt, obwohl es dazu bereits einen gegeben hat, der mit 60 Spraxhversionen verknüpft ist. Ich habe jetzt einen Löschantrag gestellt. Es wäre also leicht möglich gewesen, die Existenz festzustellen. Bitte prüfe in Zukunft, ob es nicht bereits entsprechende Einträge gibt. —Lantus 06:48, 26 August 2018 (UTC)
Adding reference(s) from Wikipedia
Hello Magnus Manske Bot, adding reference from Wikipedia in any language is not valid. Wikipedia whatever the language cannot be used as a source/reference. Please prefer VIAF et al. Thanks. Cheers! DDupard (talk) 17:42, 19 September 2018 (UTC)
- References to Wikipedia have been used here as a way to keep track of the source of information, in the hopes that this can be replaced by humans with a proper reference at some point. Do you have a specific example? --Magnus Manske (talk) 21:26, 19 September 2018 (UTC)
#quickstatements; invoked by Mix'n'match:References https://tools.wmflabs.org/mix-n-match/?#/catalog/17
https://www.wikidata.org/w/index.php?title=Q1909462&type=revision&diff=751744191&oldid=737846896 Something just went wrong with about 24 musicians declared dead / P569=P570 from beic.it https://tools.wmflabs.org/mix-n-match/?#/catalog/17 --Masegand (talk) 14:39, 24 September 2018 (UTC) it is still continuing https://www.wikidata.org/w/index.php?title=Q2272644&diff=prev&oldid=751775457.--Masegand (talk) 15:49, 24 September 2018 (UTC)
Bot's mistakes
[6], [7] and many like that. Elfhelm (talk) 15:29, 24 September 2018 (UTC)
Judi Dench (Q28054)
Hi,
Are you sure ? I think there's a problem, the person is alive and the source does not correspond.
There are many others, I didn't check them all.
Best regards --Do not follow (talk) 20:48, 25 September 2018 (UTC)
Some more errors
Both had their year of births added as a death date, and both still breathing AFAIK. These just happen to be some people I watch, I have not checked but I suspect likely to be others. Periglio (talk) 15:18, 26 September 2018 (UTC)
Filmportal-Darsteller
Habe eben Frieder Rometsch (Q55683036) gesehen, könntest du das auch mit allen Personen in [10] machen, die nur "Darsteller"/"Darstellerin" (+ ggf. Lebensdaten) und keine Suchergebnisse haben? Queryzo (talk) 20:38, 21 October 2018 (UTC)
- BTW: Filmportal.de hat vor einigen Jahren alle biographischen Datensätze in die GND eingespielt. Trotzdem finden sich immer wieder Fälle, in denen keine GND vergeben wurde. Es wäre interessant zu wissen, welche Personen das sind, dann können wir von Hand nachbessern. --Kolja21 (talk) 16:20, 22 October 2018 (UTC)
Jean-Louis Besson (Q3166634)
Hi,
Jean-Louis Besson (Q3166634) is a French association football player. Margaret Jean Adair Besson isn't the same person, she is Jamaican anthropologist.
Best regards. --34 super héros (talk) 18:21, 3 November 2018 (UTC)
Klaus Groth
Hallo Reinheitsgebot, du hattest bei Klaus Groth Q76494 Geokoordinaten und Adresse hinzugefügt, das ist ungewöhnlich und wohl auch eigentlich nicht gewollt. Sollte das vielleicht für das Klaus-Groth-Museum Q1744763 gelten ? -- Gerd Fahrenhorst (talk) 19:06, 9 November 2018 (UTC)
Incorrect interpretation of dates in Q1165538 leading to persons still alive getting P570 by Reinheitsgebot invoked by Mix'n'match
For a number of items related to still living people, Reinheitsgebot has added a date of death (P570). It seems various dates in the source Nationalencyklopedin (Q1165538) have been incorrectly used as date of death (P570). I have undone these 19 edits. --Larske (talk) 22:53, 11 November 2018 (UTC)
Chemistry
This is the second time I had to revert edits made by this bot:
- acetyl (Q481808): 1, 2; #quickstatements; #temporary_batch_1541195733806; invoked by mixnmatch:aux2wd, #quickstatements; #temporary_batch_1542070512957; invoked by mixnmatch:aux2wd.
- pyrroline (Q899707): 1, 2; #quickstatements; #temporary_batch_1541576355836; invoked by mixnmatch:aux2wd, #quickstatements; #temporary_batch_1542063308929; invoked by mixnmatch:aux2wd
Neither the person responsible for these edits nor source of the data is provided, so I can't even check what could cause this problem. Wostr (talk) 01:22, 13 November 2018 (UTC)
Wrong automatization of dates
Hi, I see that several persons have already indicated the problem. The bo does not take correctly the information of the databases it consulted, for instance for Cosette Harcourt (Q23926334), the source indicates 1968-1976 and the bot put January 1st 1968 as the date of death (in fact the correct date is 1976). Could you please correct this (or if there are several dates in a source, make the bot not take any of them) ? Thank you, --Cgolds (talk) 08:46, 18 November 2018 (UTC)
How do you guess birth and death dates from VIAF?
Hello,
I see here+here that you add birth and death dates according to VIAF, who doesn't know those dates.
How do you proceed?
Regards, --Daehan (talk) 09:04, 28 December 2018 (UTC)
bad P570 statements
https://www.wikidata.org/w/index.php?title=Q15449518&diff=prev&oldid=830475638 and https://www.wikidata.org/w/index.php?title=Q12963&diff=prev&oldid=830455798 --Masegand (talk) 18:09, 9 January 2019 (UTC)
Putting errors back in
Over the past few days I have painstakingly removed errors. I see your bot now has put them back in:
04:19, 10 January 2019 diff hist +353 Anisostachya elytraria (Q17679645) Created claim: PlantList-ID (P1070): tro-104126, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847
04:19, 10 January 2019 diff hist +353 Justicia valerii (Q17680347) Created claim: PlantList-ID (P1070): tro-102781, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847
04:19, 10 January 2019 diff hist +354 Lepidagathis alvarezia (Q17680510) Created claim: PlantList-ID (P1070): tro-50287058, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847
04:19, 10 January 2019 diff hist +353 Chasmatophyllum musculinum (Q17248247) Created claim: PlantList-ID (P1070): kew-2715967, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847
04:19, 10 January 2019 diff hist +353 Drosanthemum anomalum (Q17244917) Created claim: PlantList-ID (P1070): kew-2778129, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +353 Caralluma bhupinderiana (Q18083982) Created claim: PlantList-ID (P1070): kew-2699067, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +353 Dischidia elmeri (Q18080994) Created claim: PlantList-ID (P1070): kew-2772504, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +352 Melodinus philippinensis (Q18083139) Created claim: PlantList-ID (P1070): kew-124254, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +354 Secamone oleifolia (Q18078255) Created claim: PlantList-ID (P1070): tro-2602338, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +352 Polyscias bellendenkerensis (Q27828718) Created claim: PlantList-ID (P1070): kew-162450, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +355 Lithospermum tubuliflorum (Q17416623) Created claim: PlantList-ID (P1070): tro-50319763, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +353 Cardamine chenopodiifolia (Q17242858) Created claim: PlantList-ID (P1070): kew-2699586, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +351 Aechmea vanhoutteana (Q4687417) Created claim: PlantList-ID (P1070): kew-218298, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +354 Spergularia fasciculata (Q17241355) Created claim: PlantList-ID (P1070): tro-6301643, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +353 Calophyllum touranense (Q17565266) Created claim: PlantList-ID (P1070): kew-2693527, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +352 Ageratina calaminthifolia (Q15573812) Created claim: PlantList-ID (P1070): gcc-147859, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +353 Erica harveyana (Q17234855) Created claim: PlantList-ID (P1070): kew-2793451, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +353 Erica bruniades (Q17234951) Created claim: PlantList-ID (P1070): kew-2792806, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +350 Croton sarcocarpus (Q25673731) Created claim: PlantList-ID (P1070): kew-51099, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +352 Jatropha stephani (Q49519258) Created claim: PlantList-ID (P1070): kew-104932, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +353 Gentianella nummulariifolia (Q18078308) Created claim: PlantList-ID (P1070): kew-2822195, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +354 Smithiantha cinnabarina (Q17709077) Created claim: PlantList-ID (P1070): tro-14001042, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +353 Cryptocarya bellendenkerana (Q18077442) Created claim: PlantList-ID (P1070): kew-2745858, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +354 Bunchosia matudae (Q49524918) Created claim: PlantList-ID (P1070): kew-2686010, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +354 Cristaria molinae (Q17579749) Created claim: PlantList-ID (P1070): kew-2742853, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +353 Firmiana minahassae (Q17573784) Created claim: PlantList-ID (P1070): kew-2812997, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +354 Turraea madagascarensis (Q18078224) Created claim: PlantList-ID (P1070): tro-50217215, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +352 Melaleuca phratra (Q20721780) Created claim: PlantList-ID (P1070): kew-463477, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +353 Syzygium velarum (Q27828808) Created claim: PlantList-ID (P1070): kew-200470, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +353 Epilobium nummulariifolium (Q21257436) Created claim: PlantList-ID (P1070): kew-2790782, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +354 Amitostigma tominagae (Q15237077) Created claim: PlantList-ID (P1070): tro-50024950, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +350 Corybas muscicola (Q54922157) Created claim: PlantList-ID (P1070): kew-47752, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +350 Cymbidium munronianum (Q39336955) Created claim: PlantList-ID (P1070): kew-53299, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +350 Habenaria ferdinandi (Q56234888) Created claim: PlantList-ID (P1070): kew-94623, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +352 Pennilabium proboscideum (Q11059402) Created claim: PlantList-ID (P1070): kew-149776, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +352 Prasophyllum viretrum (Q42329972) Created claim: PlantList-ID (P1070): kew-348050, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +354 Paropsia vareciformis (Q17558939) Created claim: PlantList-ID (P1070): tro-24201625, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:19, 10 January 2019 diff hist +354 Piper verruculifolium (Q18037930) Created claim: PlantList-ID (P1070): tro-25001925, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:18, 10 January 2019 diff hist +354 Piper schlechtendalianum (Q18038103) Created claim: PlantList-ID (P1070): tro-25000906, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:18, 10 January 2019 diff hist +354 Piper mikaniifolium (Q18036805) Created claim: PlantList-ID (P1070): tro-25001710, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:18, 10 January 2019 diff hist +353 Piper canovillosum (Q18035716) Created claim: PlantList-ID (P1070): kew-2873179, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:18, 10 January 2019 diff hist +354 Russelia syringifolia (Q49519418) Created claim: PlantList-ID (P1070): tro-29200880, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:18, 10 January 2019 diff hist +352 Elymus cacuminis (Q24853084) Created claim: PlantList-ID (P1070): kew-410965, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:18, 10 January 2019 diff hist +354 Alchemilla sibbaldiifolia (Q15289093) Created claim: PlantList-ID (P1070): tro-27800469, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:18, 10 January 2019 diff hist +354 Zanthoxylum rhodoxylon (Q18084059) Created claim: PlantList-ID (P1070): tro-50100043, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:18, 10 January 2019 diff hist +354 Salix balansae (Q17561917) Created claim: PlantList-ID (P1070): tro-50146588, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:18, 10 January 2019 diff hist +354 Serjania rutifolia (Q18080588) Created claim: PlantList-ID (P1070): tro-28601475, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
04:18, 10 January 2019 diff hist +354 Pseudocyclosorus angustipinnus (Q15239168) Created claim: PlantList-ID (P1070): tro-50166326, #quickstatements; mixnmatch:microsync for catalog 1005 #temporary_batch_1547093922847 current
The purpose of Wikidata is not to track every error anybody ever made on the www and import them into Wikidata to perpetuate them. - Brya (talk) 05:15, 10 January 2019 (UTC)
- Apologies for causing you additional work. Looking at your first example, Anisostachya elytraria (Q17679645), it appears you "added" Plant List ID (Royal Botanic Gardens, Kew) (P1070) yourself by merging, then removed it again. Since no "P1070" shows up in the item's history, it is, for all practical purposes, impossible to automatically detect that such a property was ever edited/deleted.
- Your remarks imply that I and my bot should, somehow, magically, have known what is correct and what is not, and that we have offended you, Wikidata, and the world by not acting on that fictitious knowledge. No worries, once I finish my Skynet AI project, no more errors will be tolerated. Until then, shit happens. --Magnus Manske (talk) 10:08, 10 January 2019 (UTC)
- Hi, Magnus Manske, thank you for responding. Yes, I did merge the erroneous item, and then deleted the erroneous content, as the fastest way to create a redirect. And creating a redirect is faster than putting in a request for deletion.
- As to "I and my bot should, somehow, magically, have known what is correct and what is not", in view of the very many errors in The Plant List ("P1070"), it seems safe to assume that if by this time "P1070" is not there, there is a reason for its absence. - Brya (talk) 05:16, 11 January 2019 (UTC)
- The bot does not work for Plant List ID (Royal Botanic Gardens, Kew) (P1070), it runs on all of Mix'n'match, which is >50M entries in >2K catalogs, of which The Plant List is only a small part. It appears there are still >12K Plant List items not matched to Wikidata. Assuming they are all wrong is quite a leap (disclaimer: I don't care about Plant List, specifically, one way or another). --Magnus Manske (talk) 10:10, 11 January 2019 (UTC)
- 12K sounds like a lot, but in view of the size and nature of The Plant List it would not be a stretch to say there can easily be two or three times that number of errors in that database. The Plant List is an aggregator of other databases, so basic quality is never better than that of the component databases. At least one of these is quite good, but at least one other is out of date by some twenty years. So the basic quality is less than top of the line. The big problem is that one of the component databases (in itself pretty good) was miscopied, creating massive error.
- Those 12K may not be all wrong, but it seems safe to bet that >95% is indeed wrong. Otherwise they would already have an item here. - Brya (talk) 11:51, 11 January 2019 (UTC)
Duplicate references with a false date of consultation
Hello. From time to time and since several months you duplicate the source of HDS, with a false date of consultation (always the same, 9 October 2017). Like in [11]. Greetings, Sapphorain (talk) 20:28, 14 January 2019 (UTC)
- ... you did it again, in (Q30731373). Sapphorain (talk) 10:35, 5 February 2019 (UTC)
- ... and again, in (Q3173987). Can you please stop that?. Sapphorain (talk) 10:39, 5 February 2019 (UTC)
- ...and again in (Q3154793), (Q42803664), (Q3579819)... Sapphorain (talk) 11:51, 5 February 2019 (UTC)
- ...and again (Édouard Diodati (Q3579819), Jean Pierre Girard-dit-Vieux, (Q3173987)Antoine Maurice (Q30731373)). Can you please stop that nonsense? Sapphorain (talk) 05:14, 5 March 2019 (UTC)
- ...and again ( Jean Frédéric Théodore Maurice (Q42803664)). Please DO stop vandalizing. Sapphorain (talk) 09:45, 5 March 2019 (UTC)
- ... and again 3 more. Hello? Is there somebody behind this stupid bot? Is this somebody stupid also, or just a vandal? Sapphorain (talk) 16:09, 28 March 2019 (UTC)
- And again... Please do stop vandalizing. Sapphorain (talk) 22:38, 19 April 2019 (UTC)
- ...And again... Moreover, the date of consultation is false (2017), and thus misleading (the old edition available in 2017 does not exist anymore). Please stop vandalizing. Sapphorain (talk) 08:12, 1 November 2019 (UTC)
- And again... Please do stop vandalizing. Sapphorain (talk) 22:38, 19 April 2019 (UTC)
- ... and again 3 more. Hello? Is there somebody behind this stupid bot? Is this somebody stupid also, or just a vandal? Sapphorain (talk) 16:09, 28 March 2019 (UTC)
- ...and again in (Q3154793), (Q42803664), (Q3579819)... Sapphorain (talk) 11:51, 5 February 2019 (UTC)
- ... and again, in (Q3173987). Can you please stop that?. Sapphorain (talk) 10:39, 5 February 2019 (UTC)
Strange edit
Hi,
Was this edit Special:Diff/151448396 done on purpose? It seems like the bot removal was a mistake. If I'm correct, it created a bug and we should now merge The Hall (Q17537656) and The Hall (Q17582911). What do you think?
Cheers, VIGNERON (talk) 20:40, 16 January 2019 (UTC)
Weird dates of death
It looks like your bot is claiming that some people died in 2050. You probably want to keep your time traveling under wraps and not claim that 31 years in advance... Sumurai8 (talk) 23:03, 18 January 2019 (UTC)
- Should be fixed now.--Magnus Manske (talk) 10:51, 5 February 2019 (UTC)
1900 birth dates from VIAF
Your bot has given a lot of living academics 1900 birth dates with a VIAF ref (e.g. Rosamund Bartlett (Q7367298), Kim Barrett (Q29653444), Jerome Groopman (Q6182708)). --ASchedulingError (talk) 19:58, 10 February 2019 (UTC)
bad P570 statements
seems to use the entry of a different person https://www.wikidata.org/w/index.php?title=Q4098419&type=revision&diff=857308682&oldid=839632159 --Masegand (talk) 13:26, 13 February 2019 (UTC) This one https://www.wikidata.org/w/index.php?title=Q159913&action=history twice--Masegand (talk) 15:41, 14 February 2019 (UTC) Here it is also wrong twice: https://tools.wmflabs.org/mix-n-match/?#/search/klaus%20kinkel?include=875 --Masegand (talk) 21:35, 14 February 2019 (UTC)
- Hi. The bot keeps adding wrong P570 statements. Example: Klaus Kinkel (Q159913). --FShbib (talk) 18:59, 15 February 2019 (UTC)
- Also today, but only a small amount and again from catalog 875-CANTIC-ID: Q1528001 and Q1153825--Masegand (talk) 16:42, 21 February 2019 (UTC)
- The same here for a just created entry: https://www.wikidata.org/wiki/Q61831008 --Toniher (talk) 21:23, 23 February 2019 (UTC)
- Today here: https://www.wikidata.org/w/index.php?title=Q212730&type=revision&diff=866841213&oldid=861425324 --Masegand (talk) 09:41, 25 February 2019 (UTC)
Invalid P4491 statements
Hello, over the past months, your bot has added the sources' URLs instead of IDs as Isidore identifiers (property P4491) to several objects. Examples: Special:Diff/819817353, Special:Diff/863631057, Special:Diff/878996613. Regards, Axolotl Nr.733 (talk) 08:45, 10 March 2019 (UTC)
Dates of birth and death for "Giovanni Gabrielli" (Q18823013)
Your bot added added a date of birth of 1550 and date of death of 1551 to Q18823013, Gabrielli's Wikidata page, based on his VIAF page and/or the linked Library of Congress page, but I see no evidence for those dates on either of those pages, so I deleted them from Wikidata. In any case, his dates of birth and death are not known, that I know of. I don't know why your bot did this, but I believe it was done in error. --Robert.Allen (talk) 04:48, 15 March 2019 (UTC)
bad edit
Something wrong with this bot edit. The person would have lived 200 years. Also I could not find those dates at the page used as a reference. --Jarekt (talk) 13:39, 18 March 2019 (UTC)
There are a lot of duplicate items created where the names do not exactly correspond to existing pages. Most (if not all) Canadian MPs already have a Wikipedia page and thus a Wikidata item. Ebe123 (talk | contributions) 13:32, 2 April 2019 (UTC)
Bad P214 statements
A recent batch introduced junk values for VIAF ID (P214) in several objects:
- Volk (Q16089832)
- Gregory (Q1333689)
- Morris (Q2780710)
- Robertson (Q2159662)
- Duke of Norfolk (Q521109)
I'd love to fix myself, but I don't have a source. — Yerpo Eh? 08:50, 19 April 2019 (UTC)
P813: Abgerufen an
Hello Magnus Manske,
Thanks for all your work on Wikidata.
I just saw your bot doing this edit on the Wikidata-article on 'Elisabeth von Österreich'. I'm curious about the given date for retrieved (P813), which is put to 'October 9, 2017'. Did your bot keep this information for one and a half year for itself? Or is the information retrieved more recently, but is the date hard coded?
Thanks in advance, RonnieV (talk) 22:05, 3 May 2019 (UTC)
- Same problem [12] on May 23, 2019. Does your bot work with a fixed date for 'retrieved'? I hope you can have a look at it. Thanks, RonnieV (talk) 09:23, 13 June 2019 (UTC)
Conflating different people
I have reverted [13] and [14] because they appear to conflate two different people. I don't know how this bot works, but I don't think you should be either: a) adding a second ORCID to an item that already has one; or b) adding an ORCID that is already present on a different item. Cheers, Bovlb (talk) 22:34, 1 July 2019 (UTC)
- Also [15] [16] [17] Bovlb (talk) 22:48, 1 July 2019 (UTC)
- @Magnus Manske: Here is another example. Are we assuming they're the same person because they both have the last name "Smith"? Bovlb (talk) 16:42, 4 July 2019 (UTC)
Redundant references
Please explain the merit of this edit which adds yet another reference to the death date of Galileo, even though there are already many references to reliable sources for this fact. Jc3s5h (talk) 16:55, 3 July 2019 (UTC)
Guggenheim fellows ID as reference for date of death
Hello, the bot is currently adding the Guggenheim fellows ID as references for dates of death (example), which is fine. However, it's incorrect to say that the info is stated in the Solomon R. Guggenheim Museum, because Guggenheim fellowships are awarded by the John Simon Guggenheim Memorial Foundation, a different organization. --Axolotl Nr.733 (talk) 20:32, 3 July 2019 (UTC)
Manual worker
The bot added a statement to Q55720463 that Jane Elizabeth Robbins was a manual worker, but I cannot find it in the given source. It was probably mistaken by the phrase "settlement house worker". I have corrected it but decided to let you know in case the bot made more mistakes like this one. --Jan Kameníček (talk) 17:49, 11 August 2019 (UTC)
Normdatei der Spanischen Nationalbibliothek
Die BNE importiert wie Wikidata ihre Daten ohne Überprüfung aus allen möglichen Quellen; bei der GND gerne auch Platzhalter, Dubletten und Namensvetter. Die "Normdaten" sind daher mit Vorsicht zu gebrauchen. Die Beschreibung bei Augusto Córtina Aravena (Q55947699) lautet sinngemäß: "Entdecken Sie Cortina, Augusto, im offenen und verlinkten Datenportal der spanischen Nationalbibliothek", d.h. raten Sie mit, um wen es sich handeln könnte. Vermutlich handelt es sich um Augusto Córtina Aravena (Q56318455). Aber bei häufiger vorkommenden Namen sinkt die Trefferquote schnell unter 50 Prozent. Daher sollte man aus der BNE imho nur jene Datensätze importieren, die das Geburtsjahr oder den Beruf der Person enthalten. Gruß --Kolja21 (talk) 15:18, 31 August 2019 (UTC)
Death date for someone still living?
Hi,
Your bot added a death date for this writer in 2014, without any source. We (OTRS) received today an email about the writer still beeing alive. Indeed, she published a book in 2016! A wrong information of this nature can cause harm to people. Please be more careful and cite source!
Jules78120 (talk) 18:29, 7 October 2019 (UTC)
- Ok, your bot imported a false information from en-wp ([18]). Please do not make your bot import information without source when related to living persons: instead of having a vandalism only on English Wikipedia, we had it too on French Wikipedia… Jules78120 (talk) 18:39, 7 October 2019 (UTC)
Fraglicher Edit bei Michael Chapman (Q70022813)
@Magnus Manske: Dieser Edit ist aus meiner Sicht wenig schlüssig: Erstens hat der VIAF als solcher keinen Quellenwert, sondern nur die verlinkten Normdaten. Zweitens ergibt auch das „named as: Michael Chapman“ keinen Sinn, weil der VIAF keine feste Ansetzform bereitstellt. --Emu (talk) 07:36, 8 October 2019 (UTC)
Very unlikely aliases being added (at least in June)
I've run across several cases like this, here's the edit I just undid. Why would you be adding such aliases to researchers? It looks like something is confusing one person with another who has the same first name but very different last name??? ArthurPSmith (talk) 15:35, 8 October 2019 (UTC)
- Just hit another here, also from June. ArthurPSmith (talk) 15:41, 8 October 2019 (UTC)
- Here's another. That same item also had a more recent (September) edit from Quickstatementsbot that seemed to follow the same pattern of adding an entirely bogus "alias" value: here. It looks like whatever was doing this has migrated from Reinheitsgebot to Quickstatementsbot ??? The edit summaries are extremely undescriptive also "Updated item". ArthurPSmith (talk) 19:21, 8 October 2019 (UTC)
- David-Alexandre Trégouët (Q47502434) is probably the item for the alias added to David M. Evans (Q30348322). Maybe they authored a paper together (I couldn't find that) or it was the next row in some list. --- Jura 19:50, 8 October 2019 (UTC)
- Looking at [19], the above seem rather exceptional .. --- Jura 20:05, 8 October 2019 (UTC)
- I keep running across these - here's another: Nicholas Eriksson and Nicholas Wood surely different people. ArthurPSmith (talk) 21:00, 8 October 2019 (UTC)
- And another: He Gao added as alias to Ruifang Li-Gao ??? ArthurPSmith (talk) 21:43, 8 October 2019 (UTC)
bad death date
https://www.wikidata.org/w/index.php?title=Q47002805&type=revision&diff=1028737281&oldid=1028693998 but https://viaf.org/viaf/76349591/ does not have 1999 as death date.--Masegand (talk) 15:26, 9 October 2019 (UTC)
- https://www.wikidata.org/w/index.php?title=Q881154&type=revision&diff=1043448692&oldid=1042629181 but https://www.blackpast.org/african-american-history/people-african-american-history/wilder-lawrence-douglas-1931/ does not have 1994 as death date.--Masegand (talk) 18:35, 2 November 2019 (UTC)
suggest meta user page for user:Reinheitsgebot
Hi. With this bot editing at wikidata, yet the user links showing on each wiki as a red link, can I ask for you to consider creating a user page for your bot at metawiki. It will enable users to better understand the edits being made. Thanks. — billinghurst sDrewth 21:32, 31 October 2019 (UTC)
- If you need assistance getting it done at meta, then please welcome to ping me. — billinghurst sDrewth 21:33, 31 October 2019 (UTC)
Duplicate reference to NooSFere
Hi,
Reinheitsgebot is now adding references to NooSFere (Q55790874), but it's a redirection to NooSFere (Q3343389) and they already exist, as here . — eru [Talk] [french wiki] 09:43, 2 November 2019 (UTC)
Another instance of this error. Jc3s5h (talk) 17:53, 2 November 2019 (UTC)
- @Magnus Manske, Matěj Suchánek: it's worse, MatSuBot fixes a redirect, and Reinheitsgebot recreates it, see this diff. ~~
- Hi @Magnus Manske:, still appending [20]. — eru [Talk] [french wiki] 06:50, 15 March 2020 (UTC)
Unreasonable P813 inserted
@Magnus Manske: See this edit where a reference to the date of death (P570) of Robert Mugabe (Q10707) is inserted. The retrieved (P813) of 9 October 2017 is unreasonable as the date of death (P570) is 6 September 2019, almost two years later!
- Here is another one
- Here is another one
- Here are 100 more (all with the fake retrieved (P813), most of them 9 October 2017)
Please don't insert such fake retrieved (P813). Correct or remove the already insterted. --Larske (talk) 12:55, 6 November 2019 (UTC)
False information reverted
I have reverted the false information about the calendar use for a birth date at Caroline Matilda of Great Britain (Q57668). Jc3s5h (talk) 21:03, 11 November 2019 (UTC)
Unnötige Wiederholungen
Ihr Bot scheint mehrmals dieselbe Information zu ergänzen, wie zum Beispiel in den folgenden Einträgen: [21], [22], [23], [24], [25]. Maitake (talk) 16:37, 13 November 2019 (UTC)
Request approval of Reinheitsgebot be revoked
See Wikidata talk:Bots#Request approval of Reinheitsgebot be revoked Jc3s5h (talk) 16:22, 19 December 2019 (UTC)
Alfred Mayer Q13424327
Hallo, die Aussagen "Geburtsdatum", "Sterbedatem" und "Deutsche Biographie ID" gehören zu einem anderen "Alfred Mayer". Die hier behandelte Person hat laut IPNI-Eintrag 1995 fünf Strandflieder-Arten aus Sardinien und Kreta beschrieben. Ich kenne ihn persönlich, er ist 1958 geboren und meines Wissens quicklebendig. Viele Grüße --RLJ (talk) 16:48, 13 February 2020 (UTC)
- Vielen Dank, sollte jetzt behoben sein! --Magnus Manske (talk) 09:07, 14 February 2020 (UTC)
Please stop adding wrong death dates
Three times within two days this bot has added "1651" as the year of Duchess Catherine of Cambridge's death, even though she wasn't born until 1982 and luckily seems to be alive and well. There are numerous other examples. Please make it stop. --SorenRK (talk) 17:18, 14 February 2020 (UTC)
- Hi, first when a bot adds something like this, you can just mark it as deprecated.
- In this case the problem does not come from Reinheitsgebot, the first add was made 2 days ago by Thierry Caro but it just confirmed a bad automatch.
- The mix'n'match for Deutsche Biographie (GND) ID (P7902) is 1619, and I found Catharina here, then I just deleted the bad match, now the bot won't add it anymore.
- Similarly here and here. — eru [Talk] [french wiki] 17:48, 14 February 2020 (UTC)
- @eru: when a bot adds something like this, you can just mark it as deprecated This advice is new to me. Is it documented anywhere more public? What should we put as reason for deprecated rank (P2241)? Cheers, Bovlb (talk) 14:48, 19 March 2020 (UTC)
- @eru: This is terrible advice. Deprecation should be use to indicate that a normally-credible source has published an erroneous value. It should not be used to indicate crap from out-of-control wikimedia tools or bots. Jc3s5h (talk) 16:15, 19 March 2020 (UTC)
- @Jc3s5h: I'm assuming that the motivation for this advice is that some/many bots have the feature that they will not add a claim that already exists, and thus deprecating a claim rather than deleting it is a way to prevent a bot from re-adding an erroneous claim. Obviously it is suboptimal, as not only do we retain the erroneous claim, but it is also slower for a human to do when dealing with a bot that makes many errors. Bovlb (talk) 17:16, 19 March 2020 (UTC)
- Hi, in this cas it was a bad match, so no, we shouldn't mark them deprecated, but remove them from wikidata and mix'n'match.
- Deprecated must be use when a bot (or a human) makes a good match, but the source is wrong.
- But if you don't know how to use mix'n'match and a bot adds it again, mark them deprecated can prevent this and can be use if you add reason for deprecated rank (P2241) => applies to other person (Q35773207) for example.
- It's not offical, just what I usually do. — eru [Talk] [french wiki] 17:45, 19 March 2020 (UTC)
- @Jc3s5h: I'm assuming that the motivation for this advice is that some/many bots have the feature that they will not add a claim that already exists, and thus deprecating a claim rather than deleting it is a way to prevent a bot from re-adding an erroneous claim. Obviously it is suboptimal, as not only do we retain the erroneous claim, but it is also slower for a human to do when dealing with a bot that makes many errors. Bovlb (talk) 17:16, 19 March 2020 (UTC)
VIAF date with flourished or lived
Hi Magnus Manske,
Imports from viaf with a flourished or lived dateType should not be add as a year date like here, but with a precision of a century. — eru [Talk] [french wiki] 17:42, 19 February 2020 (UTC)
God on Discogs and Songkick
Why this happened? --Obsuser (talk) 21:57, 13 March 2020 (UTC)
- @Obsuser: We seem to be getting some crazy false results from the "mixnmatch" tool. I have reverted some more errors on the same item. See the topic "Blocked" below for more discussion. Bovlb (talk) 17:24, 19 March 2020 (UTC)
URL-encoded Artnet artist IDs
Hi! Such edits cause mass constraint violations, as only decoded IDs are permitted by the format constraint (which seems quite reasonable, given that external identifier links are URL-encoded anyway by the software, so the current link is http://www.artnet.com/artists/ilona-barab%25c3%25a1s/past-auction-results—the “á” is double encoded, notice the %25
s). Please decode these IDs with your bot. (Fortunately the links work even in this double-encoded form, so I can leave these two items in their current state for you as test cases.) Thanks in advance, —Tacsipacsi (talk) 22:49, 13 March 2020 (UTC)
Doppeleinträge bei Findagrave
Hallo Magnus! Ich gehe mal davon aus, dass Du die Diskussionsseite Deines Bots liest und ich mein Anliegen hier loswerden kann. Es geht darum, dass Dein Bot offenbar in Datenelementen Findagrave-IDs ergänzt, aber offenbar ohne Prüfung, ob dort schon solche enthalten sind. Ein aktuelles Beispiel ist dieser Edit. Dies führt aber dann in den Artikeln, hier also bei de:Jonathan Cilley, zu dem sehr unschönen Ergebnis, dass dort nun zu lesen ist: "23303799 Jonathan Cilley in der Datenbank von Find a Grave (englisch)". Verlinkt ist dort der erste Findagrave-Eintrag aus dem Datenelement, aber der zweite wird offenbar als Teil des Namens ausgelesen. Das kann sicher nicht der Weisheit letzter Schluss sein. Vielleicht lässt sich das ja ändern. Liebe Grüße! --Secretary McCord (talk) 15:46, 16 March 2020 (UTC)
- Hi, das ist aber eher ein Problem der de-Vorlage? Beide Find-a-Grave-IDs scheinen korrekt zu sein, und sollten deshalb auch auf Wikidata sein, auch wenn die Property ein "single value constraint" hat, glaube ich. --Magnus Manske (talk) 09:09, 17 March 2020 (UTC)
- Hallo Magnus! Dazu kann ich Dir, wie Du verstehen magst, natürlich auch nichts sagen, da ja die de-Vorlage nun nicht von mir stammt. Ich wende mich an Dich halt in der Hoffnung, dass Du weißt, wie ein solches Problem zu beheben ist. Letztlich finde ich halt schon ärgerlich, wenn dieser Darstellungsfehler in den WP-Artikeln steht, weil es einen Doppeleintrag in Wikidata gibt - und es wäre echt super, wenn es irgendwie zur allgemeinen Zufriedenheit repariert werden könnte. Ich bin auch froh, dass ich es hier bei Dir auf Deutsch darlegen kann, weil mein Schulenglisch zwar für den Hausgebrauch reicht, aber vermutlich nicht für komplexere Angelegenheiten. In den Tiefen des hiesigen Systems kenne ich mich natürlich auch nicht so aus. Liebe Grüße! --Secretary McCord (talk) 09:49, 20 March 2020 (UTC)
Blocked
I have blocked the bot as it seems to be making incorrect and somewhat disruptive errors at a high rate. See this thread for details. Unblocking is at the botop's discretion, once they have had a chance to investigate and can assure us that the problem is fixed. Bovlb (talk) 01:26, 19 March 2020 (UTC)
- Unblocked per botop Bovlb (talk) 14:57, 21 March 2020 (UTC)
old ETIS ID-s
why do you add back old ETIS ID-s [26] ? --WikedKentaur (talk) 08:29, 29 March 2020 (UTC)
- It's automatic from a mix'n'match catalog, but it was deactivated yesterday : Topic:Uw1xiumimpg2gwfb
- You can delete it now. — eru [Talk] [french wiki] 08:59, 29 March 2020 (UTC)
Where can I request that all changes by Reinheitsgebot to add Property:P2953 would get reverted? --WikedKentaur (talk) 08:18, 9 April 2020 (UTC)
- Has there been any development with this? Old ID-s are still a problem. Kruusamägi (talk) 22:03, 22 April 2020 (UTC)
Wrong links to inaturalist.org
Hi, Magnus Manske. Your bot created a lot of wrong claims for Q27046 and it has just created a new one. Q27046 or Carabidae is a family, there is no need to link it with every genus or specie within this family. Track13 (talk) 20:21, 29 March 2020 (UTC)
- The link to then bad match is in the comment, it must to be dissociated here for exemple : https://tools.wmflabs.org/mix-n-match/#/entry/11656413 — eru [Talk] [french wiki] 20:36, 29 March 2020 (UTC)
- Ok, thank you for the explanation, I've removed associations. Track13 (talk) 08:59, 30 March 2020 (UTC)
WORK not HUMAN
In this edit, your bot added an ID from the University of Barcelona to a data item. The ID was identified as that for a creative work, but the item it was matched to is a human. This is the same mistake that was happening before. --EncycloPetey (talk) 14:19, 30 March 2020 (UTC)
Same data item, same problem: [27] --EncycloPetey (talk) 00:56, 2 April 2020 (UTC)
Several errors in French churches
Hello, your BoT generates errors when adding wrong links to the property P3963 (eg error1 or error2 or error3. Pls check it, --FHd (talk) 19:39, 8 April 2020 (UTC)
- Hi, you can cancel it by clicking on the link in the comment, such as https://tools.wmflabs.org/mix-n-match/#/entry/24064435
- I canceled the 3, the first was automatic, but the others were matched by Ayack and AyackBot. — eru [Talk] [french wiki] 21:02, 8 April 2020 (UTC)
Work/edition duplicates
Please see Property_talk:P648#Work/edition_duplicates, I think it was this bot --Bultro (talk) 10:16, 16 April 2020 (UTC)
Wrong IDs
- Markus Ebner (Q119378)
- pnd133528537 was removed https://www.wikidata.org/w/index.php?title=Q119378&diff=1121111897&oldid=1114578553
- your bot adds it again as 133528537 https://www.wikidata.org/w/index.php?title=Q119378&diff=1134298424&oldid=1126950518
- your bot adds wrong CERL https://www.wikidata.org/w/index.php?title=Q119378&diff=1126950518&oldid=1121112021
- Johann Theodor Roscher (Q126802)
MrProperLawAndOrder (talk) 02:47, 10 May 2020 (UTC)
- @MrProperLawAndOrder: I cleaned up the matches for Markus Ebner (Q119378) here and here.
- It was already unmatched for Johann Theodor Roscher (Q126802) here. — eru [Talk] [french wiki] 07:51, 10 May 2020 (UTC)
- eru, this is a systematic problem. Why are the wrong IDs there in the first place? The Deutsche Biographie ID is tied to GND ID, but the items had no GND ID. 690 at https://www.wikidata.org/wiki/Wikidata:Database_reports/Constraint_violations/P7902#Item_P227 and currently 632 https://w.wiki/QUj - maybe all wrong. MrProperLawAndOrder (talk) 00:44, 11 May 2020 (UTC)
- Theodor Koch (Q101652) 1905-1975!!!
- user:Thierry Caro batch-adds pnd136084222 - this is for a person 1786-1863
- your bot adds CERL for a person 1786-1863!!! with comment "#quickstatements; invoked by Mix'n'match:add_person_dates" ... but what is the basis?
MrProperLawAndOrder (talk) 01:05, 11 May 2020 (UTC)
- Hi, both are matched by "Auxiliary data matcher" here and here, starting the at 2020-02-13 08:51:01, after the #temporary_batch_1581486542647 , so because it was already match for Deutsche Biographie, and because of the common GND id for CERL.
- Thierry Caro, what was this batch ? (could be just a apply for waiting match on mix'n'match 1619) — eru [Talk] [french wiki] 06:47, 11 May 2020 (UTC)
- It came from Mix'n'match. Thierry Caro (talk) 06:49, 11 May 2020 (UTC)
- Thanks, so I can't say why the auxiliary data matched, since there is no gnd id on the item.
- Let's notify Magnus Manske. — eru [Talk] [french wiki] 16:36, 11 May 2020 (UTC)
- It came from Mix'n'match. Thierry Caro (talk) 06:49, 11 May 2020 (UTC)
Still wrong CERL values [28] MrProperLawAndOrder (talk) 19:25, 25 May 2020 (UTC) @Eru, Epìdosis: can catalog 1619 be disabled? On top of complete nonsense, the GND IDs can change leading to duplicates etc. DtBio can be linked via tracking GND IDs and SPARQL detecting differences. MrProperLawAndOrder (talk) 19:30, 25 May 2020 (UTC)
- I've asked for deactivation ten days ago. --Epìdosis 19:35, 25 May 2020 (UTC)
- Ibrahim Themo 19 century != Themo 14 century [29] MrProperLawAndOrder (talk) 21:58, 25 May 2020 (UTC)
- Peter Maier volapukologist != Peter Maier +1542 [30] MrProperLawAndOrder (talk) 22:36, 25 May 2020 (UTC)
- Eugen Buri volapukologist in Bern ?? where is the evidence that person is the same as Eugen Buri 1835 bis 1906 Beruf/Funktion Liederkomponist ; Geheimer Regierungsrat https://www.deutsche-biographie.de/135706408.html#ndbcontent? [31] MrProperLawAndOrder (talk) 23:38, 25 May 2020 (UTC)
- Peter Hafner (Q12798618) 1958-2009 != Peter Hafner 1526-1592 https://www.deutsche-biographie.de/pnd141296534.html [32] MrProperLawAndOrder (talk) 00:34, 26 May 2020 (UTC)
- Samuel Kellner (Q12775817) 1789-1855 != Samuel Kellner +1622 https://www.deutsche-biographie.de/pnd14098884X.html [33] MrProperLawAndOrder (talk) 00:51, 26 May 2020 (UTC)
- Dirceu Dresch (Q10266727) 1964- != Dresch died 19th century https://www.deutsche-biographie.de/pnd136470289.html [34] MrProperLawAndOrder (talk) 02:16, 26 May 2020 (UTC)
- Eduardo Schroeder (Q10270774) Brazil != Eduard Schroeder Mitinh. d. Seidenfa. Wilh. Schroeder & Co., Krefeld https://www.deutsche-biographie.de/1082206539.html [35] MrProperLawAndOrder (talk) 02:35, 26 May 2020 (UTC)
- Erwin Bartke (Q10274182) world war II U-boat commander != Erwin II. +1193 https://www.deutsche-biographie.de/136099033.html [36] MrProperLawAndOrder (talk) 02:49, 26 May 2020 (UTC)
False match then false ID then year of death from the false match
Not only mismatches and wrong IDs, but then also adding year of death from mismatch. Can an admin intervene?
[37] - are you going to fix? MrProperLawAndOrder (talk) 20:57, 25 May 2020 (UTC)
- https://www.deutsche-biographie.de/pnd141656042.html erwähnt 1918, gestorben 1925 Beruf/Funktion kommunistischer Funktionär ; Jugendfunktionär der Kommunistischen Internationale ; Schriftsteller
- https://cs.wikipedia.org/wiki/Johann_Leitner Uvádí se jako německý liberál (Německá pokroková strana)
maybe he switched from Deutsche Fortschrittspartei to Jugendfunktionär der Kommunistischen Internationale or the other way around. But no evidence presented. MrProperLawAndOrder (talk) 21:07, 25 May 2020 (UTC)
- Heinrich Michel, volapük Linz vs Rektor Bensheim [38] MrProperLawAndOrder (talk) 22:53, 25 May 2020 (UTC)
Imperfect references
Hi! I've noticed the addition of these references: they are correct, but they should have stated in (P248)Internetowy Polski Słownik Biograficzny (Q96022943), not stated in (P248)property (Q1400881). Could you launch a massive substitution? Thanks, --Epìdosis 09:21, 3 June 2020 (UTC)
- I would also very much like to see a correction. Firstly, property (Q937228) ≠ property (Q1400881). Secondly, for this type of reference, the property stated in (P248) is always used. —putnik 14:14, 22 June 2020 (UTC)
...
Cousin Emmy (Q55720156) wurde von dir erstellt. Muss das einzeln geführt oder kann das eventuell mit Cousin Emmy (Q5178819) gemerged werden? Gruß -- MovieFex (talk) 20:28, 5 June 2020 (UTC)
Pilze Deutschland ID
Hallo Magnus, seit dem letzten Lauf im September 2019 scheinen die Einträge auf pilze-deutschland.de neu organisiert worden zu sein. Kannst Du bitte den Import noch einmal starten, damit die Einträge hier korrigiert werden? Im Moment enden alle Links, die ich geprüft habe, auf einer Fehlerseite. (#quickstatements; mixnmatch:microsync for catalog 2767 (5d73363101767874917373)) Vielen Dank, Matthias.Wolf (talk) 17:15, 10 June 2020 (UTC)
- Erledigt, cleanup läuft. --Magnus Manske (talk) 09:57, 16 June 2020 (UTC)
- Cool - danke! Matthias.Wolf (talk) 19:44, 16 June 2020 (UTC)
Number as label for human
https://www.wikidata.org/w/index.php?title=Q55988514&oldid=722054660 MrProperLawAndOrder (talk) 03:52, 12 June 2020 (UTC)
- "Wird beschrieben in URL" mit Link auf Bach-digital-Eintrag zu Friedrich Justin Bertuch (Q213775) (1747-1822). Der Verlag Beyer und Maring wird in der Quelle nicht erwähnt. --Kolja21 (talk) 00:07, 17 June 2020 (UTC)
GND 4053219-7 not a human
https://www.wikidata.org/w/index.php?title=Q72940909&diff=next&oldid=1041621620 MrProperLawAndOrder (talk) 22:05, 16 June 2020 (UTC)
- Gleiches Problem wie oben. --Kolja21 (talk) 00:09, 17 June 2020 (UTC)
GND 6147624-9 not a human
https://www.wikidata.org/w/index.php?title=Q72941040&diff=1042759680&oldid=1041622710 MrProperLawAndOrder (talk) 22:36, 16 June 2020 (UTC)
Wrong twitter names
Today the bot added a wrong X username (P2002) to many items (at least 180 - there might be more). The value that was set for all of them was "orcid_org". Could this be somehow rolled back? czupirek (talk) 19:59, 29 June 2020 (UTC)
- czupirek I've deleted the statement on the ≈230 items I found. As far as I can tell, none of them had Twitter IDs before. --Matthias Winkelmann (talk) 15:49, 16 September 2020 (UTC)
Wrong date of death
Hi, here a wrong date of death was added for "23.2.1861 Hub (Gem. Busswil, heute Gem. Sirnach), 5.7.1944 St. Gallen" (DOD is 1944, not 1892 which is date of marriage). Best --Hannes Röst (talk) 19:49, 11 September 2020 (UTC)
Unangebrachte Beschreibungen
Special:Permalink/718871311: ??? 𝟙𝟤𝟯𝟺𝐪𝑤𝒆𝓇𝟷𝟮𝟥𝟜𝓺𝔴𝕖𝖗𝟰 (𝗍𝗮𝘭𝙠) 09:12, 24 September 2020 (UTC)
- Uff. 𝟙𝟤𝟯𝟺𝐪𝑤𝒆𝓇𝟷𝟮𝟥𝟜𝓺𝔴𝕖𝖗𝟰 (𝗍𝗮𝘭𝙠) 09:18, 24 September 2020 (UTC)
- @1234qwer1234qwer4: Das ist seit vielen Jahren bekannt und wird so hingenommen, wohl weil die Vorteile die Nachteile deutlich überwiegen. Die Description ist eine nur schwach normierte, interne Sache, da ist die Uniformität zweitrangig. --Emu (talk) 09:28, 24 September 2020 (UTC)
- Soll das ein Witz sein? Es handelt sich bei Wikidata um strukturierte Daten, wie es hier immer so schön wichtigtuerisch heißt. Stattdessen kübelt jetzt ein Bot ellenlange Informationsbrocken ungequirlt in die Beschreibungen? Unfassbar, aber mal wieder typisch... --J.-H. Janßen (talk) 16:09, 10 October 2020 (UTC)
- Der ganze Witz bei den Beschreibungen ist ja, dass sie nicht besonders strukturiert sind. Natürlich wäre es sehr wünschenswert, wenn auch diese manierlich gepflegt wären, aber wie gesagt: Wo ist die Alternative zu Einsatz von Reinheitsgebot? Wir haben mindestens um den Faktor 10 oder 20 zu wenige Mitarbeiter/innen. --Emu (talk) 16:21, 10 October 2020 (UTC)
- @Emu: Man betrachte die Beschreibung im ersten Beispiel: "1866 bis 1943 Beruf/Funktion Bankier ; Geheimer Oberfinanzrat ; Mitglied des Reichsbankdirektoriums Konfession keine Angabe Namensvarianten Budczies, Arnold". Sie lässt sich in folgende Teile zerlegen:
- "1866 bis 1943": Lebensjahre, in der Form nicht zwingend nötig
- "Beruf/Funktion Bankier ; Geheimer Oberfinanzrat ; Mitglied des Reichsbankdirektoriums": Beruf, die einzig wichtige Information
- "Konfession keine Angabe": Angabe zur Konfession, für die Beschreibung nicht relevant
- "Namensvarianten Budczies, Arnold": Namensvarianten, für die Beschreibung irrelevant (stattdessen sollte diese Information als Alias untergebracht werden)
- Wo ist jetzt das Problem, einfach den Teil mit "Beruf/Funktion" davor zu erkennen und den als Beschreibung zu verwenden? Wenn man etwas mehr Zeit hat, kann man noch einstellen, dass " ;" automatisch in "," umgewandelt werden soll, aber das ist wirklich zweitrangig. 𝟙𝟤𝟯𝟺𝐪𝑤𝒆𝓇𝟷𝟮𝟥𝟜𝓺𝔴𝕖𝖗𝟰 (𝗍𝗮𝘭𝙠) 16:14, 10 October 2020 (UTC)
- Na sicher sind die Lebensjahre, Konfessionen und andere Angaben relevant, wenn es um die Identifizierung der Person geht – speziell bei solchen mit häufigen Namen. Sicher, in einer besseren Welt sind diese Daten alle in strukturierten und bequellten Statements hinterlegt, aber das ist halt nicht die Realität in Wikidata. Ein nicht zu kleiner Teil meiner Arbeit besteht darin, Personen auseinanderzudröseln und da hilft mir der Beruf nur teilweise. Das mit dem „einfach so“ ist halt auch so eine Sache, es gibt tausende Mix’n’Merge-Kataloge (aus denen wird das generiert). Möglich ist alles, die Frage ist nur, ob das ein sinnvoller Einsatz der (s.o.) sehr knappen Ressourcen ist. --Emu (talk) 16:21, 10 October 2020 (UTC)
- In der Form finde ich nicht, dass man die Konfession beibehalten sollte. Es gibt einige Leute, die mit QS Beschreibungen hinzufügen, aber ich habe noch nie gesehen, dass mehr als Tätigkeit und Lebensjahre benutzt wurden. Wenn andere automatische Prozesse die Beschreibungen sinnvoll formatiert hinzufügen können, kann Reinheitsgebot das auch. 𝟙𝟤𝟯𝟺𝐪𝑤𝒆𝓇𝟷𝟮𝟥𝟜𝓺𝔴𝕖𝖗𝟰 (𝗍𝗮𝘭𝙠) 16:26, 10 October 2020 (UTC)
- Na sicher sind die Lebensjahre, Konfessionen und andere Angaben relevant, wenn es um die Identifizierung der Person geht – speziell bei solchen mit häufigen Namen. Sicher, in einer besseren Welt sind diese Daten alle in strukturierten und bequellten Statements hinterlegt, aber das ist halt nicht die Realität in Wikidata. Ein nicht zu kleiner Teil meiner Arbeit besteht darin, Personen auseinanderzudröseln und da hilft mir der Beruf nur teilweise. Das mit dem „einfach so“ ist halt auch so eine Sache, es gibt tausende Mix’n’Merge-Kataloge (aus denen wird das generiert). Möglich ist alles, die Frage ist nur, ob das ein sinnvoller Einsatz der (s.o.) sehr knappen Ressourcen ist. --Emu (talk) 16:21, 10 October 2020 (UTC)
- Soll das ein Witz sein? Es handelt sich bei Wikidata um strukturierte Daten, wie es hier immer so schön wichtigtuerisch heißt. Stattdessen kübelt jetzt ein Bot ellenlange Informationsbrocken ungequirlt in die Beschreibungen? Unfassbar, aber mal wieder typisch... --J.-H. Janßen (talk) 16:09, 10 October 2020 (UTC)
- @1234qwer1234qwer4: Das ist seit vielen Jahren bekannt und wird so hingenommen, wohl weil die Vorteile die Nachteile deutlich überwiegen. Die Description ist eine nur schwach normierte, interne Sache, da ist die Uniformität zweitrangig. --Emu (talk) 09:28, 24 September 2020 (UTC)
Das ist ja wohl die Höhe. 𝟙𝟤𝟯𝟺𝐪𝑤𝒆𝓇𝟷𝟮𝟥𝟜𝓺𝔴𝕖𝖗𝟰 (𝗍𝗮𝘭𝙠) 11:12, 25 October 2020 (UTC)
- Die Kritik wäre vielleicht etwas glaubwürdiger, wenn deine folgende Änderung nicht sowohl die Richtlinien für englische Beschreibungen verletzen würde als auch inhaltlich falsch wäre. Der Historiker wird kaum 1951 gestorben sein, wenn er noch fünfzig Jahre später neue Bücher herausbringt. --Emu (talk) 21:05, 25 October 2020 (UTC)
- Ja gut, die voerherige Beschreibung dürfte wohl kaum weniger inhaltlich falsch gewesen sein. Mein Format wurde von einigen Nutzern massenweise angewandt und ist wenigstens lesbarer als der vom Bot eingefügte Salat. 𝟙𝟤𝟯𝟺𝐪𝑤𝒆𝓇𝟷𝟮𝟥𝟜𝓺𝔴𝕖𝖗𝟰 (𝗍𝗮𝘭𝙠) 00:55, 20 August 2022 (UTC)
Bibsys id disappeared
Hi,
since you made an edit on Q76483 (Rainer Maria Rilke) at 23 september the identifier "BIBSYS id" has disappeared. Could you please check that. I don't know if it has happened on other places. Many greetings, --Dick Bos (talk) 13:59, 25 September 2020 (UTC)
Thousands of Sandbox-External identifier (P2536) statements
Hi, looks like you have inserted thousands of Sandbox-External identifier (P2536) statements - https://w.wiki/eH5. Could you remove them? Or maybe replace with what you intended to upload? --Lockal (talk) 08:30, 29 September 2020 (UTC)
Can your bot deprecate the year-only value
Can your bot deprecate the year-only value when a full birth and death date are given? It interferes with my search of people with only a year of birth or death, that I use to decide who to do research on to find the full value. --RAN (talk) 12:53, 29 September 2020 (UTC)
- Deprecation should only be used for false information. The year only value is not false. However, the full date could be set to preferred rank --Emu (talk) 13:35, 29 September 2020 (UTC)
Kapelle mit Ausstattung
Hallo, bei Kapelle mit Ausstattung Q58403778 ist widersprüchlich ob der Ort Tramm in Schleswig-Holstein gemeint ist (wie in "liegt in" angegeben) oder der Ort in Mecklenburg-Vorp (wie in Geokoordinaten angegeben). Bitte fixen. Außerdem ist das evtl. auch noch mit Q58403776 identisch?? -- Gerd Fahrenhorst (talk) 06:56, 26 October 2020 (UTC)
................a Barnstar for you ..............Reinheitsgebot
![]() |
The Tireless Contributor Barnstar | |
For your tireless contributions in wikidata.....Amazing work!! |
– The preceding unsigned comment was added by 190.155.207.57 (talk • contribs) at 00:29, 10 November 2020 (UTC).
Jan van Bergen
Hello, there is something wierd with this edit. The source says he was born or baptized in 1452 but in Wikidata your bot mentioned born on 15 jul 1890. Was that maybe later corrected in the source or was there an issue with the bot? Not to blame you, just wondering (also to find out how to solve this if two different persons were mixed up in one Q-item).- Robotje (talk) 11:20, 26 November 2020 (UTC)
Tomasz Kupś (Q96580252) twitter name
Hi,
there seems to be an error with X username (P2002) external identifier of the Tomasz Kupś (Q96580252) item orcid_org is most likely not a twitter handle of Tomasz Kupś (Q96580252). The problem is replicated with several other items like George L. Hashimoto (Q102281964). Kpjas (talk) 22:50, 6 December 2020 (UTC) This is the second time this happened this year. The items seem to be largely the same as before? --Matthias Winkelmann (talk) 17:12, 20 December 2020 (UTC)
Barak Obama death date
@Magnus Manske: this edit is not good! Please fix the bot's behavior! Kaganer (talk) 16:58, 10 February 2021 (UTC)
- @Luca.favorido : it seems there is a problem on the catalog with living people: 108112751. — eru [Talk] [french wiki] 20:42, 27 February 2021 (UTC)
- @Eru: Could you please help me understand better the problem please? I scraped the site and reported the description of the site, see here. Maybe the issue could be related to the description parser for the dates? Thanks :-) --Luca.favorido (talk) 08:03, 28 February 2021 (UTC)
- Hi Luca.favorido, I'm not sure why but in this case the birth date was set as death date, I think when there's no "-"the parser assumes it's the death date, but I don't know if it's your parser or a date matcher added by Magnus.
- I usually parse the description and separate the important data with ";". Otherwise, I don't think you can do anything.
- ps : I couldn’t find another example. — eru [Talk] [french wiki] 08:31, 28 February 2021 (UTC)
- @Eru: Could you please help me understand better the problem please? I scraped the site and reported the description of the site, see here. Maybe the issue could be related to the description parser for the dates? Thanks :-) --Luca.favorido (talk) 08:03, 28 February 2021 (UTC)
Mistake
Hello Reinheitsgebot, your bot is usually pretty good however here some incorrect data was placed in the Wikidata item. I already made a correction but I thought it is better to inform you because my feedback might help you to improve the used algorithm. - Robotje (talk) 12:02, 27 February 2021 (UTC)
- Hi Robotje, the bot didn’t make a mistake, Ldhank made a bad match here, I deleted it. — eru [Talk] [french wiki] 14:46, 27 February 2021 (UTC)
- Nope, don't blame the messenger Ldhank (talk) 20:05, 27 February 2021 (UTC)
- @robotje. mea culpa. Ldhank (talk) 20:26, 27 February 2021 (UTC)
- Hi Ldhank, reading nlwiki and the site with google translate looks like it's not the same person. — eru [Talk] [french wiki] 20:16, 27 February 2021 (UTC)
- I now see you are right. I make a new element for him. Ldhank (talk) 20:19, 27 February 2021 (UTC)
- Hi Ldhank, reading nlwiki and the site with google translate looks like it's not the same person. — eru [Talk] [french wiki] 20:16, 27 February 2021 (UTC)
- @robotje. mea culpa. Ldhank (talk) 20:26, 27 February 2021 (UTC)
- Nope, don't blame the messenger Ldhank (talk) 20:05, 27 February 2021 (UTC)
change wikidata Isabelle
Could you change the wikidata for Zulu Wikipedia of Isabelle de Charrière? She placed incorrectly at the wrong Isabelle. Isabelle Delobel (Q263928). See https://zu.wikipedia.org/wiki/Isabelle_de_Charri%C3%A8re. Thanks Boss-well63 (talk) 15:20, 6 March 2021 (UTC)
mass revert
Seems like your bot made a repeated error here. It added the same wrong twitter account to multiple different items. I reverted some of it manually but is there an easy way to revert the lot of them? BrokenSegue (talk) 03:19, 23 March 2021 (UTC)
- looks like this isn't the first time it has made this exact error and had to be reverted. see for example. BrokenSegue (talk) 13:10, 23 March 2021 (UTC)
- It seems like someone keeps removing that form from the item that it should be on, consequently, the distinct value constraint wont trigger at all. Obviously, it should trigger at least on the second edit. --- Jura 18:34, 23 March 2021 (UTC)
Question...
Are you the one who listed Andrew Durnford (Q55721071) as a physician? That assertion is backed up by a reference to David Whitten's biography of Durnford.
I haven't read that biography. But Whitten has written several scholarly articles that cover Durnford, and they don't say anything about him being a physician. One of them quoted from letters he wrote to his friend and mentor John McDonogh, that make him sound like a hypochondriac.
As a slave-owner it wouldn't surprise me if Durnford routinely administered folk-remedies to his slaves. I suggest it would be an exagerration to call him a physician. Durnford seems to have been fluent in both written and oral French and English, remarkable for a Black man in the slave-culture South. But, I assume we restrict the term Physician to individuals who have had formal medical training. If I am not mistaken, in the early 19th century an apothecary or surgeon's training could be largely or entirely through apprenticeship, but a Physician, even then, went to College. Geo Swan (talk) 01:40, 18 June 2021 (UTC)
Wikispecies backlog
Hello Reinheitsgebot,
You were doing a great job of clearing, from A to Z, the backlog of taxon pages on Wikispecies not yet linked to Wikidata items - then stopped at Moravekia; I don't know if preparatory work is needed, but if not - please continue!
Thank you, Maculosae tegmine lyncis (talk) 06:29, 15 August 2021 (UTC)
Duplicate journal item created
For some reason a duplicate item was created for a journal. It was there already as Journal of Applied Phycology (Q15755721). I have not seen this before, so perhaps this was a one-off oddity. — Finn Årup Nielsen (fnielsen) (talk) 11:45, 24 August 2021 (UTC)
Invalid username references
Please stop changing URL references into invalid references to a username, as in [39]. A URL path is not a valid username, even if the current formatter URL will resolve it to the original URL; that could change if there is a better way to reference the user specifically. Here it is using GitHub username (P2037) but the bot also creates invalid references using other username properties such as X username (P2002). Thanks. –LiberatorG (talk) 21:24, 14 December 2021 (UTC)
- I'm here for the same reason, this [40] should be a Tweet ID not a Twitter Username. AntisocialRyan (Talk) 01:13, 15 December 2021 (UTC)
Overcorrection of references as part of Mix'n'match cleanup
Hi Magnus, this edit was not correct either. Just because it matches the formatter URL that's not a BBFC reference. Probably a mass revert is in order? - Máté (talk) 06:23, 15 December 2021 (UTC)
- Hi Magnus, Wikidata is still full of malformatted external IDs in references because of this exercise. You should carry out a mass revert. – Máté (talk) 05:16, 28 August 2022 (UTC)
commonchemistry.cas.org
Please take note of this revert (including the reasoning) and skip this type of reference in further removals. --Leyo 08:25, 15 December 2021 (UTC)
Biographical Dictionary of the Czech Lands ID - new entries
Hello, you created dozens of entries like František Heydrich (Q110134439) where you added fully capitalized (upper case) surnames. Can you please fix them in future imports? Thank you. Vojtěch Dostál (talk) 08:02, 19 December 2021 (UTC)
- Should be fixed as of now. --Magnus Manske (talk) 13:26, 21 December 2021 (UTC)
Datenimport
Hallo Magnus, danke für den Import der Kallías ID (P9918). Kannst bei Datensätzen wie Carl Christian Friedrich von Pistorius (Q108812477) die in Kallías erfasste GND ID (P227) nachtragen? Das würde das Auffinden von Dubletten erleichtern. Gruß --Kolja21 (talk) 23:34, 28 December 2021 (UTC)
ausgestorbenes Taxon / fossiles Taxon
Bitte fossil taxon (Q23038290) anstatt extinct taxon (Q98961713) benutzen. --Succu (talk) 21:48, 27 January 2022 (UTC)
Unterschiedliche Objekte für Kirche Bargteheide unklar
Hallo, du hattest für diese Kirche mal fast gleichzeitig die Datenobjekte Q58400269 und Q58400274 angelegt - in der Beschreibung ist nicht klar was der Unterschied sein soll, könntest du das bitte in der Beschreibung ergänzen, oder können die Objekte zusammengelegt werden? -- Gerd Fahrenhorst (talk) 14:50, 13 February 2022 (UTC)
March 2022
Hello, I'm Eihel. I wanted to let you know that one or more of your recent description edits to Q55725080 didn't meet the Wikidata description guidelines. Descriptions should appear as though they were in the middle of a sentence, typically start with a lowercase letter, and written from a neutral point of view. For example, "pop singer" would be a better description than "He is the best pop singer." If you think I made a mistake, or if you have any questions, you can leave me a message on my talk page. and Q55724775, Q55670737 Eihel (talk) 21:29, 6 March 2022 (UTC)
@Magnus Manske: —Eihel (talk) 21:30, 6 March 2022 (UTC)
Janine Bourgnon
Hello Magnus Manske, yesterday your bot created Q111529145. Apart from the information that Janine Bourgnon is/was a female born in 1963, there is nothing there (also no source). Today another bot linked that Q-item to a Commonscat about a tennis player with 1 photo of her in 1960. In that periode there was a Swiss female with that name mentioned in newspapers who participated in tennis tournaments like Wimbledon. It looks like the date of that photo is correct. So either this Q-item is about another Janine Bourgnon or the year of birth in the Q-item is incorrect. Can you explain where you got the year of birth for Janine Bourgnon? - Robotje (talk) 08:41, 9 April 2022 (UTC)
- Hi, these and other recent item creations are based on Commons categories of women. I do not manually check internal consistency of the information from Commons. --Magnus Manske (talk) 13:22, 9 April 2022 (UTC)
- Thanks for your response. So it looks like the error was introduced here. She participated in the 1952 Wimbledon Championships – Girls' singles, so she was born around 1935. I will correct the error in Commons and Wikidata. - Robotje (talk) 08:31, 10 April 2022 (UTC)
Wrong leonore updates
Hello,
Your bot has done many additions of Leonore IDs, like that one: Special:Diff/1627659824. Two remarks:
- the Leonore ID LH/618/19 is obsolete and it has been replaced by LH//618/19
- the expression LH//618/19 was already linked to the wikidata element.
However, it would be interesting to have your bot changing all the old LH/XXX/XX expressions to LH//XXX/XX.
Yours,--Le Petit Chat (talk) 21:27, 28 April 2022 (UTC)
- Hello @Magnus Manske, to prevent having this issue again, could you please also update the Mix'n'Match catalogue 80? All LH\/\d{1,4}\/\d{1,3} needs to be changed to LH\/\/\d{1,4}\/\d{1,3}. Ayack (talk) 10:12, 29 April 2022 (UTC)
- Done. --Magnus Manske (talk) 14:39, 5 July 2022 (UTC)
- Thank you @Magnus Manske, but it seems you doubled all "/" while only the one after LH should have been. For example on https://mix-n-match.toolforge.org/#/entry/4934472 you changed LH/999/58 to LH//999//58 but the right value is LH//999/58. Cf. above regex. Thanks. Ayack (talk) 09:21, 6 July 2022 (UTC)
- That should be fixed now, finally. --Magnus Manske (talk) 09:29, 6 July 2022 (UTC)
- Great, thanks a lot! Ayack (talk) 09:47, 6 July 2022 (UTC)
- That should be fixed now, finally. --Magnus Manske (talk) 09:29, 6 July 2022 (UTC)
- Thank you @Magnus Manske, but it seems you doubled all "/" while only the one after LH should have been. For example on https://mix-n-match.toolforge.org/#/entry/4934472 you changed LH/999/58 to LH//999//58 but the right value is LH//999/58. Cf. above regex. Thanks. Ayack (talk) 09:21, 6 July 2022 (UTC)
- Done. --Magnus Manske (talk) 14:39, 5 July 2022 (UTC)
Jeremiah Bell Jeter: Q52154650 and Q103864551
Hello Reinheitsgebot, there's two of them over here. Since you created the first one, could you take a look? Thank you for your time. Lotje (talk) 14:01, 16 June 2022 (UTC)
Invalid Ishim IDs
@Magnus Manske: Your bot has added several invalid Ishim ID (P3906) values, e.g. Special:Diff/1658946998. Ids starting with a "?" are invalid: https://w.wiki/5KLi It appears that each Hebrew character was replaced with "?". Could you please fix these? Thanks! –LiberatorG (talk) 14:57, 20 June 2022 (UTC)
- I noticed that the edit group added only invalid ids, so I hit the button to undo the edit group. –LiberatorG (talk) 20:36, 20 June 2022 (UTC)
Hi Magnus,
from which job stems this item? Wim9 (talk) 12:14, 29 June 2022 (UTC)
- A Mix'n'match-based new item creator. Requires that no item with that name exists, and that the new item will have some metadata, including external IDs. --Magnus Manske (talk) 14:28, 5 July 2022 (UTC)
This is the herbarium of the University of South Carolina (Q1024426), but in a Wikispecies-manner (lowest possible effort, no sources). Please enrich or delete this item!-- Maschinenraum (talk) 11:44, 17 July 2022 (UTC)
German translations
German translations of works should have their own data items, and not be added to this main data item, as was done here. --EncycloPetey (talk) 17:42, 25 July 2022 (UTC)
These aliases are not German
The bot has added "German" aliases that are not German. [41] --EncycloPetey (talk) 00:26, 28 July 2022 (UTC)
Staatsbürgerschaft Österreich
Hallo, country of citizenship (P27)Austria (Q40) ergibt vor 1918/19 keinen Sinn. Beispiel Anton Hübner (Q97930370). Bitte den Bot entsprechend anpassen. Danke! --Emu (talk) 09:49, 2 August 2022 (UTC)
Who is triggering edits of this account?
Hey Magnus, it is not clear whether all of the edits made by this bot have been triggered by yourself—or whether this bot can make edits invoked by other users via your tools, just as User:QuickStatementsBot did in the past. Given that responsibilities are unclear and considering that there is relatively little interaction with users showing up with issues here, I am currently prepared to block the bot account.
According tot he approved task, one of the reasons to utilize a bot account—high edit rates—is not valid any longer anyways, since bot accounts are rate limited just as all other accounts for quite some time now. The only reason for this account is edit visibility on user watchlists, as much as I am aware—and given its rate of mistakes, this should maybe rather be avoided as well.
So, I think this bot should be considered for immediate retirement at this time. Is there anything which I am missing here? —MisterSynergy (talk) 07:49, 3 August 2022 (UTC)
- @MisterSynergy: If the issue is the visibility of problematic edits, isn’t it better to (ask bureaucrats to) revoke the bot right instead? (By the way, Wikidata bots’ edits are already more visible than those of other wikis’ bots, since edits to entities cannot be marked as minor.) —Tacsipacsi (talk) 21:25, 3 August 2022 (UTC)
- The main issue is the apparent lack of accountability in my opinion. I have no idea whether Magnus alone is triggering edits of this bot, or if any user of (some of) his tools can do so as well. On top of that, I see that many of the complaints reported above stay unanswered which is bad practice for bot operation. Those issues might perhaps be related, but in total I think this does not seem to be an acceptable setting for Wikidata (any longer). —MisterSynergy (talk) 21:34, 3 August 2022 (UTC)
- I see. Yeah, in this case it’s probably really the only solution is to block the bot until Magnus explains/fixes the issues. —Tacsipacsi (talk) 19:16, 4 August 2022 (UTC)
- The main issue is the apparent lack of accountability in my opinion. I have no idea whether Magnus alone is triggering edits of this bot, or if any user of (some of) his tools can do so as well. On top of that, I see that many of the complaints reported above stay unanswered which is bad practice for bot operation. Those issues might perhaps be related, but in total I think this does not seem to be an acceptable setting for Wikidata (any longer). —MisterSynergy (talk) 21:34, 3 August 2022 (UTC)
- The bot is still adding wrong IDs. Example: In 2019 Reinheitsgebot add a CERL Thesaurus ID (P1871) for a person active 1625 to Hermann Busen (Q19285449): German architect (1913-1971). VIAF 77210508 took over the error from Wikidata and now, in the third step, Reinheitsgebot imported the wrong GND from VIAF. As always, no source citation, so you have to hunt back the errors laboriously. --Kolja21 (talk) 13:41, 8 August 2022 (UTC)
- I like to confirm the formentioned statement. This Bot Reinheitsgebot is adding lots of false LCCN-Id's which have not been cross-checked previously. It is very anoying! --Stephan Tournay (talk) 14:27, 13 August 2022 (UTC)
- I came here because this edit seems very suspicious and I would like ask for a source to the claim. --Dipsacus fullonum (talk) 06:02, 18 August 2022 (UTC)
- CERL id has got has got two auxiliary identifiers: VIAF and GND, CERL match also. Through this data the tool searched for q-id with one/two (i'm not sure) id's and matched to it. This is source. It is based on logics: If A = B and C=B then C=A Matlin (talk) 17:22, 31 August 2022 (UTC)
- I like to confirm the formentioned statement. This Bot Reinheitsgebot is adding lots of false LCCN-Id's which have not been cross-checked previously. It is very anoying! --Stephan Tournay (talk) 14:27, 13 August 2022 (UTC)
I have now blocked the account. We need to know who is responsible, and where to complain when something goes wrong. There are way too many unattended complaints on this page. —MisterSynergy (talk) 18:03, 24 August 2022 (UTC)
- I was away for August and didn't see this until now. Reinheitsgebot performs a variety of tasks across tools, which shoukld all be annotated. In fact, the last bot edits show where they come from, in this case the Mix'n'match tool. I can link to the respective entry, or add a reference if that's required. Mix'n'match especially relies on third-party data, which may not always be correct, but is impossible to check automatically. If there are examples of systematically wrong edits, please bring them to my attention. But please unblock the bot so it can continue its work. --Magnus Manske (talk) 15:35, 31 August 2022 (UTC)
- Update: I believe the current policy is that external IDs such as LCCN should not have a reference as they are "their own" reference. If desired, I can add source information as a reference there as well. --Magnus Manske (talk) 15:40, 31 August 2022 (UTC)
- I can also block the addition of LCCNs specifically, if deemed necessary. --Magnus Manske (talk) 15:51, 31 August 2022 (UTC)
- (@Magnus Manske: this is a reply to all of your comments.)
- The main issue right now is that it is not clear who makes edit through this account. It seems that tool users, particularly via Mix'n'match, can do this, but it is all pretty intransparent. Minimum requirement would be that the user in question, who I hope is logged in to the tool with their Wikimedia account, is mentioned in the edit summary. If this is possible, I think we can unblock the bot for now. However, I think we should phase out proxy bot accounts such as this one, since OAuth should make it possible that the tool edits directly under the account of the tool user.
- Re. references: if the LCCN entry specifically linked to a Wikidata item and this is how you found the connection, then it would be valid as a "self" reference. Otherwise not, and particularly not just to put something there into the references section of newly added claims. If you have nothing else to add, it might be valuable to add the Mix'n'match entry as a reference, in order to track data provenance.
- I don't think we need to block specific content such as LCCN identifiers at this point. Actually, I like the approach in general even though it is not perfect. It just needs to be much clearer to other users how this thing actually works, and where to complain if something goes wrong.
- —MisterSynergy (talk) 17:11, 31 August 2022 (UTC)
- Ah, I see! So there are several scripts, in Mix'n'match as well as other tools. that run on their own (on a cronjob etc) that use Reinheitsgebot to edit. So technically it's me doing these edits. There is one case (not Mix'n'match) where data from GND can be added to an item on a specific request by a user, but other than that, it's my scripts using this bot. All other user interactions run through OAuth these days, and the GND one will come soon, and is low-volume anyway. --Magnus Manske (talk) 19:57, 31 August 2022 (UTC)
- Thanks for the explanation. If I understand correctly, this is the background worker that adds claims which are displayed as "Auxiliary data matcher", "Automatic name/date matcher" (or others?) in Mix'n'match and other tools—is this correct (except for this GND thing that you mentioned)? Does it process user-generated catalogs with edits?
- If this is indeed correct (please confirm!), I am prepared to unblock the bot—but I'll leave two requests here:
- It would really be helpful if there was more information about data provenance accessible, for each and every edit the bot makes. Particularly Mix'n'match is quite difficult to navigate so that many (including me) fail to understand what has happened when the bot edits (incorrectly). A deeplink to the mix'n'match entry or similar content in other tools in the edit summary would certainly be helpful.
- There are *plenty* of complaints above on this talk page and on several other venues that have not been addressed. The amount is not unusual given the bot has done almost 19.000.000 edits, but I think we need to do better here and actually respond. The perception shared by many editors that this could be a proxy bot account comes in part from the fact that you apparently do not seem to care about what happens here on this page; this bot in particular has pretty poor reputation in the community for that reason. I understand that you have a plethora of requests related all your various activities here on Wikimedia projects—but please give it a serious try. We do expect this from other bot operators as well.
- —MisterSynergy (talk) 21:10, 31 August 2022 (UTC)
- +1 and please try to avoid creating conflated items (by using for a new item only a citable source instead of a VIAF cluster).
- —Kolja21 (talk) 23:16, 31 August 2022 (UTC)
- It's already unblocked so that I do not forget to do this before the day ends. —MisterSynergy (talk) 23:12, 31 August 2022 (UTC)
- @Magnus Manske: [42]. No explicit source. The edit group contains a bunch of authority control databases, so no implicit source either. And it’s a string containing several comma-separated names, which should have been added as separate statements. Or rather not, since they aren’t pseudonyms, they are his real name in different variations how librarians ended up recording it, a pseudonym is a different thing. —Tacsipacsi (talk) 07:30, 9 September 2022 (UTC)
- It's already unblocked so that I do not forget to do this before the day ends. —MisterSynergy (talk) 23:12, 31 August 2022 (UTC)
- Ah, I see! So there are several scripts, in Mix'n'match as well as other tools. that run on their own (on a cronjob etc) that use Reinheitsgebot to edit. So technically it's me doing these edits. There is one case (not Mix'n'match) where data from GND can be added to an item on a specific request by a user, but other than that, it's my scripts using this bot. All other user interactions run through OAuth these days, and the GND one will come soon, and is low-volume anyway. --Magnus Manske (talk) 19:57, 31 August 2022 (UTC)
- (@Magnus Manske: this is a reply to all of your comments.)
German edition in translation
In this edit, the "full work" of an ancient Greek play is listed as being a modern German translation. Editions should always be given their own data items. --EncycloPetey (talk) 16:22, 1 September 2022 (UTC)
Adding undifferentiated identifiers
Kaum entsperrt, fügt der Bot wieder Tn = Platzhalter ein.[43] (Erläuterung: de:Hilfe:GND#Personen.) Immherhin jetzt mit der Angabe: https://mix-n-match.toolforge.org/#/entry/31007026, wo behauptet wird, dass die fehlerhafte GND von der Spanischen Nationalbibliothek (BNE) stammt. Dort findet sie sich allerdings nicht. Magnus hat als Erläuterung zu dem Bot hinzugefügt: "Many edits depend on third-party data (eg via Mix'n'match), so individual errors are to be expected", was dem Namen "Reinheitsgebot" Hohn spricht. @MisterSynergy: Imho sollte der Bot gesperrt bleiben, solange grundlegende Fehler nicht ausgeschlossen sind. -- Kolja21 (talk) 00:44, 2 September 2022 (UTC)
- Es gibt zurzeit einige hundert Tn-Identifikatoren, die von User:Reinheitsgebot in den letzten Wochen ergänzt wurden.
- Die Sache mit den Tns ist allerdings eher komplex und IMO nicht unbedingt in einem solchen Bot zu berücksichtigen. Eher können wir einen weiteren Bot aufsetzen, der Tns dann zeitnah/täglich richtig auszeichnet (missbilligen und entsprechenden Qualifikator dran tun). Ich habe den Quellcode für einen solchen Bot längst fertig; soll ich da einen Task beantragen und das mal aufsetzen?
- Viele Grüße! —MisterSynergy (talk) 08:01, 2 September 2022 (UTC)
- Bitte herzlich darum! --Emu (talk) 09:07, 2 September 2022 (UTC)
- Wikidata:Requests for permissions/Bot/MsynBot 9 —MisterSynergy (talk) 09:42, 2 September 2022 (UTC)
- Please block this bot until MsynBot 9 starts working. Reinheitsgebot is again flooding Wikidata with wrong IDs. Reference: Empty (no citable source, no information given). History: Notes like #quickstatements; via https://mix-n-match.toolforge.org/#/entry/31017756 ; details; invoked by mixnmatch:aux2wd. This is mix-n-guess playing hide-and-seek. See also de:Kategorie:Wikipedia:GND in Wikipedia fehlt, in Wikidata vorhanden. --Kolja21 (talk) 23:33, 16 September 2022 (UTC)
- Wikidata:Requests for permissions/Bot/MsynBot 9 —MisterSynergy (talk) 09:42, 2 September 2022 (UTC)
- Bitte herzlich darum! --Emu (talk) 09:07, 2 September 2022 (UTC)
Adding shortened IDs with mix-n-match
For Herbert Greth (Q100383468) the bot added two IDs both with the note: #quickstatements; via https://mix-n-match.toolforge.org/#/entry/15291362. GND 12593002X ("Datensatz ist nicht mehr Bestandteil der Gemeinsamen Normdatei") and GND 12593002 without the X (empty page). This error has occurred frequently. --Kolja21 (talk) 23:09, 17 September 2022 (UTC)
- Same here: GND 10581105 without the X. Catalogus Professorum Lipsiensis: GND 10581105X --Kolja21 (talk) 23:22, 17 September 2022 (UTC)
- Remove it from MnM. This match is present from 10-2020 XDDDD It's one click (a few seconds). Matlin (talk) 18:53, 8 October 2022 (UTC)
Adding VIAF identifiers of a Greek person to the MALE entity
For male (Q6581097)
- the bot added two identifiers - that seem incorrect at 22:01, 24 September 2022
- VIAF ID (P214): 16939651, #quickstatements; via https://mix-n-match.toolforge.org/#/entry/142958414
- Library of Congress authority ID (P244): no2004080454, #quickstatements; via https://mix-n-match.toolforge.org/#/entry/142958414
- Subsequently TWO Other people/bots "Built" on the error by updating a VIAF redirect and adding more incorrect identifiers based on the VIAF:
- 00:28, 29 September 2022 Matlin talk contribs 65,598 bytes +346 Created claim: National Library of Greece ID (P3348): 180546, batch #99501 Tag: quickstatements
- 01:22, 30 September 2022 KrBot talk contribs 65,612 bytes +14 Set a claim value: VIAF ID (P214): 174149066616265602140
The incorrect edits: [44]
- Remove wrong MnM entry: https://mix-n-match.toolforge.org/#/entry/117517715 Matlin (talk) 18:56, 8 October 2022 (UTC)
Empty items
Hi Magnus,
why are you creating items like Julie Deter (Q114404085)? Besides of the name they are completely empty.-- Kop Buster (talk) 09:04, 17 October 2022 (UTC)
- See whatlinkshere. --Magnus Manske (talk) 10:34, 17 October 2022 (UTC)
- But is it really a good idea to create items about people we know absolutely nothing about? How is it better than using author name string (P2093), apart from potentially introducing conflations? Did you take any steps to prevent conflations? If yes, you probably used some kind of identifier – why didn’t you record it? —Tacsipacsi (talk) 09:42, 18 October 2022 (UTC)
- Scopus author ID (P1153) is already present in Association between the DQA MHC class II gene and Puumala virus infection in Myodes glareolus, the bank vole (Q33450316) source page. If there is a risk, it is risk of duplicates of authors IDs, not conflations (especially not koljaisms - rare conflations in VIAF). Matlin (talk) 20:17, 20 October 2022 (UTC)
- Where? I don’t see it; the author statements link, and used to link, only to PubMed publication ID (P698), which is an ID for works, not for people. And even if I’m just blind, the last question stands: why didn’t the bot add the Scopus ID to the people’s items? —Tacsipacsi (talk) 09:06, 21 October 2022 (UTC)
- [45] -> Julie Deter (below the title) -> View in Scopus (left, under the name). It would be good if the bot would add this ID right away. Matlin (talk) 19:42, 24 October 2022 (UTC)
- Where? I don’t see it; the author statements link, and used to link, only to PubMed publication ID (P698), which is an ID for works, not for people. And even if I’m just blind, the last question stands: why didn’t the bot add the Scopus ID to the people’s items? —Tacsipacsi (talk) 09:06, 21 October 2022 (UTC)
- Scopus author ID (P1153) is already present in Association between the DQA MHC class II gene and Puumala virus infection in Myodes glareolus, the bank vole (Q33450316) source page. If there is a risk, it is risk of duplicates of authors IDs, not conflations (especially not koljaisms - rare conflations in VIAF). Matlin (talk) 20:17, 20 October 2022 (UTC)
- But is it really a good idea to create items about people we know absolutely nothing about? How is it better than using author name string (P2093), apart from potentially introducing conflations? Did you take any steps to prevent conflations? If yes, you probably used some kind of identifier – why didn’t you record it? —Tacsipacsi (talk) 09:42, 18 October 2022 (UTC)
Bad author match
Hi Magnus - can you explain this edit? How is 'R.T. Thompson' matched to 'Paul Thompson'?? Those names are not compatible. ArthurPSmith (talk) 19:21, 17 October 2022 (UTC)
More bad matching - please stop and revise!
See this edit. You have matched "M. Zhang" as "Liang Zhang", "P. Schmidt" as "Brian Schmidt", "B. Smith" as "Alexis Smith", "J. Steinlechner" as "Sebastian Steinlechner", "K. A. Thorne" as "Kip S. Thorne", "M. C. Edwards" as "Philip Edwards" and MANY more bogus matches. I don't have any idea whether the matches where the names are close are even correct - there are many papers like this with multiple authors having the same last name. I've seen papers with over 10 people with the same last name ("Martin", or "Wang"). Please stop and let's review what you're actually doing here before proceeding further! ArthurPSmith (talk) 19:28, 20 October 2022 (UTC)
- I thought that came from link between ORCID (or another scientist ID) and WD. For example, Patricia Schmidt (Q61824609) has Inspire-Hep ID, where "Multi-messenger..." is stated. There is problem with reconciliation between short name and WD ID, but probably authors are really linked with the article itself. It need further investigation. Matlin (talk) 20:11, 20 October 2022 (UTC)
- It doesn't look to me like the bot is looking at ORCID or any other ID in making these matches, just looking at co-author graphs perhaps? That is totally unreliable when you have very large author lists like this! ArthurPSmith (talk) 13:20, 21 October 2022 (UTC)
- After a little research, I think, that your point of view is closer to the truth. Magnus published a source of the bot. At the beginning of the code there is a description, which describes authors reconciliation process. It is not based on any auxiliary ID's. Therefore I think, that work of this bot should be probably fixed. Matlin (talk) 13:59, 24 October 2022 (UTC)
- It doesn't look to me like the bot is looking at ORCID or any other ID in making these matches, just looking at co-author graphs perhaps? That is totally unreliable when you have very large author lists like this! ArthurPSmith (talk) 13:20, 21 October 2022 (UTC)
Peter Lewis (b. 1945) = Tom Lewis (1901-1988)
Peter Lewis (Q7175449): American musician was born 1945. Reinheitsgebot add a person with an other name b. 1901 (as usual without citing the source) and poor Peter Lewis has been declared dead. @Magnus Manske: Since there are no regulations limiting the errors of this bot could you please at least change en:Reinheitsgebot to a more appropriate name? --Kolja21 (talk) 03:26, 23 October 2022 (UTC)
- Yet again not deleted(+ [46]) match from MnM and referring to an outdated situation (currently the bot indicates a match in MnM). Plus semi-personal attack. @Emu:, based on your previous steps, you should block both users of this discussion XD . Matlin (talk) 14:07, 24 October 2022 (UTC)
- @Matlin I’m not sure what you are trying to achieve. If you want to be blocked from all namespaces, as you have previously told us, please find another admin to comply with your request. --Emu (talk) 14:36, 24 October 2022 (UTC)
- You are writing partially off topic. I want to remind and show you that there is no reason for the blocking that you used, otherwise you would have blocked Reinheitsgebot bot already, due to Magnus' lack of response (and stopped trusting Kolja21's statements). By the way (and partially off topic): If you would unblock me by acknowledging your mistake, i.e. amicably, it would be better for me than to submit appropriate applications (which will be submitted anyway, the delay is due to my health and personal matters). If I have a choice (a possibility), I choose the ability to edit Wikidata - but not on a fictitious rules, but just like the other users. Matlin (talk) 20:27, 25 October 2022 (UTC)
- @Matlin I’m not sure what you are trying to achieve. If you want to be blocked from all namespaces, as you have previously told us, please find another admin to comply with your request. --Emu (talk) 14:36, 24 October 2022 (UTC)
The birth and death dates are wrong according to [47]. The dates in CERL are just floruit dates and I have deleted them from the entry. Cheers, --Ecelan (talk) 16:46, 19 November 2022 (UTC)
State for persons
Please undo mass additions of P17 in human items from "Sanctions database" (like this edit) - this violates constraints (humans should have country of citizenship (P27), not country (P17)). Infovarius (talk) 11:03, 23 November 2022 (UTC)
- In database there is "country", not "country of citizenship", so if a statement is sourced - it is quite good. No delete is required to sourced statements, but seting deprecated rank with the appropriate qualifier. Matlin (talk) 14:48, 25 November 2022 (UTC)
- But this is a modeling issue, not a data issue. country (P17) isn’t suitable for human (Q5) --Emu (talk) 20:04, 2 December 2022 (UTC)
- I support Infovarius. Adding of this property generates a great deal of violations. At the same time we don't get any extra information, because country doublicates citizensships in all items. Ksc~ruwiki (talk) 17:55, 2 December 2022 (UTC)
- @Matlin, Emu, Infovarius: This happened again see Yan Permenas Mandenas (Q96987684).
- Reinheitsgebot please do not add P17 (Country) to humans. Germartin1 (talk) 02:16, 21 February 2023 (UTC)
- Especially because OpenSanctions initally imported the data from Wikidata in the first place. Germartin1 (talk) 04:54, 21 February 2023 (UTC)
P2348
This property time period (P2348) should not be added to data items that are a literary work (Q7725634). The "works" are not an event and did not occur at a single point in time.
Since your bot was used to import such data incorrectly [48], is there a way to prevent it, with a warning about the correct properties to select from? Such as date of first performance (P1191), inception (P571), or publication date (P577)? --EncycloPetey (talk) 20:55, 27 November 2022 (UTC)
Same problem with:
These are all literary works, and "time period" does not specify whether it is a date pertaining to the origin of the work, the publication of the work, or the time period in which the events described in the literary work are set. --EncycloPetey (talk) 21:50, 2 December 2022 (UTC)
- Could you please have your bot stop making these edits over and over? --EncycloPetey (talk) 17:53, 4 December 2022 (UTC)
- The property documentation does not prohibit the addition of such statements. They are not "wrong". Secondly, you must first ask Wolnelektury.pl to remove them (in example page you can clearly see: "epoka: starożytność"). If you do not know a given foreign language and do not know what an source says, ask the appropriate person. Thirdly, if a statement is added correctly and sourced (and it is), and yet for some reason it is incorrect, then the statement should be not removed, [but should be given an deprecated rank with qualifiers. It refrains from adding the statement again. This isn't the first time you've done this. Even so, I don't see any reason (or there are very weak reasons) why this should be deprecated. Matlin (talk) 20:22, 12 December 2022 (UTC)
I'm assuming this is the meteorologist, but can't find the paper you've created him with SourceMD for.--88-Henry (talk) 10:32, 4 January 2023 (UTC)
taxon name added to an item that is not a taxon
I caught this edit using mix-and-match and your bot, where a "taxon name" was added to an item that is not a taxon. The data item is an orthographic variant (i.e. a misspelling). The sources linked point out that it is indeed a misspelling, is not accepted, and is invalid. Under the botanical Code, an item that is invalid cannot be a taxon name. --EncycloPetey (talk) 05:28, 2 February 2023 (UTC)
German translation full works
I've seen several edits like this one recent;y made using your bot. A "full work" at Gutenberg is an edition of a work, and should be placed on a data item for that edition, and not on the data item for the main work. For one, the language of the "full work" in this case is German, but the original was written in ancient Greek, so it is not only an edition, but a translation into another language. --EncycloPetey (talk) 18:11, 4 February 2023 (UTC)
Also, I see in the Mix-n-Match entry that this is identified as a "written work", which is wrong. It's an edition, not a work. --EncycloPetey (talk) 18:25, 4 February 2023 (UTC)
Same problem, but on another work --EncycloPetey (talk) 14:31, 6 February 2023 (UTC)
Francis Ford Coppola not dead yet
See Special:Diff/1826328612 and Talk:Q56094. Bovlb (talk) 22:04, 4 February 2023 (UTC)
Country for persons
Hi. Thanks for your contributions. You've added countries to persons with OpenSanctions as source (see here). This raises warnings, because there is another statement already for citizenship. Could you either look at the warning or possibly revert those specific edits? Dajasj (talk) 08:20, 22 February 2023 (UTC)
- Ah, I see this was already raised... Dajasj (talk) 08:22, 22 February 2023 (UTC)
Reinheitsgebot-Import von 30. Januar 2023 - Baudenkmäler in Hessen - Dubletten
Hallo, siehe bitte Topic:Xd8p9xcy9s67ufa7. Vielen Dank! M2k~dewiki (talk) 13:37, 26 February 2023 (UTC)
Adding death for wrong person
This bot today added a death date for Jonas Hellborg (Q762693) based on https://www.musik-sammler.de/artist/jonas-hellborg-shawn-lane/. It's just that there's 2 people mentioned in that article and the death date is related to the other guy. Hjart (talk) 10:22, 4 March 2023 (UTC)
Bot
Hi. Your bot added a death date with "source" [52]. There is no such information in this source [53]. Elfhelm (talk) 21:35, 10 March 2023 (UTC) (same for example here).
Orth. var.
In this edit, your bot inferred that ITIS target was a taxon name, but the ITIS entry specifically says that it is "not accepted - orthographic variant (misspelling)", which means that this is not the name of a taxon. It is only a misspelling, and the ITIS entry exists merely as a pointer because the misspelling is frequent enough to note. Under the Code, misspellings are never the names of taxa. --EncycloPetey (talk) 16:53, 11 March 2023 (UTC)
I tried to remove the linkage at toolforge, but that page will not load for me. --EncycloPetey (talk) 16:55, 11 March 2023 (UTC)
Your bot made the same incorrect edit again. Please keep this from happening. --EncycloPetey (talk) 05:31, 30 March 2023 (UTC)
Reinheitsgebot running amok again with author names
this edit matched "Bruce Allen" with "G. Allen", "Jonathan Anderson" with "S. B. Anderson", "Carl Blair" with "D. G. Blair", "R Gustafson" with "E. K. Gustafson", "Andrew M. Hopkins" with "P. Hopkins", "Bryn Jones" with "D. I. Jones", etc. etc. Please stop until you have a better name-matching algorithm, and particularly avoid any articles with more than a few dozen P50 or P2093 statements. I have reverted thousands of these bad replacements when I run across them, but I thought you had stopped doing this late in 2022; now I find new ones. ArthurPSmith (talk) 19:50, 23 March 2023 (UTC)
- I've reverted 16 of the roughly 100 author name replacements that were done in this edit - based on given name mismatches like the above. I also looked through some other recent edits of this sort by Reinheitsgebot and it seemed like at least 1/3 of the edits had bad name matches like this, affecting something like 10-50% of the names edited in each case. This is a really high error rate! ArthurPSmith (talk) 21:07, 23 March 2023 (UTC)
- I spot checked some more that the bot ran last night, and those look ok (no mismatching author names in the ones I checked). Has something changed? Or perhaps it depends on the particular domain of research it's working with, what the author collaboration patterns and lists are like... Anyway, I think it's generally doing good work but it hits areas where it's making really bad decisions that take a lot of work to fix. ArthurPSmith (talk) 13:42, 24 March 2023 (UTC)
- Nothing changed. Not sure why the malfunction. I'll turn it off for now. --Magnus Manske (talk) 13:56, 24 March 2023 (UTC)
- Please also undo the old edits: Here an other "SourceMD author name string resolver script" error. --Kolja21 (talk) 23:45, 2 May 2023 (UTC)
- Nothing changed. Not sure why the malfunction. I'll turn it off for now. --Magnus Manske (talk) 13:56, 24 March 2023 (UTC)
Vandalism?
Tuomas M. S. Lehtonen (Q16983483): Difference between revisions - Wikidata? Person is alive. 176.72.27.171 14:00, 26 March 2023 (UTC) Also Eero Vaara is alive.(https://research.aalto.fi/fi/persons/eero-vaara ) Why does the bot do this? -176.72.27.171 14:23, 26 March 2023 (UTC)
"full work", but in German
Here's another instance where a "full work available at URL" was added incorrectly. I have removed the mix-n-match link.
In this instance a 1922 German translation was added as the "full work" of a book by an American that was originally published in English. --EncycloPetey (talk) 18:20, 26 March 2023 (UTC)
n.b.: Looks like I forgot to add the link before. I moved the link target to its own data item, as it should be, and linked that item as an edition of the work of literature. --EncycloPetey (talk) 05:33, 30 March 2023 (UTC)
Q58398526
Hallo, bei diesem Eintrag für ein Haus in Brokdorf ist ein Bild aus Flensburg angegeben, Versehen oder Absicht? Außerdem unverständlicher Kommentar "Vgl. Bootshafen der Marineschule Mürwik" - was ist damit gemeint, beim Bootshafen steht nichts von Brokdorf?! -- Gerd Fahrenhorst (talk) 11:38, 4 April 2023 (UTC)
- Da keine Antowrt kam, habe ich die zweifelhaften Informationen gelöscht. -- Gerd Fahrenhorst (talk) 07:57, 7 April 2023 (UTC)
Conflation: Bot legt vermischte Datenobjekte an
Oxford-DNB-Eintrag: Richardson, John (1647–c. 1725), see Richardson, William.
- Reinheitsgebot hat 2014 daraus die Person John Richardson mit der Beschreibung "William Richardson (1698–1775), antiquary and college head" geschaffen (Q18527512), die in den folgenden Jahren zu "John Richardson, Antiquar, geboren 1698" wurde und zu entsprechender Verwirrung geführt hat.
Diese Fälle sollte man versuchen aufzuspüren. Für einen Filter wären zwei Informationen nötig:
- Datenobjekt wurde von einem Bot erstellt
- Quelle, in diesem Fall Oxford Dictionary of National Biography ID (P1415), ist eine Weiterleitung
--Kolja21 (talk) 20:06, 5 April 2023 (UTC)
- Of course another person had to remove wrong match 😂😂😂 Matlin (talk) 21:06, 5 April 2023 (UTC)
- @Matlin: We all know how many errors you made by blindly using Mix'n'match. That's why your account has been blocked. The case of John Richardson has been solve. The questions is, how we can find all the other conflated items created over the last ten years. --Kolja21 (talk) 23:45, 5 April 2023 (UTC)
- @Solidest: The problem with the cross references might interest you, since you left this note. --Kolja21 (talk) 05:46, 6 April 2023 (UTC)
- I was sorting duplicated ODNB IDs for individual persons yesterday. I removed those IDs where the dates differed significantly from the dates on WD. The remaining 4 duplicated IDs whose dates are very close are left here: https://mix-n-match.toolforge.org/#/sync/1 . And after a few ID moves into correct Qitems, 26 IDs got free: https://mix-n-match.toolforge.org/#/list/1/unmatched (everything on the list from Nockalls Cottingham to Elizabeth Campbell) - these people don't seem to have pages on WD yet. Most of these entries on MnM have incorrect descriptions - they were either updated on the Oxford site at some point, or the original database scraping was incorrect and the descriptions of entities with the similar names of different people turned out to be the same. John Richardson was among them.
- MnM has 3 entries with identical "Richardson, William (1698-1775)" descriptions:
- I am not sure we should keep Q18527512 as p31=conflation. I guess it's better to just remove incorrect IDs from the item. And merge John Richardson (Q117414735) back into this again. While MnM entries can be unlinked here: https://mix-n-match.toolforge.org/?#/search/Q18527512. The problem here is really just a mismatch between the description on MnM and the actual text on the Oxford site. So I think this is the reason why the bot created this entity conflated. Btw, I am planning to update all the descriptions on MnM for this catalogue later today. Solidest (talk) 09:34, 6 April 2023 (UTC)
- The conflated item has been deleted. Now:
- John Richardson (Q117414735) (1647- abt. 1725), fellow of Emmanuel College Cambridge = cross-reference
- William Richardson (Q8017549) (1698–1775), antiquary and college head [nephew of no. 1]
- Robert Richardson (Q18527420) (bap. 1731, d. 1781), rector of Wallington, Hertfordshire [son of no. 2] = cross-reference
- --Kolja21 (talk) 21:06, 9 April 2023 (UTC)
- The conflated item has been deleted. Now:
- @Solidest: The problem with the cross references might interest you, since you left this note. --Kolja21 (talk) 05:46, 6 April 2023 (UTC)
- @Matlin: We all know how many errors you made by blindly using Mix'n'match. That's why your account has been blocked. The case of John Richardson has been solve. The questions is, how we can find all the other conflated items created over the last ten years. --Kolja21 (talk) 23:45, 5 April 2023 (UTC)
Automatic name/date matcher
The Automatic name/date matcher seems to identify people based on life dates (years) only:
- Umar Wirahadikusumah (Q76306) (1924-2003) is, according to Mix'n'match/Reinheitsgebot, identical with
- Alfons J. Keller (Q117454637) (1924-2003)
One person was born in March, the other person in April. They have nothing in common except that they're both men. Reinheitsgebot first add a wrong SIKART ID and since nobody noticed two month later a wrong GND. --Kolja21 (talk) 05:31, 6 April 2023 (UTC)
Ignoring basic informations
Reinheitsgebot still ignores basic informations:
- Gregório Alonso Aparicio (Q47003711) (b. 1982) is obviously not identical with Gregorio Alonso (b. 1974) = Gregorio Alonso (Q117460360)
- The bot should at least check the year of birth
- All edits are still made without naming a source
@Pintoch: FYI. --Kolja21 (talk) 00:37, 7 April 2023 (UTC)
- @Kolja21: not sure what I have to do with this? − Pintoch (talk) 20:53, 11 April 2023 (UTC)
- Hi Pintoch, you are linked in the history of this edit: https://editgroups.toolforge.org/b/CB/642e7aa2a2ea0186446761/ (contact the author) so I thought it might interest you. --Kolja21 (talk) Kolja21 (talk) 01:46, 12 April 2023 (UTC)
Ambigous item: supposed source offline
There seems to be a problem with imports from an unknown source using Semantic Scholar author ID (P4012):
- Q64855521: Attila Rácz, researcher, ID 48694451 (offline)
No further information, no retrieval date: "We can't find the page you're looking for." --Kolja21 (talk) 01:43, 9 April 2023 (UTC)
Done Thanks to User:Epìdosis, see Wikidata:Bot requests#Request to massdelete non-notable Semantic Scholar authors (2023-04-16). --Kolja21 (talk) 16:42, 19 April 2023 (UTC)
Format constraints
The harvesting of identifiers by Reinheitsgebot is reaching a new level:
- Library of the Senate, Rome (Q27056071): GND IT-RM0427
- Biblioteca musicale del Conservatorio “Santa Cecilia” (Q113535627): GND IT-RM0266
- Biblioteca privata Moscatelli (Q117192191): GND IT-PI0402
- Netherlands Music Institute (Q1919182): GND Q1919182
It is obvious that these IDs do not exist. @Bovlb, Emu, MisterSynergy: Block this bot till someone feels responsible for these edits? --Kolja21 (talk) 12:35, 16 April 2023 (UTC)
- User:Podonnellfm imported those into Mix'n'match. Maybe use one minute of your time to check where the bad data comes from, instead of shooting the messenger? I have cleaned the catalog from the obvious wrong ones, but I don't want to delete all of the IDs because some of them are clearly correct. --Magnus Manske (talk) 08:58, 20 April 2023 (UTC)
- Hi, this is my first experience using Mix'n'match. The data comes from the RISM file of institution authorities. After I imported the catalog it wasn't clear to me how to clean up or in any way manipulate it. I'm happy to do so but have to learn how... Maybe someone can delete them, then I can clean them up in Open Refine and re-import? Podonnellfm (talk) 09:25, 20 April 2023 (UTC)
- I think it would be helpful (again) if these edits were made under the user's account via Oauth, instead of the bot account. Mix'n'match is a super complicated tool with processes that many users do not understand, so the source of errors often can't be identified properly. —MisterSynergy (talk) 11:35, 20 April 2023 (UTC)
References in edits to Database of British and Irish Hills items
In December there was a series of edits in which OS grid reference (P613) statements were added to items with DoBIH Number (P6515) identifiers (example: https://www.wikidata.org/w/index.php?title=Q106710623&diff=prev&oldid=1782949773). The main statements look correct, but topographic prominence (P2660) values have been added as DoBIH Number (P6515) to the reference instead of the identifiers. Also the references are the same source so should probably be one reference. Peter James (talk) 13:50, 19 April 2023 (UTC)
a taxon is not a rank
https://www.wikidata.org/w/index.php?title=Q45203581&diff=prev&oldid=1878637548 Succu (talk) 19:34, 19 April 2023 (UTC)
Dublette
https://www.wikidata.org/w/index.php?title=Q102272&diff=1541058777&oldid=1541058733&diffmode=source Oursana (talk) 10:05, 29 April 2023 (UTC)
SourceMD batch from 6. March apparently wrong
SourceMD batch from 6. March apparently wrong. Jimmy Wales have been wrongly linked:
- https://www.wikidata.org/w/index.php?title=Q60511239&diff=next&oldid=1795810421
- https://www.wikidata.org/w/index.php?title=Q58041587&diff=prev&oldid=1846275664
- See more at https://scholia.toolforge.org/author/Q181
— Finn Årup Nielsen (fnielsen) (talk) 19:45, 11 May 2023 (UTC)
GND-Kennung (P227): 7.23E+21
Looks like vandalism: GND-Kennung (P227): 7.23E+21. No source given. History: #quickstatements; via https://mix-n-match.toolforge.org/#/entry/151357380 by Reinheitsgebot (Q167255). The bot is still unable to do a simple check to see if an ID is valid. --Kolja21 (talk) 22:47, 30 May 2023 (UTC)
Mistaken edit
This man is alive, but this edit said the opposite. TaronjaSatsuma (talk) 07:11, 31 May 2023 (UTC)
Error detection
Reinheitsgebot added in incorrect VIAF identifiers. See for example: John Inglett-Fortescue (Q26265809). See: Wikidata:Database reports/items with P569 greater than P570 for one of the error detection searches I monitor. I just corrected eight errors. The problem cascaded as a User added in incorrect birth and death dates based on your incorrect VIAF entries. RAN (talk) 16:07, 14 June 2023 (UTC)
- It's quite interesting, that no one has deleted wrong matches: https://mix-n-match.toolforge.org/#/entry/98693218 https://mix-n-match.toolforge.org/#/entry/63534239 https://mix-n-match.toolforge.org/#/entry/53265261 https://mix-n-match.toolforge.org/#/entry/111294683 https://mix-n-match.toolforge.org/#/entry/141850181 https://mix-n-match.toolforge.org/#/entry/5746723 Matlin (talk) 18:28, 17 June 2023 (UTC)
Dubletten
Mylnikov (Q4309377) is the same as newly created Mylnikov (Q119458509). Please correct, thank you! Anna Pavlova
- Nope. First is about surname, second about wikimedia disambig page. Matlin (talk) 11:48, 17 June 2023 (UTC)
Done I've merge Q119458509 with Mylnikov (Q1759942): Wikimedia disambiguation page. --Kolja21 (talk) 17:14, 17 June 2023 (UTC)