Wikidata:Property proposal/number of signatories

From Wikidata
Jump to navigation Jump to search

number of request signatories

[edit]

Originally proposed at Wikidata:Property proposal/Organization

Descriptionnumber of persons who signed a given petition, initiative, request for a referendum, or other large scale proposal/request.
Data typeQuantity
Allowed valuesintegers, >100
Allowed unitsnone
Example 1Massachusetts Sales Tax Relief Act (Q6784356) → 19,000 [1]
Example 2California is Not for Sale (Q25831350) → 250,000 [2]
Example 3Citizens initiative "The Netherlands Gets New Energy" (Q5501681) → 45,939 [3]
Example 4Romanian constitutional referendum, 2018 (Q28726192) → 3,000,000 [4]
Example 5Referendum "Biodiversity & Natural Beauty in Bavaria" (Q61507381) → 1,745,383 [5]
Example 6Citizens' Initiative Recognize ME (Q17300248) → 54,000 [6]
Example 7Dutch Ukraine–European Union Association Agreement referendum, 2016 (Q21117876) → 427,000 [7]
Planned useto some from [8] when applicable
Expected completenesseventually complete (Q21873974)
See also
  • number of support votes (P8683): number of votes that a referendum, proposal, bill, etc. has received support in a vote. Other vote-counting properties are aimed at general elections where voting translates into support, but there is no explicit option to vote for / against
  • votes received (P1111): qualifier stating the number of votes for a candidate in an election
  • ballots cast (P1868): total number of ballot(s) cast, including invalid or blank ballots

Motivation

[edit]

While we have several properties to describe proposals that have actually been formally voted on, we lack the above for what is generally the preceding steps. Please help complete with a few representative samples. (Add your motivation for this property here.) --- Jura 12:51, 24 February 2021 (UTC)[reply]

Useful to differentiate from the number of support votes (P8683) when there was a ballot afterwards. @Amadalvarez, Pere prlpz, FranSisPac, ESM, Arnaurs: @Davidpar: --- Jura 08:10, 26 February 2021 (UTC)[reply]

Discussion

[edit]
When the options are excluding candidates (individuals or parties) the votes that count are represented by votes received (P1111); there are no votes against, and blank and null votes do not specifically affect any of the options.
When there is an option (referendum Yes / No) that must be weighed with a pro / cons, the way to reflect it are the number of negative votes (P8682) and number of support votes (P8683). Those referendums with more than one option (ex.:Puerto Rican status referendum, 2017 (Q30044084)) are similar to a excluding election, because you vote pro to the option you want, it is P1111.
The "consequences of the result" are usually the election of candidate and end of the process. In other cases, such as primaries or two-round elections, they represent a pre-selection of candidates who will be part of another subsequent election. Therefore, the way to represent the votes is the P1111 in different elections or election phases. Another example of a "consequence of results" is when, after a parliamentary election, the newly elected chamber have to vote for one (or more) candidates for President, among them. Deputies usually vote for / against / abstention for each candidate, ie on a weighted basis. In this case, P8682 / P8683 / number of abstentions (P5043) are used again (ex.:Presidential investiture in Catalonia, 2016 (Q97659738)).
Therefore, the case to which the new proposal applies, would be an election with a direct vote (signatures collected), where those who do not want to sign are equivalent to abstaining, and the "consequence of the result" may occur - if it exceeds the certain threshold - in another electoral process.
Therefore, I do not see the need to create this new property, as it is equivalent to P1111 of the “electoral process of collecting signatures or supporters”. It is equivalent to collecting endorsement among members that is done in some social organizations by the election of their board. If you get a certain number of endorsements, pass the cut to be a candidate, if don't, you are not excluded from voting. Thanks, Amadalvarez (talk) 22:45, 26 February 2021 (UTC)[reply]
  • @Amadalvarez: I don't think it would apply to elections, just to requests for ballots/votes. Requests may need to be backed by a number of persons, thus this proposal. Maybe you could outline how to add the data from the 6 samples to the items with existing properties. --- Jura 23:14, 26 February 2021 (UTC)[reply]
  • @Jura1: I understood that. But the example about Romanian constitutional referendum, 2018 (Q28726192) confuse to me, because is clearly a referendum and should be elimited as exemple to avoid misunderstanding. I updated its statements (P8682 / P8683 already exits) and you can see how it looks in cawiki infobox. As it is not a vote, but a support for a petition that may end in a legislative change, the difference between "vote" and "support" in the examples and the definition of property must be very clear and strong, otherwise we would have had enough with the P1111. For instance, I suggest to change definition to "number of persons who signed a given petition, popular initiative, referendum, or other large scale proposal/request that doesn't produce a direct legislative or political change, but may force that it happens". This property is not "the number of signatory (P1891)", because in P1891 the signatories of an agreement, contract, law, etc. cause the validity and application. I'll change my vote if we agree on it. Thanks, Amadalvarez (talk) 08:51, 27 February 2021 (UTC)[reply]
  • @Amadalvarez: Thanks, but you didn't add the number of 3,000,000 at all (see discussion with FranSisPac below). I changed "referendum" to "request for a referendum". I take it that it's quite auomatic that once the 3,000,000 signed, the vote had to happen. --- Jura 08:57, 27 February 2021 (UTC)[reply]
  • @Jura1:. This is my concern. The item is about the referendum. The popular initiative was the has cause (P828) and should be a different item, because are different moments, different participants and different consequences. That's why I talked about the similarity with P1111, but I accept the difference between both properties. However, the two actions (initiative and referendum) should be linked but separated. Regarding the description, IMHO should incorpore that has not legislative or political effects, because is not a vot. Thanks,  – The preceding unsigned comment was added by Amadalvarez (talk • contribs).
    • It's just a step in the process as for #1, #2, #5, #7. I expanded the description a bit and added sample #7.
      If one would make separate items for that aspect on #1,#2,#4,#5,#7, it's still unclear what property one would use. --- Jura 10:09, 27 February 2021 (UTC)[reply]
  • @Jura1:. It is obvious that our debate is about small nuances. We're closer to an agreement. An initiative / proposal can be made by anyone and does not have to be approved. A popular initiative / proposal made from outside the legislative body, to deal with it, usually has conditions of acceptance for processing, but in no case does it mean that it is approved. A referendum involves an implication of the outcome. Before a referendum or an election there are other phases, I know. And that's why I insist that they must be separate items, as are primaries to choose candidates are with respect to the vote for those candidates to become part of the chamber. We do not put in one item all the steps of an electoral process. You say, and I accept that these are not elections, but I do the simile because not all phases of a process have to go together in the same item. The #1 did not change taxes on November 19, 2009, but it did force the law to change. The item describes the bill (Nov-2010) and the article has a section to describe the popular petition. Item #2 describes an initiative that did not thrive and is therefore not the first phase, but the only one. Item #5 describes the initiative and does not talk about the law, i.e. the case contrary to #1. #7 is the same as #4. It is a referendum and the previous stages that ended up triggering the referendum, must have their own item.
  • In reference to your comment that, if we have to have a separate item, we should have its ontology defined, I agree. If you want us to work together, I will be delighted. To begin, I've created Popular initiative for a referendum against same-sex marriage in Romania (Q105700835) as a first phase of #4 and linked it with Romanian constitutional referendum, 2018 (Q28726192). I also did California is Not for Sale (Q25831350), the #2 & Citizens initiative "The Netherlands Gets New Energy" (Q5501681), the #3. Observe that I used P1111 until we have Pnnnn of this property. Take a look to constraint messages to avoid them in the future property. Your suggestions will be wellcome. Amadalvarez (talk) 11:39, 27 February 2021 (UTC)[reply]
  • I meant that we probably need the proposed property whether it's added to a separate item or not. Personally, I find the use of "votes received" (P1111) in (e.g.) California is Not for Sale (Q25831350) misleading. I don't mind if the proposed property is added to separate items when they exits. In some cases, it can mean that substantial parts of the other item end up being duplicated. --- Jura 08:35, 28 February 2021 (UTC)[reply]
  • Sorry, @Jura1:. Perhaps we have a misunderstanding. Above, I told you "I talked about the similarity with P1111, but I accept the difference between both properties", it is, I agree to create this property understanding that it will apply in the previous steps items, when an organizer just "get supports" to propose a voting. Later, in response to your request, I have created a few examples of the separate items that I advocate. And, to make the structure I propose understandable, I included the P1111 as a dummy of the future property. Look my sentence "Observe that I used P1111 until we have Pnnnn of this property". Then, do you agree, now?. So, I'll changed my vote. Please, be sure that final description of new property define what we agree. Thanks, Amadalvarez (talk) 19:20, 2 March 2021 (UTC)[reply]
  •  Oppose /  Comment In order to avoid confusions, this property should explicitly exclude situations where both a "yes" and a "no" possition are possible, namely referenda. In your "example 4" it seems your 3,000,000 figure is the not updatet "number of support votes (P8683)" property, so there is property collision in these cases. By the same token, in multioption polls / surveys, another definition (another property) would be required to account for the number of participants.--FranSisPac (talk) 20:39, 26 February 2021 (UTC)[reply]
    • @FranSisPac: The enwiki article mentions "over three million signatures" (above threshold: 500,000) which I understood as being different from the 3,531,732 votes in favor. The property shouldn't be used when pro/con applies, but it's unclear what suggests that it would. Feel free to edit the description. --- Jura 21:26, 26 February 2021 (UTC)[reply]
    • Thanks for asking, @Jura1: my apologies for having been absent from de debate for so long. Nevertheless, the subject has been in my daily to-do list. Even, occasionally, I have tried to write my "updated" position but didn't find the right way to express it, partly because I am a bit lost about what is the "final" status reached after your discussion / agreements with Amador (c.c.@Amadalvarez:). My strong position, and recomendation, for this being a useful and successful property that does not create confusion, is to keep it apart enough from any kind of voting process, even in cases where the outcome could be someting to be voted... and I realize that the latter is in the birth of your proposal. The way I see it, these are a few implications and some of my contributions:
  1. Change the name of the property, to make it more specific and differentiated, when needed, from the WikiData element / item to which it applies. I see this as particularly important in the many cases that "would require a separate item" (like Q105700835 created by Amador), while in the Wikipedia article the initiative or petition does not deserve more than a single sentence. I have worked with potential synomyms of the terms "petition", "initiative", "request" and "proposal", already appearing in the description, plus "demand", "proposition" and "claim". My suggestion would be "number of request signatories", because "request" appears as te more neutral and applicable synonym.
  2. Link it to public participation (Q541936).
  3. Specify potential attributes of this property:
    1. "applies to part, aspect, or form (P518)" (many times "citizens' initiative (Q660064)").
    2. "point in time (P585)".
  4. Stress that this is not to be confused with some voting associated properties, even though they can also apply to the same item if this one results from the request / initiative / demand being quantified by this property, and so coexist. Namely I refer to the ones currently appearing as "see also":
    1. number of support votes (P8683).
    2. votes received (P1111).
    3. ballots cast (P1868).
  5. Include examples from other environments, like "Change.org (Q3090589)" or "We the People (Q7977910)" petitions. This could be, for instance: "Reaction to the Sandy Hook Elementary School shooting (Q7300271)" (or maybe "Sandy Hook Elementary School shooting (Q906933)", present in more languages) → 151,000 [9] // applies to part, aspect, or form (P518) → gun control (Q1328508) // point in time (P585) → 17 December 2012.
  6. Current examples to be reviewed / completed.
  7. Include "integer constraint (Q52848401)".
Surely there are a few more things I could say, but I will stop here. Hoping to be useful.--FranSisPac (talk) 23:54, 16 March 2021 (UTC)[reply]
  • Thanks for your input. Most helpful. I have to give this some thought before commenting. --- Jura 16:37, 17 March 2021 (UTC)[reply]
  • @FranSisPac:
    re "#1: change the name of the property" ✓ Done excellent idea!
    re "#2." seems good, but I don't quite know where to do that.
    re: "#3." I suppose you refer to possible qualifiers. If there is some ambiguity object of statement has role (P3831) should do. "point in time" can be needed if the proposal is still open for signature.
    re: "#4." yes, the "see also" properties are there to indicate mainly from which properties it should be distinguished or which properties could be useful as well.
    re: "#5." sounds good, except that none currently have items. The two items you mention (Q7300271, Q906933) seem too broad and not suitable from their domain. I think we should avoid using the property on such items qualified with "applies to part".
    re: "#6.": I removed the Romanian sample as I'm not really convinced by Q105700835. It might be mislabeled, but I don't know much about Romanian terminology. As I wrote before, I find P1111 ("votes received"), added to some of these sample items since made this proposal, highly problematic.
    re: "#7.": added "integer" to allowed values. --- Jura 12:30, 5 April 2021 (UTC)[reply]

Thanks, @Jura1: let me insist that, from my point of view, it is critical to differentiate a propiety of the "request" from any property specific of the potentially resulting "voting event". The following table can help us understand my concern. Let's call "P" the proposed property, "R" the request event and "V" the potentially resulting voting event.

Case A B C
Situation There is no "V" Both "R" and "V" are separate items in WikiData "R" has no separate item, so it is embedded in the "V" item
P Makes sense Makes sense in "R", but not in "V" except if having the proposed qualifier "applies to part, aspect, or form (P518) R", manadatory in this case Makes sense in "V"
number of support votes (P8683) Does'n apply Does'n apply in "R" May coexist with "P" (in "V",of course)
number of negative votes (P8682) Does'n apply Does'n apply in "R" Idem
ballots cast (P1868) Does'n apply Does'n apply in "R" Idem
votes received (P1111) Does'n apply Does'n apply in "R" Idem, but only if "R" appears as one of the voting alternatives

As we can see, the initial statement «If followed by a vote, add...» should be removed and the "see also" changed to different from (P1889) (hence my point #4). We could say that these properties belong "to another world", even if linked.

--FranSisPac (talk) 16:04, 18 April 2021 (UTC)[reply]

    • @FranSisPac: sounds good. I removed the part of the description. I like your summary table, except the that I'm not sure if the last part ("Idem, but only if "R" appears as one of the voting alternatives") is that clear. I suppose it means one could vote either for the request (R) or some other alternative. BTW different from (P1889) doesn't work with properties as values. It would always be "see also". Shall we move ahead with this? --- Jura 21:14, 23 April 2021 (UTC)[reply]
Thanks again, @Jura1: there isn't really a possibility of showing that these three properties (four if we include the negative votes) do not apply here and should not be used? You know that I see this as a critical point that would clear the way. On your comment about the last part of the table, yes I undestand that, by definition and in a broad sense, votes received (P1111) should only be used when applied to an item that is one of many voting alternatives (one of the "candidates").--FranSisPac (talk) 17:59, 24 April 2021 (UTC)[reply]
  • Bravo, @Jura1:, good job !. I believe this discussion gather a lot of specifications about "how and when to apply" that should be summarized as a first point of the new property talk page in order to ensure good application without re-read this long proposal debate. The "restriction properties" are good tool to ensure values, classes, etc. but doesn't describe aspects related with ontology that have been agreed here, and just a few people comes back to proposal discussion to understand it. cc:@FranSisPac, NMaia, Davidpar:, Salut ! Amadalvarez (talk) 04:53, 16 July 2021 (UTC)[reply]
Retrieved from "https://www.wikidata.org/w/index.php?title=Wikidata:Property_proposal/number_of_signatories&oldid=1460921869"