Property talk:P5102: Difference between revisions
Content deleted Content added
Amadalvarez (talk | contribs) |
|||
| (16 intermediate revisions by 10 users not shown) | |||
Line 3:
{{ExternalUse|
* [[:ca:Plantilla:Format P170]]
* [[:ga:Teimpléad:Format P170]]
}}
== Label for this property ==
Line 31 ⟶ 32:
:Their meaning is identical, one being the noun and the other it’s form as an adjective. Both the other allowed values as well as the specific phrasing ‘’’‘’nature’’’ of statement’’ favor the noun. There are more cases like this, such as {{Q|10585806}} and {{Q|55935291}}. --[[User:Matthias Winkelmann|Matthias Winkelmann]] ([[User talk:Matthias Winkelmann|<span class="signature-talk">{{int:Talkpagelinktext}}</span>]]) 22:21, 28 September 2020 (UTC)
== Former attribution ==
I've added {{Q|Q70918737}} as an allowed value, as parallel to "attribution". - [[User:PKM|PKM]] ([[User talk:PKM|<span class="signature-talk">{{int:Talkpagelinktext}}</span>]]) 21:39, 17 April 2021 (UTC)
== Should we allow prediction as well as estimate for values? ==
We already allow {{Q|Q37113960}}, maybe we should allow {{Q|Q748250}}, which has a similar but different meaning, and is usually used for forecasting what will happen, when, and with what probability, vs estimate which seems to be broader.
== Conflict with [[Property:P1480|P1480]] to be solved ==
{{ping project|Data Quality}} Hi all! As of now, as [https://w.wiki/5L7j this query] shows, 27 usable values are shared between this property and {{P|1480}}. This makes queries harder, of course; it affects some qualifiers used for dates (see [[Help:Dates]], which prescribes {{P|1480}}). I think we should establish, for each value, only one suitable property and then perform the appropriate replacements. The need of having this property was established in [[Wikidata:Requests for comment/Close-out of statements formerly using P794|this RfC]] (2018) as proposed by {{ping|Deryck Chan}}, in order to have two different properties for the qualification regarding the reference or the qualification regarding the statement. While the problem is complex and I would like to reflect more on it, it seems that the distinction has blurred during this four years and that probably P1480 has over-expanded against P5102; I also tend to think that, since P1480 regards the source of a statement, we should reflect on the possibility of restricting its use to references, excluding it from qualifiers. Opinions are welcome! --[[User:Epìdosis|'''Epì''']][[User talk:Epìdosis|<span style="color:green">'''dosis'''</span>]] 17:27, 22 June 2022 (UTC)
:Interesting. I remember P5102 was created because the curators of {{P|1480}} were very strict about their constraints, so as a compromise we defined a conceptual boundary between "the state of human knowledge is fuzzy" (P1480) vs "the fundamental truth is fuzzy" (P5102), with the expectation that P5102 would be less strict about constraints. Among the shared values, I can see many of them can be reasonably used for both with different meanings, e.g. {{Q|Q10585806}} and {{Q|Q10578722}}. The P1480 use case will be "10 people measured the high tide independently, reporting values of min. 2.0m and max 2.1m", whereas the P5102 use case will be "the water level at this harbour is min. -1.8m and max +2.1m [due to tidal range]". In terms of your proposed intervention, I think you'll need to ask the curators of P1480 would even entertain the idea of moving all their current use cases from qualifier to reference... [[User:Deryck Chan|Deryck Chan]] ([[User talk:Deryck Chan|<span class="signature-talk">{{int:Talkpagelinktext}}</span>]]) 10:40, 23 June 2022 (UTC)
::{{ping|Deryck Chan}} I mostly agree. While effectively the examples you make for {{Q|Q10585806}} and {{Q|Q10578722}} demonstrate that having them on both sides is fine, I am mainly thinking to the qualifiers used for dates (e.g. {{Q|5727902}}, {{Q|30230067}}, {{Q|56644435}}), which should probably belong to single references instead of to the entire statement. Let's make a concrete example, I have come across this many times: two sources report the birth date of a person, one as year X and the other as year X ''circa''; this could be modeled in two ways: either two statements, one with the first source and the second with the qualifier P1480/P5102={{Q|5727902}} and the second source; or one single statement, with the first source and the second source containing in itself P1480={{Q|5727902}}. Both options have pros and cons; I tend to think that the first is the most used (I usually act in this way), but I'm not completely sure it is the best one. --[[User:Epìdosis|'''Epì''']][[User talk:Epìdosis|<span style="color:green">'''dosis'''</span>]] 20:13, 23 June 2022 (UTC)
== Inference? ==
Could we include {{Q|Q408386}} here? Example - for {{Q|Q123513698}}, to support the statement that Mackensen was born in Scotland, I cited a published source in which a historian explained the process by which he inferred that Mackensen had emigrated from Scotland to Poland. [[User:DragonflySixtyseven|DS]] ([[User talk:DragonflySixtyseven|<span class="signature-talk">{{int:Talkpagelinktext}}</span>]]) 19:57, 22 November 2023 (UTC)
:I don't see why not. [[User:Swpb|Swpb]] ([[User talk:Swpb|<span class="signature-talk">{{int:Talkpagelinktext}}</span>]]) 20:11, 28 November 2023 (UTC)
::Okay. And what steps are necessary to do this? [[User:DragonflySixtyseven|DS]] ([[User talk:DragonflySixtyseven|<span class="signature-talk">{{int:Talkpagelinktext}}</span>]]) 14:32, 7 February 2024 (UTC)
== Grades of trueness from obviously false to absolutely true ==
Imagine there is a task to characterize the statement, and the degree of evaluation has to be carefully chosen and exactly corresponding to the state of things.
I wonder if there are enough values for [[Property:P5102|P5102]] to specify exactly the degree that is needed.
From the full list of the existing values, I could select the following (sorted from false to true):
'''False'''
[[Q99841874|non-canon]] -> [[Q107217620|unsubstantiated]]
'''Probably false'''
[[Q104378399|dubious]]
'''Unknown'''
[[Q1321906|open problem]] -> [[Q41719|hypothesis]] / [[Q18603603|hypothetically]] / [[Q18706315|hypothetical entity]] -> [[Q319141|conjecture]] -> [[Q3144351|scientific hypothesis]]
'''Unsure'''
[[Q18912752|disputed]] -> [[Q28831311|unconfirmed]] -> [[Q30230067|possibly]] / [[Q56644435|probably]] / [[Q13649246|uncertainty]] -> [[Q3962655|opinion]] / [[Q855395|interpretation]] -> [[Q652476|retrospective diagnosis]]
'''True with exceptions'''
[[Q28962310|rarely]] -> [[Q100349848|partially]] / [[Q110143752|sometimes]] / [[Q115667526|depends on jurisdiction]] -> [[Q90177495|does not always apply]]
'''True but only one side of range limitation is known'''
[[Q110290991|no earlier than]] / [[Q110290992|no later than]] -> [[Q10585806|minimum]] / [[Q111305333|maximum]] / [[Q47035128|greater than]] / [[Q52834024|less than]] / one more [[Q10578722|maximum]] -> [[Q55935291|greater than or equal to]] / [[Q55935272|less than or equal to]]
'''The true value is unknown but is located in some range'''
[[Q50376823|expected]] / [[Q744069|extrapolation]] / [[Q18122778|presumably]] / [[Q37113960|estimate]] / [[Q187631|mathematical interpolation]] -> [[Q202785|average]]
'''True with known limitations'''
[[Q20820099|de jure/de facto]] -> [[Q5727902|circa]] / [[Q91013007|mainly]] / [[Q28962312|often]]
'''True'''
[[Q101072|definition]] -> [[Q712144|de facto]] -> [[Q1520777|certainty]] -> [[Q116142274|illegal mark]] -> [[Q9510|reality]]
'''Questions'''
How do you look at this? I see a lot of gaps and lots of values for P5102 that could fill the gaps.
Could it be a problem to make some more of the existing items allowed to be used as values for P5102? What if there will be as much new available values as there are already listed above? Would it be a problem? [[User:Nikolay Komarov|Nikolay Komarov]] ([[User talk:Nikolay Komarov|<span class="signature-talk">{{int:Talkpagelinktext}}</span>]]) 20:31, 6 February 2024 (UTC)
:wouldn't be a problem to add more allowed values but would want to make sure we aren't making things more ambiguous. [[User:BrokenSegue|BrokenSegue]] ([[User talk:BrokenSegue|<span class="signature-talk">{{int:Talkpagelinktext}}</span>]]) 22:13, 6 February 2024 (UTC)
== Consensus decision-making as a allowed value ==
Items of {{q|Q877358}}, have {{P|P9681}} with the voter organization (mostly, Security Council) and the number of votes fav./con./abs. as a qualifiers with they correspondant properties (Ex.:[[Q860231#P9681|Resolution 757]]). However, sometimes that result is "uncountable", because the decissions is "by consensus", without indication about number of present voters. Then, I used {{P|P5102}} with {{q|Q188577}}.
As this value is not in the {{q|Q21510859}} value allowed, these cases generate a [[Wikidata:Database reports/Constraint violations/P5102|constraint]].
So, do you agree to include {{q|Q188577}} as a valid value ?.
CC: @[[User:Nikolay Komarov|Nikolay Komarov]], for ping to me about constraint.
Thanks, [[User:Amadalvarez|Amadalvarez]] ([[User talk:Amadalvarez|<span class="signature-talk">{{int:Talkpagelinktext}}</span>]]) 15:11, 14 February 2024 (UTC)
:No news, good news. I proceed !. Thanks, [[User:Amadalvarez|Amadalvarez]] ([[User talk:Amadalvarez|<span class="signature-talk">{{int:Talkpagelinktext}}</span>]]) 06:22, 25 February 2024 (UTC)
| |||