Property talk:P735
Documentation
first name or another given name of this person; values used with the property should not link disambiguations nor family names
Description | The given name or the first name of a person. See WikiProject Names for more information. This only applies to persons who have a first name. This provides a machine readable version of the label for person items. See also: family name (P734). | ||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Represents | given name (Q202444) | ||||||||||||||||
Data type | Item | ||||||||||||||||
Domain | According to this template:
persons
According to statements in the property:
When possible, data should only be stored as statementsperson (Q215627), fictional character (Q95074), legendary figure (Q13002315), character that may or may not be fictional (Q21070598), alter ego (Q201662), human (Q5), mascot character (Q386208), pet (Q39201), personification (Q207174), fictional entity (Q14897293), individual animal (Q26401003), mythical creature (Q2239243), human fetus (Q26513) or processional giant (Q340069) | ||||||||||||||||
Allowed values | given name (Q202444) (note: this should be moved to the property statements) | ||||||||||||||||
Usage notes | values used with the property shouldn't link disambiguations nor family names | ||||||||||||||||
Example | Jimmy Wales (Q181) → Jimmy (Q4166211) Donal (Q1239626) | ||||||||||||||||
Formatter URL | http://www.namepedia.org/en/firstname/$1 http://www.namepedia.org/de/firstname/$1 | ||||||||||||||||
Tracking: usage | Category:Pages using Wikidata property P735 (Q29454937) | ||||||||||||||||
See also | family name (P734), Roman praenomen (P2358), name (P2561), given name version for other gender (P1560), Vietnamese middle name (P8500), name in native language (P1559) | ||||||||||||||||
Lists |
| ||||||||||||||||
Proposal discussion | Proposal discussion | ||||||||||||||||
Current uses |
| ||||||||||||||||
Search for values |
List of violations of this constraint: Database reports/Constraint violations/P735#Conflicts with P31, SPARQL
List of violations of this constraint: Database reports/Constraint violations/P735#Type Q215627, Q95074, Q13002315, Q21070598, Q201662, Q5, Q386208, Q39201, Q207174, Q14897293, Q26401003, Q2239243, Q26513, Q340069, SPARQL
List of violations of this constraint: Database reports/Constraint violations/P735#Value type Q202444, Q49614, SPARQL
List of violations of this constraint: Database reports/Constraint violations/P735#allowed qualifiers, SPARQL
List of violations of this constraint: Database reports/Constraint violations/P735#Scope, SPARQL
List of violations of this constraint: Database reports/Constraint violations/P735#Conflicts with P131, search, SPARQL
List of violations of this constraint: Database reports/Constraint violations/P735#Conflicts with P625, SPARQL
List of violations of this constraint: Database reports/Constraint violations/P735#Conflicts with P31, search, SPARQL
List of violations of this constraint: Database reports/Constraint violations/P735#Conflicts with P1566, search, SPARQL
List of violations of this constraint: Database reports/Constraint violations/P735#Target required claim P1705, SPARQL, SPARQL (by value)
List of violations of this constraint: Database reports/Constraint violations/P735#Target required claim P282, SPARQL, SPARQL (by value)
List of violations of this constraint: Database reports/Constraint violations/P735#Entity types
Two possibilities: (1) change P735 value if incorrect item is used, (2) move P214 statement from value if property applied incorrectly, (3) .. (Help)
Violations query:
SELECT DISTINCT ?item { [] wdt:P735 ?item . ?item wdt:P214 [] . }
List of this constraint violations: Database reports/Complex constraint violations/P735#Value should not have VIAF ID
todo: TBD, e.g. change rank to deprecated (Help)
Violations query:
SELECT ?item ?value WHERE { ?item (p:P31/ps:P31) wd:Q1747829; wdt:P735 ?value. FILTER(!(wikibase:isSomeValue(?value))) }
List of this constraint violations: Database reports/Complex constraint violations/P735#Items with notname wouldn't have a value
Value N/A (Q929804) will be automatically replaced to value NOVALUE. Testing: TODO list |
Value Israel (Q801) will be automatically replaced to value Israel (Q19819746). Testing: TODO list |
|
|
Statistics
[edit]Name | Usage |
---|---|
44668 | John (Q4925477) |
26021 | William (Q12344159) |
20728 | Robert (Q4927937) |
18282 | David (Q18057751) |
17758 | Thomas (Q16428906) |
17611 | James (Q677191) |
17447 | Charles (Q2958359) |
16094 | George (Q15921732) |
15947 | Peter (Q2793400) |
15481 | Paul (Q4925623) |
14081 | Michael (Q4927524) |
13722 | José (Q2190619) |
13456 | Richard (Q1249148) |
12000 | Joseph (Q471788) |
11580 | Jan (Q12173670) |
10943 | Hans (Q632842) |
10108 | Karl (Q15731830) |
9948 | Henry (Q1158477) |
9656 | Jean (Q7521081) |
9360 | Johann (Q11122389) |
9020 | Ivan (Q830350) |
8515 | Edward (Q278835) |
8422 | Daniel (Q14516546) |
8148 | Juan (Q475210) |
7926 | Pierre (Q2093070) |
7764 | Vladimir (Q2253934) |
7714 | Nicholas (Q42771) |
7654 | Louis (Q2897866) |
7492 | Frank (Q220546) |
7482 | Walter (Q499249) |
7390 | Carl (Q2529610) |
7361 | Alexander (Q923) |
7297 | Martin (Q18002399) |
6667 | Antonio (Q7141520) |
6602 | Albert (Q577011) |
6556 | Maria (Q325872) |
6514 | Christian (Q18001597) |
6235 | Arthur (Q1778168) |
6166 | Mark (Q13610143) |
6046 | Friedrich (Q14038597) |
5824 | Alfred (Q3480335) |
5729 | Wilhelm (Q11027623) |
5697 | Carlos (Q364753) |
5610 | Franz (Q4925932) |
5603 | Manuel (Q11113719) |
5447 | Josef (Q15730712) |
5237 | Heinrich (Q2018484) |
5214 | Anna (Q666578) |
5186 | Tom (Q3354498) |
5134 | Samuel (Q629347) |
5112 | Georg (Q1985538) |
5068 | Otto (Q18029644) |
4918 | Francisco (Q7201446) |
4903 | Johannes (Q2117521) |
4834 | Bill (Q18245781) |
4832 | Giovanni (Q1158906) |
4818 | Mike (Q361309) |
4768 | Jacques (Q941049) |
4718 | Jack (Q1159009) |
4659 | Ernst (Q292691) |
4581 | Andrew (Q18042461) |
4579 | François (Q1166840) |
4572 | Sergei (Q12902079) |
4561 | Harry (Q668885) |
4469 | Chris (Q339346) |
4355 | Luis (Q2598693) |
4334 | Jim (Q15868042) |
4287 | Joe (Q17862013) |
4220 | Hermann (Q1158570) |
4186 | Patrick (Q18002623) |
4171 | Rudolf (Q18028491) |
4148 | Anton (Q5401576) |
4079 | Francis (Q1441346) |
4026 | Henri (Q1605665) |
4024 | Johan (Q10989273) |
4000 | Adam (Q347181) |
3997 | Bob (Q18105736) |
3964 | André (Q6298851) |
3923 | Giuseppe (Q15720844) |
3889 | Steve (Q18336566) |
3873 | Max (Q16422172) |
3832 | Stephen (Q4927100) |
3820 | Mary (Q734578) |
3817 | Roger (Q4925304) |
3768 | Brian (Q15930574) |
3604 | Andreas (Q4926263) |
3518 | Eric (Q12788459) |
3504 | Mario (Q3362622) |
3483 | Pedro (Q15897419) |
3452 | Michel (Q14626626) |
3399 | Herbert (Q4926833) |
3392 | Francesco (Q2268455) |
3328 | Simon (Q4117588) |
3290 | Victor (Q539581) |
3283 | Alan (Q294833) |
3282 | Frederick (Q3273004) |
3271 | Bernard (Q14649171) |
3270 | Kevin (Q605834) |
3235 | Marie (Q632104) |
3229 | Stefan (Q4927128) |
3176 | Anthony (Q12241622) |
3170 | Wolfgang (Q2589157) |
3142 | Mohammed (Q5837762) |
3142 | Tony (Q15238167) |
3137 | August (Q370731) |
3084 | Stanisław (Q19801658) |
3035 | Georges (Q15921764) |
3028 | Miguel (Q15620295) |
2988 | Marc (Q17520944) |
2978 | Fernando (Q18115751) |
2975 | Fred (Q913073) |
2971 | Christopher (Q1084472) |
2967 | Jorge (Q14346236) |
2932 | Anne (Q564684) |
2918 | Philip (Q827311) |
2894 | Yakov (Q544124) |
2868 | René (Q2624417) |
2838 | Fritz (Q1158596) |
2835 | Maurice (Q1472321) |
2829 | Claude (Q17523984) |
2798 | Roberto (Q15905580) |
2775 | Alex (Q13258171) |
2769 | Gabriel (Q4925914) |
2751 | Benjamin (Q4926472) |
2749 | Philippe (Q15727189) |
2739 | Erik (Q750186) |
2724 | Bruno (Q1874605) |
2662 | Emil (Q989320) |
2659 | Werner (Q676823) |
2651 | Kurt (Q1158489) |
2637 | Ludwig (Q14159020) |
2634 | Aleksandr (Q17501806) |
2606 | Gustav (Q746076) |
2602 | Ernest (Q595105) |
2588 | Marco (Q17520955) |
2576 | Alberto (Q16799105) |
2550 | Carlo (Q4205426) |
2546 | Jonathan (Q1158394) |
2538 | Barbara (Q153957) |
2526 | J. (Q19803507) |
2525 | Ali (Q17865069) |
2525 | Józef (Q19688659) |
2521 | Elizabeth (Q385468) |
2516 | Hugo (Q1242555) |
2505 | Adolf (Q18145837) |
2489 | Scott (Q2260734) |
2461 | Tim (Q1369663) |
2432 | Andrzej (Q14941830) |
2432 | Dave (Q16275316) |
2426 | Luigi (Q4245067) |
2421 | Nicolas (Q7029481) |
2419 | Gary (Q1820274) |
2414 | Gerhard (Q7996169) |
2389 | Ian (Q18336315) |
2365 | João (Q16916120) |
2344 | Ben (Q816407) |
2338 | Donald (Q13422248) |
2293 | Julius (Q1102114) |
2271 | Klaus (Q15635230) |
2262 | Lars (Q15635262) |
2261 | Rafael (Q16424000) |
2252 | Eduard (Q18031774) |
2244 | Dan (Q2487038) |
2235 | Raymond (Q2127728) |
2202 | Josep (Q16429820) |
2199 | Matthew (Q4927231) |
2199 | Harold (Q14647205) |
2196 | Hugh (Q839387) |
2161 | Heinz (Q11682369) |
2160 | Antoine (Q15235674) |
2151 | María (Q18088640) |
2136 | Don (Q16275555) |
2135 | Ralph (Q18156180) |
2124 | Marcel (Q4927589) |
2122 | Viktor (Q16430022) |
2108 | Jerzy (Q17536409) |
2099 | Matt (Q1158973) |
2087 | Jeff (Q18609696) |
2075 | Ken (Q2645657) |
1994 | Eduardo (Q18130894) |
1979 | Jason (Q2630093) |
1978 | Ferdinand (Q3276175) |
1965 | Antoni (Q17480899) |
1960 | Margaret (Q4963612) |
1955 | Sam (Q13858350) |
1932 | Jimmy (Q4166211) |
1931 | Guy (Q1159023) |
1927 | Anders (Q8843357) |
1913 | Erich (Q15277196) |
1904 | Ahmad (Q406294) |
1887 | Alexandre (Q16002466) |
1886 | Bernhard (Q221978) |
1881 | Leo (Q606764) |
1877 | Boris (Q666112) |
1876 | Oscar (Q1951683) |
1872 | Pavel (Q1242655) |
1862 | Ron (Q2165388) |
1857 | Piotr (Q15897269) |
1849 | Andy (Q13627273) |
1845 | Sarah (Q18201513) |
|
Surnames:
Personal names:
|
Navigation
[edit]Discussion
[edit]two given names
[edit]Thanks for creating this. Question: How to handle John F. Kennedy (Q9696)? Littledogboy (talk) 22:46, 31 July 2013 (UTC)
- Use series ordinal (P1545) to indicate the order of a person's given names. Senator2029 17:32, 2 November 2019 (UTC)
Mess
[edit]Having the datatype as Item creates a lot of mess when it comes to translation versus transliteration. For example, English first name Joseph transliterated as Джозеф in Ukrainian, but Joseph gives just a Ukrainian variant for it which is Йосип. So Joseph Stalin is Йосип (because of the original Russian Иосиф), but Joseph Priestley is Джозеф. So currently I don't see a way to use this property in any consistent way in Wikipedia templates. --DixonD (talk) 16:53, 29 January 2014 (UTC)
- That's not such a big surprise, considering that the property was added with neither consensus nor majority vote, and disregarding serious objections against its datatype. Just don't use it and let it rot out...--Shlomo (talk) 18:20, 18 August 2014 (UTC)
- Or you could just use several values of P735 for an item, with "prefered value" for the name in the person's original language and devalued values for all the others first names generally associated with them… Like every other case where there is several informations? Like Francis of Assisi (Q676555)… --Harmonia Amanda (talk) 05:23, 3 October 2014 (UTC)
- Francis of Assisi (Q676555) is a good example why the property is useless with item datatype. As for now, it has 8 values, but if you look through connected pages, you can see about 50 other language versions of his name, that could be added. Now what's the point of having links to >50 items about the name, if the person has only one name? And what's the point of having link to item, whith Czech / English / Italian / etc. labels "Franz" if the person is not called "Franz" in any of these languages (and a similar link for every of the other >49 languages)?--Shlomo (talk) 18:52, 4 October 2014 (UTC)
- Or you could just use several values of P735 for an item, with "prefered value" for the name in the person's original language and devalued values for all the others first names generally associated with them… Like every other case where there is several informations? Like Francis of Assisi (Q676555)… --Harmonia Amanda (talk) 05:23, 3 October 2014 (UTC)
- This also is a problem with fictional characters. Kaladin is not a name that I've ever heard given to a real person in any language, but it's the name of a character from The Stormlight Archive. But this fucking constraint is in the way, so I can't actually use the property the way that it's meant to be used. There's a Similar issue with the father, mother, and parent properties, which have (at least I think) number constraints, even though there are fictional characters (such as James Holden, who has 9) with more than two parents! Even outside of fiction, there are some places (even in the US) where you can have more than two *legal, non-step* parents. There are actually a number of properties that have constraints which screw things up when considering fictional or mythological characters. LiftedStarfish (talk) 08:24, 2 November 2021 (UTC)
The above project aims to improve and expand the use of this property. --- Jura 03:09, 7 October 2014 (UTC)
- @Jura1: Instead of using Q18336849 do you think it would be possible to use has characteristic (P1552):given name (Q202444) on classes requiring a given name? --Micru (talk) 08:33, 22 October 2014 (UTC)
- The idea is to check if only "P31:Q5" or fictional characters/humans use P735 and not others. Before this could be done with a "Type: person (Q215627)" constraint, but as you changed Q5, this broke this and most related checks.
- Don't hesitate to add other constraints above and see what results this gives. Afterwards we can keep the one(s) that work(s) best. --- Jura 11:03, 22 October 2014 (UTC)
Middle names
[edit]How do we handle middle names? They're not on the same level as given (first) names.
Notified participants of WikiProject Names --AmaryllisGardener talk 22:58, 29 December 2014 (UTC)
- Usually we use given name (P735) for all the given names, and the first name as "Prefered value". -Ash Crow (talk) 00:32, 30 December 2014 (UTC)
- Ah, makes sense. Thanks! :) (BTW, your edit was marked as a bot edit, mistake?) --AmaryllisGardener talk 00:51, 30 December 2014 (UTC)
- Oops. I had the flood flag because I did mass edits with my account. I removed it. -Ash Crow (talk) 01:55, 30 December 2014 (UTC)
- Ah, makes sense. Thanks! :) (BTW, your edit was marked as a bot edit, mistake?) --AmaryllisGardener talk 00:51, 30 December 2014 (UTC)
I advise against using the "preferred rank method" because the attribute of the name we want to indicate isn't that the first name is "better" than the middle name. Rather, it is the order in which each names appears. Therefore, use series ordinal (P1545). Senator2029 17:28, 2 November 2019 (UTC)
Expansion to house, animals and other objects
[edit]As we have named after (P138) to describe the naming of objects, I undid this change, attempting to expand the use of this property for first names to items that are not for about people (mainly items with instance of (P31) = human (Q5)). --- Jura 03:00, 24 March 2015 (UTC)
- It seems to me that you are misinterpreting the difference between the two properties. The basic difference is not about the items they can be applied to, but the items which can be used as target for the properties. This property (P735) links a subject to its name. Items to be linked are to be about that name. Corresponding Wikipedia articles could for example describe the origin or the frequency of the name. On Wikidata:WikiProject Names there is a list of items for given names.
- On the other hand, named after (P138) links an item to an "entity or event that inspired the subject's name, or namesake (in at least one language)" (my emphasis). So named after (P138) should not link to name-items. It would be strange to say that a name is its own inspiration. And 'namesake' means: another subject with the same name, not the name itself. In yet other words, the difference is between named after and bearing a name. So it is wrong if you add for example to Dolly (Q171433) that it is 'named after' Dolly (Q1235797) while in fact the sheep was called Dolly and named after Dolly Parton (Q180453).
- By the way, the word "subject" in the description op P138 can apply very well to humans. After all, people are often named after other people. And in my opinion it is awfully anthropocentric to talk about 'houses, animals and other objects'. Bever (talk) 04:20, 24 March 2015 (UTC)
- And note that the constraint restraint below the changed one does say "Exceptions are possible as rare values may exist" while both restraints have a large overlap (both mention human (Q5)). So the restraints contradict each other. Bever (talk) 04:32, 24 March 2015 (UTC)
- This property is only for first names. Animals don't have first names. You'd need to use another property for sheep, etc. --- Jura 05:30, 27 March 2015 (UTC)
- What do you mean with "Animals don't have first names"? Whether you like it or not, animals are given name. Sometimes these names would never be used for humans, but when a cat is called Tom (Tom Cat (Q1839152)) or a bull is called Herman (Herman the Bull (Q2241459)), this is exactly the same name as for example Tom Cruise (Q37079) or Herman van Veen (Q503717).
- Also it is not quite correct to speak about 'first names', as Hungarians, Chinese and several others have the names the other way around. If somebody has only a given name, this would automatically the first name, I suppose, but I wondered if your statement "This property is only for first names" implies that there should be a last name as well. What about Thomas the Apostle (Q43669) or Herman (Q1428284) (humans with only one name)?
- Further it is not consequent that you allow fictional characters but do not allow animals: this already means that you cannot calculate the use of names for real people with counting this property only. Animals are even more real than fictional characters. And what about mixed characters like the main characters of Miss Minoes (Q1049558) and The Golden Ass (Q1044767), who transform between human and animal during the story? Bever (talk) 19:40, 27 March 2015 (UTC)
- If there other items where it shouldn't be, of course it should be removed as well. --- Jura 23:47, 1 April 2015 (UTC)
- What items are you talking about? Animals of fictitious characters? I did not mention "items where it shouldn't be", the examples I gave where it could be (a few Hermans and Toms) do not have this property at present, except for the apostle.
- In fact you did not respond at all to what I said. Why do you want to exclude animals while including fictitious and mythical characters? Bever (talk) 00:38, 21 April 2015 (UTC)
- If there other items where it shouldn't be, of course it should be removed as well. --- Jura 23:47, 1 April 2015 (UTC)
- This property is only for first names. Animals don't have first names. You'd need to use another property for sheep, etc. --- Jura 05:30, 27 March 2015 (UTC)
- And note that the constraint restraint below the changed one does say "Exceptions are possible as rare values may exist" while both restraints have a large overlap (both mention human (Q5)). So the restraints contradict each other. Bever (talk) 04:32, 24 March 2015 (UTC)
I see there was debate about this issue in 2015, and I'm not really understanding the outcome.... Nevertheless I've just made this edit [1] which gives a "Given name" to an individual known Turkey (animal). This throws a constraint violation, as Zelda (Q8068703), the turkey, isn't a person. We have several thousand individually known animals on Wikidata, and a good portion of them have non-unique given names. Currently only a handful have a "given name" statement but it seems obvious to me at least that this property should be allowed to apply to them too. Wittylama (talk) 18:07, 24 November 2017 (UTC)
Qualifiers constraints
[edit]I don't understand very well why we are taking as valid instance of (P31), language of work or name (P407), point in time (P585), series ordinal (P1545), nickname (P1449). From my point of view:
- instance of (P31): Should be the same as the value type and it should be specified there.
- language of work or name (P407): Should be specified as a property of the value.
- point in time (P585): I don't any situation where a person should have a name only in one specific moment.
- series ordinal (P1545): I don't understand the logic of this one... Maybe a list of all the people with that name?
- nickname (P1449): This should be used as a property of the item itself, not as a qualifier of the name.
I think about removing them from the constraints, but first I would like to know what other people think about it. Specifically Jura1, who added (if I checked it correctly) all of them in the last months. -- Agabi10 (talk) 03:14, 20 October 2015 (UTC)
- Possible: I added P1449 (and others) just to avoid that they fail the constraint reports ;) I might have cleaned-up a few others that were also in use.
- I think P407 is being used for items like Q676555#P735 with many language variations included, P1545 to indicate the order of given names (first name, second name, etc.)
- Personally, I used P31 for things like Q460300#P735.
- Maybe you can find a sample for P585 where someone found it useful. BTW P585 indicates what was correct at given point in time, not necessarily only at a specific moment. --- Jura 06:12, 20 October 2015 (UTC)
- @Jura1: OK, fair enough. I didn't think on the possibility of a person whose name has variations in different languages. I was only thinking about people having one name (or combination of names) which is the same in all the languages and where language of work or name (P407) specifies the name of origin, no the language in which the elements given name is that one. Using series ordinal (P1545) for specifying the order of the names of a person also makes sense, because even if we can know having the names separately and know which is the order it is less obvious for a machine.
- For the use you are using as an example of instance of (P31) personally I don't like it... I think it should be a better way to specify that kind of things. From my point of view instance of is a property that should only be used as a property, not as a qualifier. Also it looks weird having "novalue" with a qualifier specifying its value. Shouldn't be more logical using it as instance of Q460300 instead?
- Finally related to the use of point in time (P585) I think that in that property is semantically inferred that something happened in an specific moment, for wider ranges we have start time (P580), end time (P582), earliest date (P1319) or latest date (P1326), depending of the meaning we want to give to the specific date. -- Agabi10 (talk) 12:25, 20 October 2015 (UTC)
- Q19968968 isn't a given name. It's just an explanation why there is no given name.
- point in time (P585): Not really, have a look at the English label ("time and date .. a statement was true"). If it wasn't this qualifier, we would need another qualifier for the same. --- Jura 16:53, 21 October 2015 (UTC)
Ranks
[edit]Some items used the "depreciated" rank for first names that aren't generally used. → Help:Ranking#Deprecated rank
Personally, I think the given name that is generally used can have preferred rank, but the others should have normal rank.
I changed a few items to this. --- Jura 15:35, 12 November 2015 (UTC)
- Jura you forgot to mention the possibility that the person changed their name legally, this is something usual with gender change operations and things like that for example, but probably there are more cases. In this case at least from my point of view the old name should be a deprecated statement because it is not like they are not generally used, they are more like at the moment incorrect statements that were correct in the past. I would put those as deprecated with end time (P582) as qualifiers. -- Agabi10 (talk) 19:27, 12 November 2015 (UTC)
- I didn't fix any of those, but I'm not sure if statement depreciation would be appropriate. You could use a qualifier to specify end and start dates for names that changed.
- A given name could have depreciated rank if it's one that keeps getting misreported ("Paul" for "Peter Jones"). --- Jura 07:00, 13 November 2015 (UTC)
- I agree with Jura. "Deprecated" to me means "wrong", not "obsolete". If a name was changes, use a end time (P582) qualifier and use "preferred" for the current name. --Srittau (talk) 01:53, 19 June 2016 (UTC)
Baptismal name
[edit]What sholud I do if a person has different baptismal names (given right after birth in hospital) and different legal names (used in official documents)? Marciooo (talk) 16:57, 29 September 2016 (UTC)
- Generally, I add only the currently used given name in this property, but you could include all (past and present given names).
- Ideally, you'd add the full name at birth (including surname) with birth name (P1477) and subsequent ones with name in native language (P1559).
--- Jura 17:08, 29 September 2016 (UTC)- Alas, according to the description of name in native language (P1559) you can't use it exclusively for names in the person's native language. For name in other languages acquired later (e.g. in connection with migration, conversion, assimilation...) there is no suitable property now, AFAIK.
- birth name (P1477) is not a solution for Marciooo's problem either, since you can't tell, which part of the value (=full name) is the baptismal name.--Shlomo (talk) 07:43, 30 September 2016 (UTC)
- I add it in P735, and I use P31 as a qualifier with Christian name (Q2396897) as value. --Harmonia Amanda (talk) 09:34, 30 September 2016 (UTC)
Multiple given names: ranks and series ordinal (P1545)
[edit]There was a recent discussion in the project chat that got me thinking about our given names. Until now, I've always thought we gave the first given name a preferred rank and add the other names as "normal" statements. This has some problems though, as some values don't show up in the query service for example. I've recently discovered that the approach with series ordinal (P1545) is way much better, the rank should be normal for all of them though. Should we set this as the new approach and how many items should be fixed?
Notified participants of WikiProject Names
Sjoerd de Bruin (talk) 14:04, 4 October 2016 (UTC)
- Actually, I use both methods: I list all the given names with the qualifier series ordinal (P1545), and I mark the actual used given name with preferred rank when it's not the first one (in France, until recent times, the given name in use daily was the last and not the first, etc. --Harmonia Amanda (talk) 14:09, 4 October 2016 (UTC)
- In the Netherlands, a lot of people also have a "roepnaam" what seems close to "given name in use daily". Someone called Hendrik (Q1158181) could have Henk (Q1605227) as "roepnaam" for example. I'm still wondering how we can include information like that. I mostly add the birth name (P1477) and short name (P1813), but how can I indicate that Henk (Q1605227) is the "roepnaam" of someone? Sjoerd de Bruin (talk) 14:20, 4 October 2016 (UTC)
The first name is David, but in real life he was called Theo (see, for example, his FRS bio article, Q47481409). His publications were always signed D. T. N. Williamson; David appears only in a few formal occasions. Which name should be in P735? Retired electrician (talk) 11:27, 10 February 2018 (UTC)
- All four. Maybe the English label should be "Theo Williamson" as well, but oddly the enwiki article title isn't using that either.
--- Jura 11:31, 10 February 2018 (UTC)- For the "Theo" part, see my comments above. Think it's time to suggest a property for that. Sjoerd de Bruin (talk) 15:37, 10 February 2018 (UTC)
Pseudonyms
[edit]Does this property apply for pen names, pseudonyms, or other aliases? For Mark Twain (Q7245), his given name is Samuel (Q629347), and "Mark" is not a given name. The pen name "Mark Twain" is a nautical phrase, not a given name and surname. Similarly George Eliot (Q131333) was a woman, and her given name is not George. --EncycloPetey (talk) 21:07, 13 June 2018 (UTC)
- There is already a related discussion going on at Wikidata:Requests_for_deletions#Q17018059. Q17018059 is the nickname of some person. But in my opinion, nicknames should not be described by given name (P735) but by nickname (P1449). Similar with pseudonyms, we should use pseudonym (P742) instead of given name (P735). --Pasleim (talk) 22:11, 13 June 2018 (UTC)
allowed qualifiers constraint
[edit]I get the message "alternate names is not a valid qualifier for given name". I wonder why. Medieval names were often in variations. In this case the names Machteld, Mechteld, Mechtild and Mechtildis (which are variations themselves of the name Mathilda/Matilda/Mathilde/Matilde) are all used for the same person. See Matilda of Guelders (Q2031173). --HRvO (talk) 20:44, 24 July 2018 (UTC)
Female given name
[edit]Hi. I have been around some time but do not know how to create a new "statement". Can someone kindly make "Resmiye (given name)" please? BTW where should I ask this kind of help? Thanks. --E4024 (talk) 01:50, 24 May 2019 (UTC)
- Hi @E4024: are you searching for the item Resmiye (Q64008871)? Regards --NicoScribe (talk) 20:36, 8 June 2019 (UTC)
- I was. --E4024 (talk) 12:28, 26 June 2019 (UTC)
- Hi NicoScribe. Indeed I managed to make that one, but I am really without much experience in WD. I also wanted to make "Buğra (female given name)" and "Buğra (male given name)" but to no avail. There is "Buğra (family name)" and that confuses me. Can you do the job for me please? Thanks. --E4024 (talk) 14:58, 5 July 2019 (UTC)
- @E4024: try Special:NewItem. I think you did just fine at Akıncı (Q65040663). I will check afterwards and fix anything that is needed. --- Jura 15:27, 5 July 2019 (UTC)
- Thanks, I just discovered that gadget. Until now I was only using the "search" button and filling in. --E4024 (talk) 15:36, 5 July 2019 (UTC)
- @E4024: I am sorry: when writing my first message above, I had not seen that you were the creator of the item Resmiye (Q64008871).
- I am not an expert in items about names, but I think that Resmiye (Q64008871), Buğra (Q65042221) and Akıncı (Q65040663) are good. You could perhaps use "instance of = female given name" instead of "instance of = given name".
- Moreover, if you create more items about names, other statements can be useful:
- When a family name already exists and is identical to the created given name, you could use family name identical to this given name (P1533), cf. example in Fatima (Q221839). So, for Buğra (Q65042221), you could add "family name identical to this given name (P1533) = Buğra (Q5003635)".
- When a disambiguation page already exists and is identical to the created given name, you could use different from (P1889) with given name has to use a different item than disambiguation pages (Q23765057), both in your item and in the disambiguation page item, cf. example in Fatima (Q221839) + Fatima (Q223848).
- When a disambiguation page already exists and is identical to the created family name, you could use different from (P1889) with family name has to use a different item than disambiguation page (Q27924673), both in your item and in the disambiguation page item, cf. example in Fatima (Q37505043) + Fatima (Q223848).
- --NicoScribe (talk) 17:49, 5 July 2019 (UTC)
- Thanks, I just discovered that gadget. Until now I was only using the "search" button and filling in. --E4024 (talk) 15:36, 5 July 2019 (UTC)
- @E4024: try Special:NewItem. I think you did just fine at Akıncı (Q65040663). I will check afterwards and fix anything that is needed. --- Jura 15:27, 5 July 2019 (UTC)
What to do when given name is actually not a given name
[edit]Hi,
We have a constraint which says that the given name should be a given name. In 99% of the cases, there is absolutely no problem.
But sometimes, people have for given name, the name of an other people. For instance, Manuel II of Portugal (Q154308) has for 11th given name Francisco de Assis so right now we have Manuel II of Portugal (Q154308)given name (P735)Francis of Assisi (Q676555) (added by Sarah Layton) which kind of make sens but break the constraint.
How should we proceed in such cases?
Cdlt, VIGNERON (talk) 10:51, 25 June 2019 (UTC)
- Simple: Remove the incorrect value and replace it with an actual item for the name. If there isn't one yet, one can be created. (This is assuming that it's actually a given name, and not just the names "Francisco" and "de Assis".) --Yair rand (talk) 17:32, 25 June 2019 (UTC)
- I was a bit skeptic to create a name for only one person ; then Francisco de Assis (Q66659044) was created by Pablo Busatto and used on Francisco de Assis Bezerra de Menezes (Q66659034) so all is good now. Cheers, VIGNERON (talk) 13:27, 5 September 2019 (UTC)
I think that this discussion is resolved and can be archived. If you disagree, don't hesitate to replace this template with your comment. VIGNERON (talk) 13:27, 5 September 2019 (UTC) |
--Zbmath authorid (talk) 14:19, 11 February 2020 (UTC) Shouldn't we impose a constraint that the target of P735:"given name" actually is a "given name" (instance of Q202444)? There are many cases not satisfying this constraint, and they are all wrong from my point of view. Either they contain a country (user's mistake), or something different, like for example Q345:"Virgin Mary" (obviously intentional vandalism):
SELECT ?item ?giv WHERE { ?item wdt:P735 ?giv. FILTER NOT EXISTS {?giv p:P31/ps:P31/wdt:P279* wd:Q202444 }
}
wondering why people add countries and cities into this property
[edit]Sometimes editors add countries or cities to statements of given name (P735). Some examples: Janka Wohl (Q1465294), Jean-George Guiguer (Q3166029), Zdeněk Lhota (Q59524629), Haruna Esseku (Q56249921), Sofia Kuvshinnikova (Q18907422). I have checked some of the editors' user page but still can't figure out in what languages the ambiguous has occurred. --Jd3main (talk) 13:58, 15 August 2019 (UTC)
- Probably mixing with other properties... --Infovarius (talk) 14:52, 6 September 2019 (UTC)
Single value constraint?
[edit]I'm not really convinced about its addition. This property can typically hold multiple values. --- Jura 16:18, 4 October 2019 (UTC)
- Yes, it can be hold multiple values. Yet, it would be wrong to write these several values without separator.
- For example: George H. W. Bush (Q23505).
- It would be wrong to write:
given name (P735): George, Herbert, Walker
without subproperties. Therefore, in this case there should be an error. - The correct entry is:
given name (P735): George (series ordinal (P1545) 1), Herbert (series ordinal (P1545) 2), Walker (series ordinal (P1545) 3)
. In this case, the database knows the order of the given names and there is no error.
- It would be wrong to write:
- Or assume a person who changes the name: Chelsea Manning (Q298423).
- It would be wrong to write:
given name (P735): Chelsea (1), Bradley (1), Elizabeth (2), Edward (2)
. In this case, the database knows the order of the given names but doesn't know which are the current two names. Therefore, in this case there should be an error. - The correct entry is:
given name (P735): Chelsea (1, start time: 2013), Bradley (1, end time: 2013), Elizabeth (2, start time: 2013), Edward (2, end time: 2013)
. In this case, the database knows which names are the current names and there is no error.
- It would be wrong to write:
- In general, there should be an error if someone creates several given names but doesn't distinguish them with a subproperty. --Eulenspiegel1 (talk) 21:59, 4 October 2019 (UTC)
- @Jura1: Do you have still any concerns? Or are you convinced, now? --Eulenspiegel1 (talk) 15:07, 6 October 2019 (UTC)
- People with multiple names do not always use a full set of the names. E. g. Omar Khayyám (Q35900) is sometimes referred as Omar Khayyam, sometimes as Abu’l Fath Omar ibn Ibrahim al-Khayyam and sometimes as Ghiyat ad-Din Abu-l-Fath Omar bin Ibrahim an-Nisaburi al-Khayyam. Should “Omar” be qualified as #1, #2 or #3 given name?
- People often use different names or different versions of their name simultaneously. E. g. Jonathan Sacks (Q336766) has (a) “civil” given name(s) Jonathan Henry, while at the same time he bears (a) Jewish given name(s) Yaakov Zvi.--Shlomo (talk) 19:05, 9 October 2019 (UTC)
- Omar Khayyám (Q35900): "ibn" means "son of", Thus, everything which is behind "ibn" is not part of the given name. In western countries it is common to speak to people with their first given name. Thus, instead of "Charles Robert Darwin" they only say "Charles Darwin" and miss all given names after the first one. In some eastern languages, it is more common to speakt to people with their last given name. Thus, intead of "Ghiyat ad-Din Abu’l Fath Omar Khayyam" they only say "Omar Khayyam" and miss all given names before the last one. Nevertheless, the correct entry would be:
given name (P735): Ghiyat ad-Din (series ordinal (P1545) 1), given name (P735): Abu’l Fath (series ordinal (P1545) 2), Omar (series ordinal (P1545) 3)
("ibn" and everything behind is not part of the given name.) - Jonathan Sacks (Q336766): His offical name is "Jonathan Henry Sacks". His hebrew name is "Yaakov Zvi". Thus, use the separator "applies to part" to distinguish between the official name and the hebrew name. --Eulenspiegel1 (talk) 20:34, 9 October 2019 (UTC)
- ad Omar Khayyám (Q35900): the “ibn…” part was not the question, no need to knock an open door ;) Well, if it's not a problem that the name used usually as a single one (i.e. the first one) has #3, I could accept it as a solution for Khayyam – and for other people, who's set of names has a stable order.
- ad Jonathan Sacks (Q336766): “Applies to part” is misleading. Yaakov Zvi and Jonathan Henry are not “parts” of his given name, they are two separate names – of the same person.
- Also Yaakov Zvi is not necessarily a Hebrew name. It is a Jewish name, used preferentially in religious context. The English, Russian or Yiddish speaking Jews would use it in the Synagogue as well; on the contrary, a (non-religious) Hebrew Israeli newspaper would probably refer to him as יונתן סאקס (=Yonathan Sacks) or maybe ג׳ונתן סאקס (=Jonathan Sacks).
- About the “official name” qualification: How do we define an “official” name, and how can we check/prove that one name is “official” while the other one is not? I'm pretty sure the Jewish name has been given at some official ceremony approved by the official Jewish community and recorded in it's official register – why should it be less “official” than the name used by the non-Jewish authorities? Remember we're not talking only about rabbi Sacks; there are people less famous whose birth records can't be found, people who lived long ago whose birth records never existed, people with double (or multiple) citizenship &c.--Shlomo (talk) 22:14, 9 October 2019 (UTC)
- ad Jonathan Sacks (Q336766): Yes, Yaakov Zvi is one given name and Jonathan Henry is another given name. Yet, if you only distinguish by series ordinal, you have:
given name (P735): Jonathan (1), Yaakov (1), Henry (2), Zvi (2)
- Thus, the two given names could be: "Jonathan Zvi" and "Yaakov Henry". To clarify which to names belong together we need a separator.
- If you know that "Jonathan (1)" and "Henry (2)" is used in the same part of society, it is clear, that "Jonathan Henry" is one name. If you know that "Yaakov (1)" and "Zvi (2)" is used in the same part of society, it is also clear, that "Yaakov Zvi" is another name.
- "Jonathn (1)" and "Zvi (2)" don't apply to the same part. Thus, the two names don't belong together and "Jonathan Zvi" is not a name.
- If you don't like the separator "applies to part", we can look for another separator (e.g. criterion used (P1013)). Yet, there should be a separator.
- The official name is the name which is written in the passport or which the government/public administration uses when it writes to you.
- For people with double citizenship exists the separator applies to jurisdiction (P1001). --Eulenspiegel1 (talk) 22:48, 9 October 2019 (UTC)
- ad:
The official name is the name which is written in the passport or which the government/public administration uses when it writes to you.
It would be virtually impossible to find authoritative sources supporting that kind of qualified statement. Besides, my experience is, that various authorities even of the same jurisdiction often use different names or variants of name, especially in the cases of multiple names, hence it wouldn't solve the problem anyway. Except you intend to use qualifiers like applies to part, aspect, or form (P518)→“official name variant used in the letter from Ministry of Silly Walks dated 2015-02-29”… - Generally, qualifiers are intended to expand the already existing useful information already included in statement. (Mis)using them for grouping single statements is a bad practice. Doing it via properties that are intended and designed for a different purpose is a very bad practice. Doing this without premediation and documentation “as you like it” is a very, very bad practice. Bad practice usually means, that the construction will break sooner or later with some maintanence, data import, or just a contribution of another editor who “likes it” other way.--Shlomo (talk) 06:10, 10 October 2019 (UTC)
- 1. Given name has not citation-needed constraint (Q54554025), Thus, there is no need of a reference. In fact, it is total usually to have no references for the values of the given names.
- 2. Of course, authorities can also write the name wrong. Wrong names shouldn't be written in "given name" - neither with qualifiers nor without qualifiers. In fact, I know no item, where a wrong written name is entered as a value.
- 3. Yes, qualifiers are intended to expand the already existing useful information. Thus, in Jonathan Sacks (Q336766) the qualifiers expand my knowledge: Because of the qualifiers I know that "Yaakov" and "Zvi" are used in a religious (jewish) circumstance.
- 4. The property separator (P4155) defines for each properties which qualifiers separates values of a property with single value constraint. To define separators of properties is the propose of this item. It's bad practice to ignore such properties like separator. It exists for 2 years and is widely used. If you think the documentation is not good enough, feel free to expand the documentation. Yet, don't ignore this property because you don't like it. --Eulenspiegel1 (talk) 23:47, 10 October 2019 (UTC)
- ad:
- Omar Khayyám (Q35900): "ibn" means "son of", Thus, everything which is behind "ibn" is not part of the given name. In western countries it is common to speak to people with their first given name. Thus, instead of "Charles Robert Darwin" they only say "Charles Darwin" and miss all given names after the first one. In some eastern languages, it is more common to speakt to people with their last given name. Thus, intead of "Ghiyat ad-Din Abu’l Fath Omar Khayyam" they only say "Omar Khayyam" and miss all given names before the last one. Nevertheless, the correct entry would be:
- While I understand that you may want to add this in one way or the other, I don't think it should be implemented through a single value constraint. If you don't want to add these qualifiers yourself, please don't add the constraint. It's always possible to match P735 values with labels in one or the other languages, with birth name or name in native language statements to determine its order. --- Jura 18:47, 10 October 2019 (UTC)
- 1. Why it shouldn't be implemented through a single value constraint? It's exatcly the use of the two properties "single value constraint" and "separator": An item shouldn't have several values which don't differ in separating qualifiers (which are defined by separator (P4155)).
- Or the other way: If an item with "single value constraint" and "separator" have several values, they should differ in the separating properties (defined by separator (P4155)).
- This is, how "single value constraint" and "separator" works in every value. Why you want an exception for "given name"? They don't have the "mandatory constraint", so it's always possible to ignore the rule. Yet, it is an mistake if there exists several values which don't differ in their qualifiers.
- 2. Wikipedia is for human persons. Wikidata is for machine reading. A human person can understand the labels. A human person doesn't need the property "given name" or "family name". The two properties "given name" and "family name" are only for the computer because the computer doesn't understand the labels.
- 3. I have the impression that you argue about "single value constraint" without "separator". That would be wrong. Yet, "separator" is often used as a qualifier of "single value constraint". "single value constraint" with "separator" as qualifier is a correct use. --Eulenspiegel1 (talk) 23:47, 10 October 2019 (UTC)
- It seems to me that you are trying to implement mandatory qualifiers through something that is meant for identifiers.
- The main advantage of P735 is that it allows to retrieve all people with the same given name. Computers can still work out the display order of a value for a person based on the preferred label or sting property they want to use. I don't see what the difficulty could be once it's identified which ones are P735. --- Jura 10:00, 11 October 2019 (UTC)
- 1. If I wantetd to make it mandatory, I would have used as qualifier
constraint status (P2316): mandatory constraint (Q21502408)
(as it is already used for conflicts-with constraint). Yet, I didn't use it. The qualifiers should not be mandatory. - 2. If I wanted to use a qualifier everytime, I would have used required qualifier constraint (Q21510856). Yet, I didn't use it.
- Of course, if a person doesn't change the given name, you can always enter as qualifier: start time: date of birth. Yet, this is useless. There is no need to enter the start time if the person didn't change the given name. (Nevertheless, if the person has changed the given name, the start time of the given names differ. In this case, it's useful to use the qualifier "start time".)
- 3. I neither choose 1. nor 2. I choose single-value constraint (Q19474404) together with separator (P4155). I already explained the reasons in my earlier textes.
- 1. If I wantetd to make it mandatory, I would have used as qualifier
- Question:
- You think the use of single-value constraint (Q19474404) together with separator (P4155) in "given name" is wrong. Can you give an example for a property in which separator (P4155) is correctly used? Or do you think separator (P4155) is completely useless and should be deleted? --Eulenspiegel1 (talk) 01:06, 12 October 2019 (UTC)
Two given names used interchangeably
[edit]Małgorzata Mazurczak (Q24577078) (w:pl:Małgorzata Urszula Mazurczak) is called either Małgorzata Urszula Mazurczak or Urszula Małgorzata Mazurczak. Is there a way to reflect this in WD ? Kpjas (talk) 13:34, 14 May 2020 (UTC)
A notebook to explore first names with Ada as example
[edit]I've written a small notebook in French which explore what we can learn about first names using Wikidata. The idea is just to take a first name. The choosen example is Ada (Q346047) but you can take anyone and ask a few questions :
- How many Ada (Q346047) on Wikidata ?
- Who is the most famous Ada (Q346047) ?
- What is instance of Ada (Q346047) ?
- WHat is the gender of Ada (Q346047) ?
- WHat is the occupation of Ada (Q346047) ?
- What is the country of Ada (Q346047) ?
etc.
You can find the notebook here : User:PAC2/Prenoms. English version is possible if you want.
--PAC2 (talk) 05:52, 6 June 2020 (UTC)
- Nice. If you change your interface language to English, you will see a series of similar queries on any talk page of given name items. --- Jura 08:40, 6 June 2020 (UTC)
Vietnamese middle names
[edit]I've proposed a Vietnamese middle name property for Vietnamese names, which don't neatly fit into the Western-centric practice of treating a middle name as a second given name. – Minh Nguyễn 💬 01:32, 15 July 2020 (UTC)
Usual given name and/or preferred rank
[edit]Notified participants of WikiProject Names Hi all! I've recently noticed two (or three) different ways of indicating which name is the usual given name:
- setting preferred rank for the usual given name (e.g. in Barack Obama (Q76))
- adding the qualifier object of statement has role (P3831)usual forename (Q3409033) to the usual given name (e.g. in Charles Darwin (Q1035))
- a combination of 1 and 2: setting preferred rank and adding the qualifier object of statement has role (P3831)usual forename (Q3409033) to the usual given name
I think that the option 1 should be avoided as too generic (the fact that the reason for preferred rank is "usual given name" should be specified in some way), so in my opinion the choice should be between options 2 and 3. Personally I agree with @Sjoerddebruin: who in this page in 2016 wrote "the rank should be normal for all of them though", so I would choose option 2. Anyway, the most important thing is to establish a clear guideline (which should also strongly advise the use of the qualifier series ordinal (P1545)) and write it somewhere. What's your opinion? Thanks, --Epìdosis 19:41, 5 September 2020 (UTC)
- I started adding the qualifier object has role = usual given name before we had reason for preferred rank, which now looks like the better choice maybe? Moebeus (talk) 19:49, 5 September 2020 (UTC)
- I prefer option 1 or 3 for querying reasons: we usually want the usual given name only, so using ranks makes the query easier. If someone wants all given names, it is always possible to query with all ranks. -Ash Crow (talk) 17:30, 7 September 2020 (UTC)
Multilingual names
[edit]Prominent figures/celebrities or even normal people might have chosen multilingual names. For example, see Traditional Chinese birth name (P1477) for Yoo Jeong-yeon (Q25374501) (this is not actually her birth name (P1477), but I'm keeping it there for the time being).
How should these be entered? Lectrician1 (talk) 16:59, 18 January 2021 (UTC)
German nobility
[edit]@Roelof Hendrickx: claims (without sources) that "in the 17th and 18th double christian names and triple christian names have been fashionable for members of the german nobility". Is anyone else familiar with this? How should we model that using P735? Do we treat them as separate given names with series ordinals in the usual way, or should we be reifying double and triple given names? Bovlb (talk) 18:34, 21 November 2021 (UTC)
- I did give sources for the use of that. But those sources disappeared when you deleted the christian name. Roelof Hendrickx (talk) 18:45, 21 November 2021 (UTC)
- Yes, compound given names exist and not only in German nobility - it is also very common for French names. The articles linked to double given name (Q1243157) describe this phenomenon. Often they are hyphenated (clarifying that the name is a compound one), but a space could be used, too. A Marie Luise with a compound given name would always be called "Marie Luise" while a Marie Luise with multiple single names has only one usual forename (Q3409033) - so usually she would either be called "Marie" or "Luise". The latter phenomenon should be modelled with separate given names with series ordinals (and maybe the usual forename (Q3409033) set to "preferred") but for the first phenomenon an own item for the compound name makes sense to me. There already exist a lot of items for compound given names in French: Jean-Marie (Q3167523), Jean Marie (Q6171108), Jean-Pierre (Q6169840), Jean-Paul (Q16798217), etc. - Valentina.Anitnelav (talk) 08:25, 22 November 2021 (UTC)
- Thanks. I stand corrected. Bovlb (talk) 17:17, 22 November 2021 (UTC)
family name (P734) as a given name?
[edit]@Quesotiotyo: Some people have given names that are (family name (P734)). I had an uncle, who got the maiden names of his two grandmothers.
Shouldn't there by a way in Wikidata to allow a user to specify a family name as a given name?
If that's not allowed, then shouldn't Wikidata prevent a user from entering a family name as a given name?
More importantly, I'm concerned that the proper way to fix a "constraint violation" of this type is NOT to just delete the offending name but to create it as a given name and then use it as a given name, if that's what people think is the best way to handle this.
EXAMPLES: User:Quesotiotyo just deleted second given names in the following two cases:
- "McLean" was deleted from the Wikidata item for J. McLean Sloughter Q98972826, leaving "J." as the given name
- "Dunster" was deleted from the Wikidata item for John Dunster Kettelle, Jr. Q106775681, leaving "John" as the given name.
I can create given names of "McLean" and "Dunster" to fix these two items. However, I think the onus for doing that should be on the user who wants to fix these constraint violations, because by deleting these middle names, they make Wikidata worse by any reasonable criterion other than the automated designation as constraint violations of family names used as given names.
I appreciate User:Quesotiotyo trying to improve Wikidata, but I'm concerned that this work is making it worse not better. ??? Thanks, DavidMCEddy (talk) 22:50, 31 January 2022 (UTC)
- "Some people have given names that are (family name (P734))."
- This does not make sense within the context of Wikidata (reading through Wikidata:WikiProject_Names#Basic_principles may provide some clarity). Each use of a particular string of characters determines what that string is (regardless of how it used elsewhere), and in turn the P31 value of a name item created to represent that string dictates which property the item should be used with.
- "I had an uncle, who got the maiden names of his two grandmothers."
- In this case, while his given names are string-equivalent to their family names, they would not be equivalent items in Wikidata. They could be if we chose not to differentiate between given name (Q202444) and family name (Q101352) and used only instance of (P31)name (Q82799) instead and if we used just the name (P2561) property with a corresponding object of statement has role (P3831) qualifier, but we do not.
- "If that's not allowed, then shouldn't Wikidata prevent a user from entering a family name as a given name?"
- If only! :)
- "More importantly, I'm concerned that the proper way to fix a "constraint violation" of this type is NOT to just delete the offending name but to create it as a given name and then use it as a given name, if that's what people think is the best way to handle this."
- I am working on this right now (or was, before getting a bit waylayed). Where an equivalent given name item existed, the value was simply changed (this is still ongoing). Some values will have to be migrated to family name (P734) in cases where the wrong property was used. New given name items will be created where appropriate and added to all people possessing those names. In more than a few cases, vandalism will need to be reverted. And after all of that is done, the inverse problem awaits! :) In all of this my goal is to fix the error, not simply make the constraint violation go away. The latter is just a sign that the former needs to be done.
- "I appreciate User:Quesotiotyo trying to improve Wikidata, but I'm concerned that this work is making it worse not better."
- There are certainly causes for concern when it comes to Wikidata, but this need not be one of them (but thank you for taking notice).
- --Quesotiotyo (talk) 03:22, 1 February 2022 (UTC)
Hello, @DavidMCEddy, Quesotiotyo:, I understand you both. But my heart is with David (who got a soul of rules) rather than with formal approach of Quesotiotyo. Quesotiotyo, formally you are right, thank you for trying to fix errors/constraints. But you didn't answer to a keypoint: why to create new item for strings which are usually surnames but occasionally (very rarely, 1-2 cases) are used as given names? I suppose that we can allow to have a little number of exceptions just for clarity and not duplication of information. But in any change of situation (growth of usecases, pronunciation variation) new items are highly appreciated, of course. --Infovarius (talk) 15:09, 2 February 2022 (UTC)
- There is no requirement to add P735 if one doesn't want to create the corresponding item(s). It's not uncommon that the values for P735 don't exist. This also avoids that one feels "disrupted" when others clean up. --- Jura 15:48, 2 February 2022 (UTC)
- @Infovarius: About 42.9% of all given name items are used with P735 on only 1 or 2 items (this is not counting the nearly 24,000 (!!) given names that are currently unused).
- [Source: https://w.wiki/4nKk]
- --Quesotiotyo (talk) 18:40, 2 February 2022 (UTC)
Proposal: Drop indirect constraints about native label and writing system
[edit]- Property “native label (P1705)” declared by target items of “given name (P735)”: If [item A] has this property with value [item B], [item B] is required to have property “native label (P1705)”.
- Property “writing system (P282)” declared by target items of “given name (P735)”: If [item A] has this property with value [item B], [item B] is required to have property “writing system (P282)”.
These two constraints ought to be reported on the given name item, not on all items that use a particular given name. These are very unhelpful for new users who don't know how to fix it and don't understand why they're being asked to fix an item they're not editing. I have seen new users add P1705 and P282 claims to a person item because of these unhelpful messages. Bovlb (talk) 23:43, 13 September 2022 (UTC)
- I agree with this and would even go as far as to say it does not need to be reported on a given name item. People are generally given names before they know to write and even if they don't write; this isn't an inherent property of names. Just as we do not have constraints like this for literary works (The Odyssey being an example of one which was not originally written), there is no need for constraints like this on names. Middle river exports (talk) 07:26, 14 September 2022 (UTC)
Proposal: Add P1813 to the allowed qualifiers
[edit]I would like to propose allowing short name (P1813) as a qualifier for this property. Some given names are (or can be) abbreviated to a certain form in some names, but not in others. For example, some William may call himself Will while some other may call himself Bill or not abbreviate William at all. Of course, we could also model this like in Joe Biden (Q6279) with nickname (P1449) and object of statement has role (P3831)most frequent value (Q71538638). Allowing short name (P1813) would, however, yield the benefit that abbreviated second names (such als W. in George W. Bush) can be included. Using unknown value this would be possible even when it is not known what they abbreviate (like in Annie L Cherry Petherbridge (Q110777770)). --2A02:8108:50BF:C694:EC05:EF9E:B0EF:C96D 14:50, 28 February 2023 (UTC)
Why all the exceptions listed?
[edit]Why are there several exceptions listed, and why? StarTrekker (talk) 08:52, 7 April 2023 (UTC)
Only show values that are subclass of: given name in autocomplete
[edit]Could you please change this so it only shows these values in the autocomplete? This prevents false additions and makes it easier and quicker to add this info. Same for family name and probably many other properties that show just any item in autocomplete (maybe it would make sense to show it if the title matches 1:1 which isn't autocomplete). Prototyperspective (talk) 10:20, 8 September 2024 (UTC)
- All Properties
- Properties with wikibase-item-datatype
- Properties used on 1000000+ items
- Properties with distinct value counts
- Properties with conflicts with constraints
- Properties with constraints on type
- Properties with qualifiers constraints
- Properties with scope constraints
- Properties with target required claim constraints
- Properties with entity type constraints
- Properties with complex constraints
- Person properties