Wikidata:Property proposal/Archive/53

This page is an archive. Please do not modify it. Use the current page, even to continue an old discussion.

This is the page currently active to archive closed property proposals.

The previous archive page is Wikidata:Property proposal/Archive/52.

number of parts (or number of components)

DescriptionThe numeric number of components
Data typeQuantity
Allowed valuesnumbers
Examplebuilding complex (Q1497364) → 5 (as qualifier)

Very often we have statements saying that A is a collection of some sort. While we would ideally have all of those parts listed under has part (P527) this isn't always possible and it isn't easy to judge, by looking at has part (P527) if that is complete. This property would allow a statement to be qualified with the number parts/components.

This should only ever be used as a qualifier and should only be used where the number of components are known and finite.

See number of parts of this work (P2635) for a similar but domain specific property.

To expand on the example above. A building complex (Q1497364) might be made up of 20 buildings where only 5 are relevant enough for Wikidata and thus only they are included in has part (P527). /André Costa (WMSE) (talk) 16:40, 2 June 2016 (UTC)


Tobias1984 (talk) TomT0m (talk) Genewiki123 (talk) Emw (talk) 03:09, 9 September 2014 (UTC) Ruud 16:15, 9 December 2014 (UTC) Emitraka (talk) 14:32, 14 October 2015 (UTC) Bovlb (talk) 19:10, 21 October 2015 (UTC) Peter F. Patel-Schneider (talk) 22:21, 23 October 2015 (UTC) ArthurPSmith (talk) 15:51, 5 November 2015 (UTC) Daniel Mietchen (talk) 20:53, 3 January 2016 (UTC) Harmonia Amanda (talk) 22:00, 27 February 2016 (UTC) Lechatpito (talk) Andrawaag (talk) 14:42, 13 April 2016 (UTC) ChristianKl (talk) 16:22, 6 July 2016 (UTC) Cmungall Cmungall (talk) 13:49, 8 July 2016 (UTC) Cord Wiljes (talk) 16:53, 28 September 2016 (UTC) DavRosen (talk) 23:07, 15 February 2017 (UTC) Vladimir Alexiev (talk) 07:01, 24 February 2017 (UTC) Pintoch (talk) 22:42, 5 March 2017 (UTC) Fuzheado (talk) 14:43, 15 May 2017 (UTC) YULdigitalpreservation (talk) 14:37, 14 June 2017 (UTC) PKM (talk) 00:24, 17 June 2017 (UTC) Fractaler (talk) 14:42, 17 June 2017 (UTC) Andreasmperu Diana de la Iglesia Jsamwrites (talk) Finn Årup Nielsen (fnielsen) (talk) 12:39, 24 August 2017 (UTC) Alessandro Piscopo (talk) 17:02, 4 September 2017 (UTC) Ptolusque 01:47, 14 September 2017 (UTC) Gamaliel (talk) --Horcrux (talk) 11:19, 12 November 2017 (UTC) MartinPoulter (talk) Bamyers99 (talk) 16:47, 18 March 2018 (UTC) Malore (talk) Wurstbruch (talk) 22:59, 4 April 2018 (UTC) Dcflyer (talk) 07:50, 9 September 2018 (UTC) Ettorerizza (talk) 11:00, 26 September 2018 (UTC) Ninokeys (talk) 00:05, 5 October 2018 (UTC) Buccalon (talk) 14:08, 10 October 2018 (UTC) Jneubert (talk) 06:02, 21 October 2018 (UTC) Yair rand (talk) 00:16, 24 October 2018 (UTC) Tris T7 (talk) ElanHR (talk) 22:05, 26 December 2018 (UTC) linuxo Gq86 Gabrielaltay Liamjamesperritt (talk) 08:44, 21 June 2019 (UTC) ZI Jony Ivanhercaz (Talk) 11:07, 15 July 2019 (UTC) Gaurav (talk) 22:39, 24 August 2019 (UTC) Meejies (talk) 04:38, 29 August 2019 (UTC) Iwan.Aucamp SilentSpike (talk) Tfrancart (talk) Sylvain Leroux TiagoLubiana (talk) 15:12, 2 December 2019 (UTC) Albert Villanova del Moral (talk) 15:43, 6 February 2020 (UTC) Clifflandis (talk) 15:10, 18 February 2020 (UTC) --Tinker Bell 16:48, 23 March 2020 (UTC) SM5POR Mkbergman (talk) 19:17, 10 July 2020 (UTC) Toni 001 (talk) 11:03, 11 July 2020 (UTC) The-erinaceous-one (talk) 22:02, 31 August 2020 (UTC) Cdo256 (talk) 06:26, 8 September 2020 (UTC) Antoine2711 (talk) 17:22, 25 September 2020 (UTC) Rehman 07:15, 12 October 2020 (UTC) Susanna Giaccai (talk) 07:22, 22 October 2020 (UTC) Nomen ad hoc So9q (talk) 10:17, 4 January 2021 (UTC) Simon Cobb (User:Sic19 ; talk page) 18:29, 8 January 2021 (UTC) Andrew Eells (User:AndrewEells ; talk page) 15:35, 7 April 2021 (UTC) RShigapov 9:48, 8 April 2021 (UTC)   Notified participants of WikiProject Ontology --Izno (talk) 11:53, 3 June 2016 (UTC)

day of week

   Done: day of week (P2894) (Talk and documentation)
Descriptionday of the week on which this item occurs, applies to or is valid on
Representsday of the week (Q41825)
Data typeItem
Domainany item where the day of the week needs to be specified directly or as a qualifier to another property
Allowed valuesMonday (Q105), Tuesday (Q127), Wednesday (Q128), Thursday (Q129), Friday (Q130), Saturday (Q131), Sunday (Q132)

The immediate reason for this proposal is to allow it as a qualifier for publications (see frequency of publication proposal above) but it could be used with many different items and properties and may (depending on the structure we eventually use) be a key component of the way to story "opening hours" currently being discussed. There may be a small overlap with valid in period (P1264) but I'm not sure that a day of the week is quite how that property is intended to be used. Thryduulf (talk: local | en.wp | en.wikt) 23:48, 24 May 2016 (UTC)

  •   Support - this seems like a useful property to have. Should there maybe be a "month of the year" property also - for example Thanksgiving (Q13959) -> November (Q125) (qualifier = US)? Anyway I definitely support this one. ArthurPSmith (talk) 18:02, 27 May 2016 (UTC)
  •   Comment It makes sense as a qualifier for "frequency of publication"-like properties – not so much for the other examples given, IMO. "The Queen's birthday is on some Saturday" is not very helpful information - there is literally an infinity of Saturdays that would fit that description.--Anders Feder (talk) 21:28, 28 May 2016 (UTC)
    • Actually knowing the Queen's official birthday (which incidentally is distinct from her actual birthday) is always on a Saturday is useful, as it can be linked with other events that happen on Saturdays for example, or dates determined with a day of week + date range. I agree that is only minorly useful for this example, but something like e.g. Jumu'ah (Q924037)Friday (Q130) it is a much more significant aspect of the item. Thryduulf (talk: local | en.wp | en.wikt) 23:24, 29 May 2016 (UTC)

@Thryduulf, ArthurPSmith, Anders Feder:   Done now day of week (P2894) --Lymantria (talk) 05:31, 8 June 2016 (UTC)

Extraction site locations

   Not done
DescriptionLocations where organisation is involved in material extraction from underground
Representsoil field (Q211748)
Data typeGeographic coordinates
Example 1MISSING
Example 2MISSING
Example 3MISSING

(Just putting location as a statement on the Eni item does not let understand that it's not the company's offices location that is being described, but the location of the extraction points.) GiordanoArman

  •   Weak oppose without an example, but if I've understood this properly I think using operator (P137) and/or owned by (P127) on the item about the extraction site (which should have coordinates) would be a better way of doing this. Large corporations like BP (Q152057) would have massively long lists for this property, and it would make it harder to track changes in ownership that way too. Thryduulf (talk: local | en.wp | en.wikt) 10:31, 8 May 2016 (UTC)
  • Oppose per Thryduulf. --Izno (talk) 14:03, 8 May 2016 (UTC)

@GiordanoArman:   Not done, no support. --Srittau (talk) 21:22, 11 June 2016 (UTC)

foreign aid outflow

   Not done
Descriptionmoney, food, or other resources flowing to other countries in the form of aid
Data typeNumber (not available yet)
Allowed valuesCurrency amounts
Example$43 billion
Robot and gadget jobsA robot should update this

Foreign aid outflow is an important macroeconomic statistic Mcnabber091 (talk) 07:24, 17 December 2015 (UTC)


  Comment @ArthurPSmith: What do you think of this property and the ones I proposed below? Your feedback would be appreciated. Thanks Mcnabber091 (talk) 06:10, 18 December 2015 (UTC)

  Comment @Mcnabber091: make sure what you are proposing actually makes sense as a property - I don't understand how the "list of sovereign wealth funds" could work as a property for example. Maybe you what you want could be done better by creating that as an item representing a class, and then other items could be indicated to be of that type via instance of (P31)? It's good if possible to pull an example from an existing wikipedia article (the article would have an item ID, the property would then allow a claim on that item with a certain value). If you have some good examples it makes it clear what a property is for and how it would work. Also I would expect to see a number of different people showing support for the property before it could be created, and at least a week's time for discussion. ArthurPSmith (talk) 15:11, 18 December 2015 (UTC)

  Comment @ArthurPSmith: Thanks for the feedback. I will update the all of the properties I proposed with more concise descriptions and examples. Mcnabber091 (talk) 04:20, 20 December 2015 (UTC)

  Comment This statistic would be a currency amount and would have qualifiers by date Mcnabber091 (talk) 04:49, 20 December 2015 (UTC)

  •   Comment Isn't this just "Government expenditure by type"? Why does it need a special property? Joe Filceolaire (talk) 16:34, 22 December 2015 (UTC)
    • @Filceolaire: This property would be for the total government + private foreign aid donation. Mcnabber091 (talk)
      • Isn't this usually a pretty small figure? Are macroeconomists really that interested? Isn't 'remittances' more important that private foreign aid in most cases? Joe Filceolaire (talk) 22:01, 23 December 2015 (UTC)
        • Some developed countries have aid outflow of 1% of GDP. Although this is small, I believe it has a macroeconomic effect and should be a Wikidata property. I think that remittances inflow/outflow should have it's own property, Mcnabber091 (talk) 09:21, 25 December 2015 (UTC)
  •   Comment What I miss here is an example of how this is supposed to be used, and what source is to be used. Apparently this is not about government expenses alone, which will make it difficult to find figures that are reliable and compatible between countries. Lymantria (talk) 13:40, 22 April 2016 (UTC)

@Mcnabber091:   Not done, without a working example this can not be evaluated and implemented correctly. --Srittau (talk) 21:59, 11 June 2016 (UTC)

@Srittau: You should check out and scroll to the bottom of a country page to see an example. Mcnabber091 (talk) 02:15, 12 June 2016 (UTC)

foreign aid inflow

   Not done
Descriptionmoney, food, or other resources flowing into a country in the form of aid
Data typeNumber (not available yet)
Allowed valuesCurrency amounts
Example$41 billion in 2013
Robot and gadget jobsA robot should update this

Foreign aid inflow is an important macroeconomic statistic for developing countries Mcnabber091 (talk) 07:24, 17 December 2015 (UTC)

  •   Comment Foreign aid inflow is a basic statistic and shouldn't be hard to incorporate into Wikidata. Mcnabber091 (talk) 04:50, 20 December 2015 (UTC)
  •   Comment Isn't this a "Government revenue by source"? Why does it need a special property? Joe Filceolaire (talk) 16:30, 22 December 2015 (UTC)
  •   Comment Not 100% of foreign aid inflow goes directly to the government. This property includes private and publc recipients of aid. Sometimes aid flows into the economy without going first through the government. Mcnabber091 (talk) 09:26, 25 December 2015 (UTC)
  •   Support. Joe Filceolaire (talk) 18:49, 29 December 2015 (UTC)
  •   Comment What I miss here is an example of how this is supposed to be used, and what source is to be used. Apparently this is not about government to government alone, which will make it difficult to find figures that are reliable and compatible between countries. Lymantria (talk) 16:51, 2 May 2016 (UTC)

@Mcnabber091:   Not done, without a working example this can not be evaluated and implemented correctly. --Srittau (talk) 22:00, 11 June 2016 (UTC)

@Srittau: You should go to, pick any of the countries, and then scroll to the bottom of the page. There you should see a good example of how the property for foreign aid inflow should look. All of the properties I have proposed are meant to replication econ factbook (see the United States article for the complete complete statistics). BTW thanks for all the suggestions on the properties so far. Mcnabber091 (talk) 01:36, 12 June 2016 (UTC)

State and local combined government revenue

   Not done
Descriptioncombined total of state and local government revenue.
Data typeNumber (not available yet)
Allowed valuesNumber, currency. positive
Example$98 billion
Robot and gadget jobsA robot should update this

state and local finance is important for macroeconomic analysis Mcnabber091 (talk) 23:06, 20 December 2015 (UTC)


  Comment this property should be a single $ amount Mcnabber091 (talk) 23:06, 20 December 2015 (UTC)

Mcnabber091 Is this a number issued by reliable sources or is it just calculated? If it is calculated then we could calculate it ourselves and we don't need this property. Joe Filceolaire (talk) 16:07, 22 December 2015 (UTC)
@Filceolaire: I think that this property is needed because I think it will be hard to internally calculate it within Wikidata. Most state and local governments don't easily provide this information, so it would be unrealistic that Wikidata could sum up every city and localities' budget entered into Wikidata and produce a good estimate. State and local combined government spending can be found for the United States in this report: I like this type of estimate because it reliable. Mcnabber091 (talk) 09:46, 25 December 2015 (UTC)
I think that for calculable properties, I would suggest having two references: the statistic from directly for a source and our internally calculated statistic. That way the reader would have both. Mcnabber091 (talk) 21:16, 25 December 2015 (UTC)
Then I   Support this. Joe Filceolaire (talk) 18:34, 29 December 2015 (UTC)

  Question @Mcnabber091: This is supposed to be a USA only property? Lymantria (talk) 06:27, 7 May 2016 (UTC)

  • @Lymantria:This property should apply to every country. It combines all of the government budgets into a total budget and then a balance is determined. Some countries provide this information such as the United States.Mcnabber091 (talk) 06:10, 17 May 2016 (UTC)
    • @Mcnabber091: The problem is that not all countries have the same form of subdivision as the USA have. Actually, many of them don't. Shouldn't the property read "combined revenue for all subnational governments within a country"? Same for the propositions on "state and local governments combined" below. Lymantria (talk) 07:11, 18 May 2016 (UTC)

@Mcnabber091:   Not done, open questions, no support --Srittau (talk) 22:01, 11 June 2016 (UTC)

@Srittau: @Lymantria: I agree Lymantria that it should be renamed to 'all subnational governments'. If I make that change will this work?Mcnabber091 (talk) 01:31, 12 June 2016 (UTC)

State and local combined government expenditure

   Not done
Descriptioncombined total of state and local government expendtiure
Data typeNumber (not available yet)
Allowed valuesNumber, currency. positive
Example$92 billion
Robot and gadget jobsA robot should update this

state and local government financial data is important for economicsMcnabber091 (talk) 23:06, 20 December 2015 (UTC)


  Comment this property should be a single amount Mcnabber091 (talk) 23:06, 20 December 2015 (UTC)

  Comment I think that for calculable properties, I would suggest having two references: the statistic from directly for a source and our internally calculated statistic. That way the reader would have both. Mcnabber091 (talk) 21:16, 25 December 2015 (UTC)

  Question @Mcnabber091: This is supposed to be a USA only property? Lymantria (talk) 06:56, 7 May 2016 (UTC)

  • @Lymantria:This property should apply to every country. It combines all of the government budgets into a total budget and then a balance is determined. Some countries provide this information such as the United States.Mcnabber091 (talk) 06:10, 17 May 2016 (UTC

@Mcnabber091:   Not done per Wikidata:Property proposal/state and local combined government revenue --Srittau (talk) 22:02, 11 June 2016 (UTC)

state and local combined government budget balance

   Not done
Descriptioncombined budget balance for all state and local governments within a country. revenue minus expenditure.
Data typeNumber (not available yet)
Allowed valuesNumber. positive or negative
Example-352 billion
Robot and gadget jobsA robot should update this

State and local government finance is important for economic analysis Mcnabber091 (talk) 23:06, 20 December 2015 (UTC)


  Comment this property should be a single amount positive or negative. qualifer by year, currency. Mcnabber091 (talk) 23:06, 20 December 2015 (UTC)

  Comment I think that for budget balance, and other calculable properties, I would suggest having two references: the statistic from directly for a source and our internally calculated statistic. That way the reader would have both. Mcnabber091 (talk) 21:16, 25 December 2015 (UTC)

  Question @Mcnabber091: Repeat the question: is this supposed to be a USA only property? Lymantria (talk) 06:57, 7 May 2016 (UTC)

  • @Lymantria:This property should apply to every country. It combines all of the government budgets into a total budget and then a balance is determined. Some countries provide this information such as the United States.Mcnabber091 (talk) 06:10, 17 May 2016 (UTC)

@Mcnabber091:   Not done per Wikidata:Property proposal/state and local combined government revenue --Srittau (talk) 22:02, 11 June 2016 (UTC)

@Lymantria: @Srittau: If I change the description to 'all subnational governements' would that work? Mcnabber091 (talk) 01:33, 12 June 2016 (UTC)

Minimum age

   Done: minimum age (P2899) (Talk and documentation)
Descriptionminimum age for, for example, movie certification
Data typeNumber (not available yet)
Example18 certificate (Q4557532) -> 18 (qualified as "years")
Sourcevarious film certification authorities

We need to think of a way of describing the items (e.g. 18 certificate (Q4557532)) used by media or game certification properties (e.g BBFC rating (P2629)). Unless anyone has a better approach? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 21:04, 22 March 2016 (UTC)

  • I would have an item with a particular rating ("E for Everyone"/"M for Mature") with this property on it. I think that's what you're proposing, so I support. No need to qualify--just make it number with units? --Izno (talk) 16:09, 30 March 2016 (UTC)
  • On the one hand, I   Support the property, if only we could find a more descriptive label. On the other hand, I'd   Comment that we need a way to indicate the nature of this "minimum age" (we really need a better name), whether it means that parental guidance is suggested for ages above, not recommended for ages below but not restricted, restricted to ages above unless accompanied by an adult, restricted to ages above, or adult content. (In Hungary we've only had the first two. It's fascinating to me that there are Western countries where you won't be admitted to a screening if you're not old enough.) – Máté (talk) 16:42, 30 March 2016 (UTC)
    Come to think of it, even the rating of "R" in America allows for parental supervision, if memory serves. --Izno (talk) 17:09, 30 March 2016 (UTC)
  • Freebase uses Minimum Accompanied Age and Minimum Unaccompanied Age. Can we do similar? -- Netoholic (talk) 02:54, 12 April 2016 (UTC)
    • "Minimum Accompanied Age" is only used on EIRIN and RSVR (RARS), nine ratings, and in all cases it's set to the same value as "Minimum Unaccompanied Age" (which is used 121 times). Even if the system made sense and was used, it doesn't begin to describe properly the complexity of the issue. On the other hand "minimum unaccompanied age" already sounds somewhat better than just "minimum age". We may have a "minimum unaccompanied age" property (this one), and an "accompanying criterion" property/qualifier with item datatype with possible values: n/a (Q929804) (when the value is 0)/parental guidance suggested/not recommended/accompanying adult required/restricted/adult content. – Máté (talk) 05:00, 12 April 2016 (UTC)
      • "accompanying criterion" would have to be a String monolingual text type. If we're going to do that, just label it "restriction" to make it more widely useful. We can also make an opposite property called "allowance". Those two along with "minimum age" could then all be used in various other contexts, like video games or any other age-limited activities. -- Netoholic (talk) 08:18, 12 April 2016 (UTC)
        • I assume, you're thinking of monolingual text, not string. In any case, item datatype would have more (research) use, as categories with similar nature could be easily grouped together, or even weighted according to their "strictness" when calculating an average rating (of a work/a system/a year) to be used as a variable. – Máté (talk) 08:42, 12 April 2016 (UTC)
          • The problem is the very subtle wording differences would make managing items pretty daunting. See en:Motion picture rating system for all the variations. I'd rather manage them as values within a single item for each rating. The restriction text isn't conceptually a different "thing" from the rating, its descriptive. If you want to group and weight, use the ratings items themselves, not the "accompanying criterion". -- Netoholic (talk) 16:10, 12 April 2016 (UTC)
            • The article you linked is exactly the example how you can easily group them and use the item-system on them, just take a look at the comparison table. This is invaluable information that you need for grouping and weighting, the rating items themselves are not enough. The two, however, are not mutually exclusive. We can store both the item-type nature of the rating and the exact wording of their description quoted from the corresponding law or guidelines. All I'm saying that the item-type is something that is useful. Especially given, that the description often doesn't even gives you the true nature of the rating, but you have to go back to the law and find about it in the context where it is defined. – Máté (talk) 16:59, 12 April 2016 (UTC)
  •   Support as proposed. Unit should be year. Use qualifiers if needed.
    --- Jura 00:10, 17 April 2016 (UTC)

@Pigsonthewing, Máté, Jura1:   Done as proposed. While there is much more that could be modelled on individual rating items, considering the various different rating systems around the world, this is a good start. Also, this property could be useful in other contexts. --Srittau (talk) 22:51, 11 June 2016 (UTC)

HKMDb film ID

   Done: HKMDB film ID (P2883) (Talk and documentation)
Descriptionfilm IDs on Hong Kong Movie Database (Q5369036)
RepresentsHong Kong Movie DataBase (Q5369036)
Data typeExternal identifier
Domaincreative work
Allowed valuesup to 5 digits
ExampleThe Bride with White Hair (Q987752)7660
SourceHong Kong Movie DataBase (Q5369036),, Template:HKMDB title (Q8140673)
Formatter URL$1

The corresponding template Template:HKMDB title (Q8140673) has already been created on 8 different wikis and is used on more than 500 pages in the English Wikipedia. HKMdb contains more than 18,000 movies and has a much better coverage of Hong-Kong and Chinese movies than IMDB. Koxinga (talk) 23:41, 19 May 2016 (UTC)


@Koxinga, Pigsonthewing:   Done now HKMDB film ID (P2883) -- Lymantria (talk) 09:30, 1 June 2016 (UTC)

frequency of publication

Descriptionstandard publication interval for daily or periodical publications
Data typeNumber (not available yet)
Template parameter"frequency" in en:template:Infobox journal
Domainnewspapers, magazines, periodicals, journals, annuals
Allowed valuespositive numbers with units of time
ExampleReader's Digest (Q371820) → 1 month; Connotations (Q501760) → 6 month; The Sunday Times (Q221986) 1 week; Mainichi Shinbun (Q1136866) 0.5 day

Looking through the constraint violations for event interval (P2257) I noticed that t was being used to indicate a publication frequency on several items, but the domain for that property is restricted to events and I think there is sufficient difference between "this item reoccurs with this nominal frequency" and "this items releases an issue/edition on this standard schedule" that a new property is better than broadening the old one. See also the "day of week" proposal below that could be used as a qualifier to this property.

I debated a couple of issues with myself when proposing this:

  • number or item datatype: some publication schedules are more complex than a standard interval, e.g. for a time period the Reader's Digest published 10 issues a year, but 0.8333 year (or 10/12 year if/when we get fraction support) can represent that. If people prefer the item route though then I will support that. Another option would be for three new properties: issues per week, issues per month and issues per year. That is getting too complicated though I think and I would need convincing before supporting that (I am open to being convinced though).
  • "publication" or "publication or broadcast": is something like this desired to represent the frequency of TV and radio programmes? As I don't really know the answer to that I've decided not to include this in my initial proposal but I'll happily and readily support that expansion if there is a desire. Thryduulf (talk: local | en.wp | en.wikt) 23:27, 24 May 2016 (UTC)
  • There's some problem with this property in that some publications do not have a regular publication schedule. I had a discussion on en.WP just this week about the subject. It may be better to infer this property by making items for the published works and linking them via "series" or "part of" or similar. --Izno (talk) 10:17, 25 May 2016 (UTC)
    • That could work for less frequent publications, but would not be realistically valid for anything publishing more frequently than about weekly, and items for each issue of daily newspapers would overwhelm wikidata very quickly. The alternative I discussed above using "item" could work here (have an item for "irregular schedule" or something), or maybe we should title this "regular publication schedule" and set it to no value for those publications that don't have a regular schedule? Thryduulf (talk: local | en.wp | en.wikt) 10:32, 25 May 2016 (UTC)
  •   Comment Isn't this proposal on publication interval (how much time between two publications) rather than publication frequency (how many times per amount of time)? I think I would prefer a real "publication frequency" property, as it is more flexible. For most newspapers you could say "6 per week", which is more accurate than a "at a 1 day interval". Lymantria (talk) 15:56, 31 May 2016 (UTC)
    • That would indeed be more flexible, but I am unsure how a property with two variables can be stored in Wikidata's structure? I came accross this same issue with ratios and gradients, for the first we fixed it as a 1:n ratio (e.g. using 1:-2 rather than 2:1) and for the latter we restricted it to a percentage. Here it would I think require four properties - issues per day, issues per week, issues per month, issues per year - making it harder to query and harder to maintain. Thryduulf (talk: local | en.wp | en.wikt) 23:47, 31 May 2016 (UTC)
  Support This would be a very nice property to have. It resembles the predicate dcterms:accrualPeriodicity. --Andrawaag (talk) 19:13, 2 June 2016 (UTC)
I share Lymantria preference for a "publication frequency" property, also to be able to store a value like "irregular". --Andrawaag (talk) 19:16, 2 June 2016 (UTC)
@Andrewag, Lymantria: How do you get around the one variable limit (see my comment above)? Thryduulf (talk: local | en.wp | en.wikt) 09:49, 3 June 2016 (UTC)
@Thryduulf: I think this should be solved by using a qualifier "per/relative to" (but not relative to in the sense of relative to (P2210)). Examples would be:
⟨ Journal of the American Academy of Audiology (Q2840472)      ⟩ "frequency of publication" search ⟨ "10" ⟩
"per" search ⟨ "1" year (Q577)  ⟩
⟨ Welsh Music History (Q7981948)      ⟩ "frequency of publication" search ⟨ "1" ⟩
"per" search ⟨ "2" year (Q577)  ⟩
Apart from that, still I am thinking over this frequency issue becoming harder to query and maintain. I think that is indeed a risk. Lymantria (talk) 10:53, 3 June 2016 (UTC)
I suppose we could partially (but not completely) alleviate that with "per day", "per week", "per month", "per year", etc items used as units of frequency but again that might make querying more difficult as the number of time periods is potentially limitless, so I still prefer the original proposal (although your point about the name is good). Thryduulf (talk: local | en.wp | en.wikt) 15:56, 3 June 2016 (UTC)
  Support the original proposal. Lymantria (talk) 05:39, 8 June 2016 (UTC)

@Thryduulf, Andrawaag:   Done Now publication interval (P2896) --Lymantria (talk) 09:43, 8 June 2016 (UTC)

Eldoblaje Movie ID

DescriptionIdentifier of dubbing in the website, a database of Spanish dubbings.
Data typeExternal identifier
Domaininstance of dubbing (Q187657)
Allowed values\d+
ExampleThor (Q24046091)22608
Formatter URL$1
Motivation is a database that stores dubbing data about thousands of movies and Spanish voice actors. Escudero (talk) 16:17, 17 May 2016 (UTC)

  •   Support Looks good to me. --Izno (talk) 20:41, 6 June 2016 (UTC)

@Escudero, Izno:   Done now Eldoblaje Movie ID (P2897) --Lymantria (talk) 05:32, 10 June 2016 (UTC)


   Not done
DescriptionFrame rate
Data typeNumber (not available yet)

The expected frame rate. Dispenser (talk) 00:13, 2 May 2016 (UTC)

last proposal: Wikidata:Property proposal/Archive/47#frame rate --Pasleim (talk) 09:57, 2 May 2016 (UTC)
  • @Máté: is this what you were looking for?
    --- Jura 15:28, 19 May 2016 (UTC)
Not exactly. Though The Hobbit trilogy was shot in 48 fps, most people watched it in 24 fps at the movies, while in 23.976 fps or 25 fps at home. This could be "filmed at frame rate". But my proposal was something different, a qualifier for duration (at what frame rate was the given duration measured), since The Hobbit has a different running time on home video in the U.S., in Europe or at the cinema. – Máté (talk) 20:32, 19 May 2016 (UTC)

@Dispenser:   Not done, no support --Srittau (talk) 22:59, 11 June 2016 (UTC)

Shakeosphere person ID

DescriptionThe Shakeosphere site is a scholarly project of the University of Iowa, collecting edition information on publications in English to 1800. It has around 54K person identifiers (authors and others).
Data typeExternal identifier
ExamplePeter Abelard (Q4295) -> 39
Formatter URL$1
Robot and gadget jobsThe Shakeosphere persons are a mix'n'match catalog.

Good scholarly information on pages in timeline form, and synergy with other early-modern British datasets. Charles Matthews (talk) 10:58, 24 May 2016 (UTC)

  • Looks good, and the early modern project rushes ever on. Pity they don't seem to have standard identifiers we can crosswalk to - not yet, at least. Andrew Gray (talk) 20:07, 24 May 2016 (UTC)
  •   Support --Magnus Manske (talk) 09:47, 26 May 2016 (UTC)
  •   Support. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:40, 26 May 2016 (UTC)

@Charles Matthews, Andrew Gray, Magnus Manske, Pigsonthewing:   Done Now Shakeosphere person ID (P2886) and ready for connection with Mix'n'match. --Lymantria (talk) 17:43, 2 June 2016 (UTC)


   Not done
DescriptionThe chancellor is the name for the highest-ranked university official in many english-speaking universities. Synonyms include President. Rector (P1075) is less commonly used in English-language schools for the same item.
Data typeItem
Domainuniversities and colleges have chancellors (or presidents) and provosts
ExampleUniversity of California, Berkeley (Q168756)Nicholas Dirks (Q7025311)

Items about universities should list the top officers; particularly in the US, this is generally the Chancellor and Provost, or President and Provost. The chancellor (or president) is in charge of the whole university; provost is in charge of the academics. There is not another property that fits this information that isn't so generic as to be meaningless, as far as I can tell (president right now redirects to chairperson, but the chairperson at a university has an entirely different meaning, typically meaning the chair of the governing board). I am hesitant for instance to list the chancellor of US universities using the current property choices, because I think it's misleading and confusing (since the chair is an entirely different person/role, and in the US "Rector" often doesn't actually exist as a title, but there's no way to indicate this). Note: this is needed in Wikipedia as it's a position that occasionally changes, and I often see wrong information in articles about who is who. Phoebe (talk) 04:41, 18 April 2016 (UTC)

The datatype should be item, not string. --Yair rand (talk) 04:51, 18 April 2016 (UTC)
fixed! thanks -- Phoebe (talk) 04:54, 18 April 2016 (UTC)
  •   Support as proposed. Thryduulf (talk: local | en.wp | en.wikt) 12:03, 19 April 2016 (UTC)
    • In the UK, and apparently often in other Commonwealth nations, Chanellor is a figurehead position and the Vice-Chancellor (or sometimes President) is the one who runs the institution as a chief executive. This should probably be represented somehow, but I don't immediately know if a separate Vice-Chancellor property is the best way to do that. Thryduulf (talk: local | en.wp | en.wikt) 12:12, 19 April 2016 (UTC)
  •   Oppose. Use existing role/ position held properties. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:00, 29 April 2016 (UTC)
  •   Support as proposed. In common wealth countries it is required. Also the term "principal" should also be added. Balajijagadesh (talk) 10:28, 3 May 2016 (UTC)
  • Pigsonthewing what properties would you use to represent this relationship for university items (e.g. UC Berkeley above)? Thanks! -- Phoebe (talk) 15:14, 2 May 2016 (UTC)

  Not done No consensus at this time. It appears a solution to use is corporate officer (P2828). --Izno (talk) 15:45, 6 June 2016 (UTC)


   Not done
DescriptionThe provost is the top academic officer at many universities, particularly in English-speaking parts of the world
Representsprovost (Q2114175)
Data typeItem
DomainColleges and universities have provosts
ExampleStanford University (Q41506)John Etchemendy (Q6231954)

Per above, particularly at universities in the United States, the Provost is the top academic position and there is not currently a way to show this relationship. Phoebe (talk) 04:48, 18 April 2016 (UTC)

  • Pigsonthewing, as per above, what would you use to indicate this for the university item? (I am not familiar with something that would work). thanks! -- Phoebe (talk) 15:19, 2 May 2016 (UTC)
    @Phoebe: Probably something like John Etchemendy (Q6231954) position held (P39) provost (Q2114175), qualifier of (P642) Stanford University (Q41506). --Yair rand (talk) 09:00, 3 May 2016 (UTC)
    @Yair rand: But that is the other way around. It is IMHO not easy to add the provost (or chancellor for that matter) to the item Stanford University (Q41506). Lymantria (talk) 11:17, 3 May 2016 (UTC) (typo fixed 19:38, 3 May 2016 (UTC))
    @Lymantria: Seems like that would be redundant data. If there's already a statement pointing in one direction, we don't need another pointing back. --Yair rand (talk) 21:37, 3 May 2016 (UTC)
    For companies we have "chief executive officer" lists, and the individuals then have "position held=chief executive officer, employer=company". I think it's often useful to have something pointing in both directions, because then queries can go both ways. --99of9 (talk) 02:33, 4 May 2016 (UTC)
    Queries can go both ways even without bidirectional statements. --Yair rand (talk) 04:44, 4 May 2016 (UTC)
    Ok -- but I'm not sure that makes sense, since the leaders of the university are an important part of the information about the institution, and having to figure out a complicated query to get that as part of the item doesn't seem super usable. -- Phoebe (talk) 01:35, 5 May 2016 (UTC)

  Not done No consensus at this time. It appears a solution to use is corporate officer (P2828). --Izno (talk) 15:45, 6 June 2016 (UTC)

Universal Product Code

   Not done
DescriptionUPC-A barcode number
RepresentsUniversal Product Code (Q1193047)
Data typeString
Allowed values[01678]\d{11} (12 digits)
  • Template:UPC (Q14397562) import (about 100 items)
  • MusicBrainz
  • MobyGames
  • Formatter URL$1 or$1 or$1
    First digit meaning
    UPC EAN Type
    0 00 Products
    1 01
    2 02 In-house: Encoded weight
    3 03 UPN codes (medicine/pharmaceutical)
    4 04 In-house use (loyalty cards/store coupons)
    5 05 Manufacturer coupons
    6 06 Products
    7 07
    8 08
    9 09 Manufacturer coupons


    From MusicBrains:

    Barcodes are numbers used as stock control mechanisms by retailers: as such they are highly standardised and well recognised, and form an invaluable identifier for communication between companies.

    So I want to add UPCs to video games as this is the only unique identifier that exists (like an ISBN). This property would make it easy for other systems to integrate Wikidata. Or the reverse such as searching Amazon by UPC.

    A cool thing we can do is auto create the barcode itself with Lua and image segments. The only question is should it be complicated by including the European superset, EAN?

    Dispenser (talk) 02:48, 17 March 2016 (UTC) Updated 01:46, 27 April 2016 (UTC)

    •   Weak oppose I believe Wikipedia/Wikidata simply isn't granular enough for this to make sense. Usually there are different barcodes for different versions, packaging sizes, multi-packs, regions, etc of a product. Even simply a change in packaging design may result in a new barcode. So that may result in an item having tens or hundreds of bar codes attached. For example, the Cheerios (Q2425291) item won't just have one value, it should have at least 71 according to Intgr (talk) 08:54, 17 March 2016 (UTC)
      For video games, UPCs are like a book's ISBN. As an example, The Orange Box (Q374556) is more findable by its UPC 0-14633-09852-5 (4,000 Google results) than by its ISBN (which is unusual to have) 0-7845-4306-2 (5 Google results). Dispenser (talk) 04:20, 25 March 2016 (UTC)
    • See Wikidata:Property_proposal/Archive/15#Barcode for a similar proposal from several years ago. - Nikki (talk) 09:39, 17 March 2016 (UTC)
    Updated with more rational. Dispenser (talk) 01:46, 27 April 2016 (UTC)
    •   Oppose Unfortunately it is not that easy. For example, the UPC/EAN for the Orange Box is not just "5030931058936". That is just the EAN for one particular edition. The German, French, and other language edition have a different code. The UPC for Cheerios is just for a specific package size. The 685 gram package listed here has a different number. UPC/EAN in general is much for volatile than, for example ISBN numbers. --Srittau (talk) 23:50, 5 May 2016 (UTC)
    •   Comment maybe we could attempt to determine a range of products where this could be helpful.
      --- Jura 15:21, 19 May 2016 (UTC)
    •   Support I am agree with @Intgr about this require more wikidata granularity but this would give a new powerfull functionality to wikidata as is explained here Probably it's early to start using wikidata in this way since nowdays there are not tools developed speciffically aimed at listing and searching objects ( would be the closest thing) but this could be reached little by little.Qupro (talk) 06:22, 12 May 2017 (UTC)

    @Dispenser:   Not done, no consensus. --Srittau (talk) 23:22, 11 June 2016 (UTC)

    L2 Cache

    DescriptionAmount of L2 Cache of the CPU in Bytes (or maybe bits?)
    Data typeNumber (not available yet)
    ExampleCore2 Quad Q6600 (Q15218754): 8MB
    SourceIntel website, for Intel processors
    Robot and gadget jobsRobots can gather info on the Intel website and fill the property with it.
    Proposed byMisterSanderson (talk)

    I want to add information to the CPU items, but there aren't enough properties to that. MisterSanderson (talk) 15:33, 21 November 2013 (UTC)

      Support I created an item for the processor. Hopefully we will get both bit and byte in the numeric datatype. --Tobias1984 (talk) 16:10, 21 November 2013 (UTC)
      Comment I think this is best left until we have the number type, and know how it will function, because we may only need one generic property for "size in bytes" which can be used as a qualifier. So, for example, we may be able to claim that a CPU has part L2 cache with qualifiers, quantity = 2, size = 8MB. Let's not create 100s of properties for speeds and dimensions. Danrok (talk) 02:11, 27 January 2014 (UTC)

      On hold To be re-opened when the right datatype is available.--Micru (talk) 13:57, 23 May 2014 (UTC)

    @MisterSanderson, Tobias1984, Danrok, Micru: So quantity datatype is now available, and we have megabyte (Q79735) and related items so in principle we can move ahead. However Danrok 's comments should be taken into account. We don't actually have any property (that I can find) that would give a generic "size" suitable for showing the number of megabytes, pixels etc - I think that property should be created first and may be all we need. Also I couldn't find an item corresponding to L2-cache: the only related one I think is cache (Q165596) (which has aliases include L2 and L3). Please look into some of these and let's have a more organized proposal for computing properties here! Is "has part" generally a suitable way to document the components of a CPU, camera, phone, etc, rather than having separate properties for each piece? That makes sense to me... ArthurPSmith (talk) 20:11, 11 December 2015 (UTC)

    We have a maximum capacity (P1083) property and this probably should become a generic "capacity" property, to be used to indicate how many things the item can hold, such as people, cars, bytes, etc. Danrok (talk) 01:27, 12 December 2015 (UTC)
    Well maximum capacity (P1083) seems very specific to counting people, has constraints that would need to be modified, it's been used by a Russian project etc (see discussion page). I think a new more generic property is warranted. Maybe let's add it under Wikidata:Property_proposal/Generic @Danrok: can you take care of making the proposal? Or maybe I will try later this week if it's not done yet. And @MisterSanderson: - what we are specifically proposing here is as an alternative to your property, to use the combination has part (P527) with an item for "L2 cache" and a qualifier specifying "capacity" (size) in MB or GB or whatever appropriate unit. This seems reasonable to me and I think it's preferable as a rule to use more generic properties where they make sense. What are your feelings on the matter? ArthurPSmith (talk) 20:34, 14 December 2015 (UTC)
    I could not understand what you said about not existing a generic size property.--MisterSanderson (talk) 16:34, 12 December 2015 (UTC)
    @MisterSanderson: I think he was thinking like
    ⟨ proc ⟩ has part (P527)   ⟨ L2 cache ⟩
    size search ⟨ whatever Mb ⟩
    . author  TomT0m / talk page 13:15, 11 January 2016 (UTC)

      Support Only allowed unit should be byte (Q8799). --Srittau (talk) 20:49, 16 May 2016 (UTC)

      Question @MisterSanderson: Why only L2 cache (and not L1)? Lymantria (talk) 05:23, 25 May 2016 (UTC)
    Lymantria, because it has been 3 YEARS waiting, I simply don't care anymore! If someone can please close this proposal so I can be left alone, I appreciate.--MisterSanderson (talk) 16:42, 4 June 2016 (UTC)
    I marked this proposal as withdrawn, since I don't expect it to get any traction if the proposer has lost interest. That said, I would support a new proposal. --Srittau (talk) 22:49, 6 June 2016 (UTC)

    Maximum sustained winds

    DescriptionMaximum sustained wind evaluated by agency.
    Data typeQuantity
    Template parameterSee en:Template:Infobox hurricane: 1-min winds, 5-min-winds, 10-min-winds
    DomainTerm — hurricanes/tropical cyclones, winter storms, European windstorms (think Freie Universität Berlin), etc.
    Allowed valuesNumeric — two-digit, three-digit
    ExampleHurricane Sandy (Q264), Hurricane Irene (Q187621), Hurricane Floyd (Q2110774), Hurricane Gloria (Q3045816)
    Robot and gadget jobsManual to avoid human error.
    There are differences in the usage of the number of minutes in different countries; this would result in misleading information, might need to be disambiguated. Also, consider adding a "classification" property — i.e. "Category 1," "super typhoon," "severe tropical storm," etc.? Hurricanefan24 (talk) 13:50, 18 March 2013 (UTC)
      Support --Goldzahn (talk) 19:10, 18 March 2013 (UTC)
    Wrong data type. Should be deferred until QuantityValue implemented. --EugeneZelenko (talk) 13:43, 19 March 2013 (UTC)
    Appears that's been fixed.
    •   Support Sven Manguard Wha? 20:19, 15 June 2013 (UTC)
    •   Question I'm a little confused about what is proposed here. There are three parameters noted, is this one property with a mandatory qualifier (duration (P2047) or determination method (P459) perhaps?) or three properties? Thryduulf (talk: local | en.wp | en.wikt) 14:36, 13 January 2016 (UTC)
      • @Thryduulf: Yes there would be mandatory qualifiers for duration because the quantity is without rigorous definition otherwise. Additionally it should be qualified with the quality of the estimate - operational, postanalysis, measured with reconnaisance, etc.--Jasper Deng (talk) 23:54, 31 May 2016 (UTC)
    • generally support though also need something for maximum wind gusts. needed for an item like January 2016 United States winter storm (Q22222608). Maybe this property could be generalized and a qualifier be used to indicate if it's wind gusts or sustained winds. Separate properties would also be ok. Aude (talk) 00:59, 24 January 2016 (UTC)
    •   Oppose Use speed (P2052) with measurement scale (P1880) qualifier pointing to maximum sustained wind (Q6795923) (or split item that into 1-min, 5-min, etc). -- Netoholic (talk) 01:05, 20 April 2016 (UTC)
      • I don't think speed (P2052) should be used in this manner to record the wind speed produced by a storm system, as I would expect a speed property on a storm item to record how fast the system was moving. Something like
        ⟨ storm ⟩ has part (P527)   ⟨ maximum sustained wind (Q6795923)      ⟩
        speed (P2052)   ⟨  50 mph ⟩
        might be OK, but I think a separate property would be clearer. Thryduulf (talk: local | en.wp | en.wikt) 10:40, 20 April 2016 (UTC)
        • I would not oppose if this were changed to something more generic like "wind speed" with "Maximum sustained" etc as qualified, point in time, references.. -- Netoholic (talk) 13:16, 24 April 2016 (UTC)
        • @Netoholic: "Speed" in the context of a storm is generally considered to be the velocity of the storm in total, so I agree with Thry on this point. @Thryduulf: The more correct property would be has quality (P1552). --Izno (talk) 13:59, 16 May 2016 (UTC)
      • @Netoholic: No, qualification using measurement scale (P1880) would be incorrect, because "maximum sustained winds" is seldom something that can be directly measured. For this reason, meteorological agencies tend to limit it to integer multiples of 5 knots and even in the presence of direct measurements of wind, always acknowledge that these are estimates.--Jasper Deng (talk) 23:51, 31 May 2016 (UTC)
    • Support either a "wind speed" property or the "has quality" proposal above, by Thryduulf. I think I would slightly prefer a dedicated wind speed property, qualified as desired, with "point in time"/"end time"/"start time"--maximum speed can be derived from that information. --Izno (talk) 13:59, 16 May 2016 (UTC)
    •   Support However, I disagree that this should not be automatic. I hope we can import the entire HURDAT database and other track files to Wikidata where they can be more easily referenced for Wikipedia's purposes.--Jasper Deng (talk) 23:45, 31 May 2016 (UTC)

      Done as maximum sustained winds (P2895). @Hurricanefan24, Goldzahn, Sven Manguard, Izno, Jasper Deng: ready to use. -- Hakan·IST 08:11, 8 June 2016 (UTC)

    @HakanIST: I am not sure this had consensus as "max sustained wind speed". I only supported a more generic property, and the same with Thryduulf and Netoholic. Sven's !vote is ancient as is Gold's, and Hurricane fan is retired. @Jasper Deng: would you support a more generic property? --Izno (talk) 14:39, 10 June 2016 (UTC)
    No, for reasons that I already gave above.--Jasper Deng (talk) 19:15, 10 June 2016 (UTC)
    I didn't actually support (or oppose) this proposal. I opposed using speed (P2052) and commented that a specific property would be better than using a construction involving has quality (P1552), but didn't put my name to this specific proposal as I was not certain about the approach to recording the duration of measurement (I think it is very important that it is somehow noted so like can be compared with like). I hadn't seen (until now) Jasper Deng's comment about measurement only being an estimate which makes me wonder more about how the estimations are performed, and how that relates to this property - are there multiple methods? If so should they get qualifiers or separate properties? I don't know if I oppose it finally, but I don't support it as it and my thinking at the moment and wouldn't have called this a consensus at this point. Thryduulf (talk: local | en.wp | en.wikt) 21:50, 10 June 2016 (UTC)
    Read any National Hurricane Center advisory and you'll see. See for example [1]. Not only are there multiple ways of measuring or estimating it, but we also need a qualification using 1-, 3-, or 10-minute.--Jasper Deng (talk) 21:57, 10 June 2016 (UTC)

    knowledge crystal result

    Descriptionquantitative result of a knowledge crystal
    Representsknowledge crystal result (Q23893445)
    Data typeNumber (not available yet)
    Template parameternot yet available
    Domainitems that are instance of (P31) knowledge crystal (Q23691297)
    Allowed valuesnumeric values -INF to INF
    ExampleDisease burden of air pollution (Q23680551) → 3297000 number of infected (Q23696805); qualifiers: of (P642) death (Q4), point in time (P585): 2010, location (P276) Earth (Q2)
    Sourceexternal websites producing knowledge crystals, e.g. Opasnet (Q7095608)
    Robot and gadget jobsnot yet available

    Knowledge crystals (knowledge crystal (Q23691297)) are versatile information objects that answer specific research questions. The answer to this question often has a written description, but typically the core of an answer is a quantitative result. Results of different knowledge crystals are measured in different units, and only a fraction of knowledge crystals have a result that can directly be represented using an existing property. In the example, the suggested property "knowledge crystal result" could be replaced by number of cases (P1603). But if the disease burden had been measured using quality-adjusted life year (Q614165) with unit quality-adjusted life year (Q23893355), there would not be a good property available.

    The point is that the relations between the question and the answer of a knowledge crystal may be very different from each other and such that there are no existing properties describing them. Therefore, it makes sense to create one property that is generic enough to describe the relation of any question and answer of a knowledge crystal, rather than creating a new property whenever a new kind of relationship emerges.

    Of course, specificity is needed in describing the relation of a question and an answer. This is done in several ways:

    • A specific unit is used to describe the result. It is easier to create units (which are items) than properties. E.g. quality-adjusted life year (Q23893355) and disability-adjusted life year (Q23893342) are specific units that could be used together with our example disease burden.
    • The result may vary e.g. between subgroups, as by observation year or country in the example. This variation and subgroup-specificity is handled by using qualifiers. Again, it is easier to use or create qualifying items than create properties.
    • The question must be known to be able to understand the result. This can be achieved by locating the question into the description of the item of a particular knowledge crystal. A benefit of this is that the question can be described in several languages.

    In this way, all essential information to interpret the question and answer of a knowledge crystal can be achieved by creating a single property as suggested here: knowledge crystal result. Jtuom (talk) 20:31, 20 April 2016 (UTC)

    •   Comment This appears to be exclusively associated with Opasnet and a Finnish health institute, and I don't see a large number of actual results that this would be applied to (at least not yet). Jtuom do you have an estimate of how many times this property could be used based on current data? We do have a generic property in the form of "numeric value" that could possibly be expanded in domain if the lack of suitable properties was really a problem, but I think it would be better to add these statements using existing properties and propose any missing properties as discovered in that process first. ArthurPSmith (talk) 19:01, 21 April 2016 (UTC)
      •   Comment @ArthurPSmith: I am trying to be proactive here. It is true that currently knowledge crystals are mostly used in National Institute for Health and Welfare (Q4354957), but they were designed in such a way that they could be used widely in science and also to support societal decisions. In my own work only (in the field of health impact assessment), I can see hundreds of knowledge crystals with immediate need. For example: disease burden estimates of dozens of risk factors in different countries; disease burden estimates of dozens or hundreds of major diseases in different countries; half-lives of toxic pollutants in the human body or in the environment; average levels of pollutants in food; average intakes of pollutants in different populations; exposure-response functions of major pollutants on causing specific diseases etc. All these are quantitative data that are of importance and interest to people interested in environmental health.
      But more importantly, I see knowledge crystals as an effective way to handle and update scientific information in any field, and therefore this should be seen as a generic property that has a large potential. If the property is found out to be useful and effective, it can actually reduce the need of new properties. So, I think it is now important to talk about the theoretical and potential merit of knowledge crystal result rather than think of its current popularity. If it fails in theory, I withdraw my suggestion. If it succeeds, it should be promoted in practice.
      I do not agree that numeric value (P1181) could be used instead of the suggested knowledge crystal result. It seems to be used with natural constants and other things that have a single true value. Knowledge crystal results are by nature uncertain, and their values may also change in time (if e.g. an infectious disease is eradicated, its disease burden may decrease dramatically -- or increase, if the health care system fails such as in the case of ebola). --Jtuom (talk) 08:45, 11 May 2016 (UTC)
    •   Comment An important feature of a knowledge crystal is that it has been produced by applying quality control and certain rules. The most important rules are that the result has been produced in an open website where anyone can see the contents and the rationale, and anyone can criticize the methods and premises used. However, critique must be based on science. Therefore, the most common critique is that the result is not in accordance with observed data.
    Because of this production process, knowledge crystals can be said to be established scientific information, and conversely, the property knowledge crystal result cannot be used about a result that has not been produced according to the rules. See also [2]. --Jtuom (talk) 11:39, 20 May 2016 (UTC)
    •   Support ArjaA (talk) 11:42, 20 May 2016 (UTC)
    •   Support I support experimenting with this but I think it needs more documentation also. This would be easier to manage if "knowledge crystals" were all coming from a single authority, but right now, the plan is to get knowledge crystal information from Opasnet or the Finnish health institute with which it partners. As I understand, the longer term plan is to use this term for any data produced by a "knowledge crystal" methodology. In the example above, a knowledge crystal query is producing data on how many people died of air pollution in all of earth in a given year. The same dataset that produced that information could produce variations of similar data that seem limitless by human standards. Opasnet is not the only organization that can provide this kind of data. So when this property is used, it would be used in the context of saying, "here is some interesting data", then apply the knowledge crystal property, then say what authority executed the knowledge crystal process. That means that this property probably would never be used without associating it with some authority.
    I have a little trouble thinking about this, because I cannot think of examples of Wikidata having been used in this way, but it seems logical to set up this property and experiment with using it. Blue Rasberry (talk) 20:28, 20 May 2016 (UTC)
    @Bluerasberry: You are quite right: this is experimenting at the moment. I also agree that at the moment knowledge crystals require an authority who curates the content and thus backs up the quality. However, I hope and believe that then when we get practical experience about this, it can be shown that the knowledge crystal methodology, when properly applied, tends to produce high-quality content even if the curator is not an expert (organisation) in the field. One reason to this is that there is a rule saying that relevant information must not be ignored. Therefore, if there are several knowledge crystal variants of the same topic, they must collaborate, resolve possible disputes, and produce a synthesis, thus preventing limitless variation. Such process has already been seen with encyclopedias: previously, there was a large variation of content within the hundreds of different encyclopedias; but Wikipedia brought all encyclopedists to a single page to discuss and synthesise a particular topic, thus reducing variation on what people can read about the topic. --Jtuom (talk) 15:11, 22 May 2016 (UTC)
    Yes, but the fact that Wikipedia is reducing variation on what people can read about the topic is, clearly, no guarantee for quality. Good content does not just happen, it takes work. - Brya (talk) 16:48, 22 May 2016 (UTC)
    I completely agree. I wanted to emphasize that reduced variation makes it possible to focus the work on one place, where it has the largest impact. --Jtuom (talk) 21:16, 22 May 2016 (UTC)

    Observed separation

       Not done
    Descriptionen:Observed separation
    Data typeQuantity
    Template parameteren:Template:Starbox relpos
    Example 1MISSING
    Example 2MISSING
    Example 3MISSING
    Proposed byGZWDer (talk) 11:04, 8 July 2013 (UTC)
    @Paperoastro: Shouldn't this have units (degrees?) Filceolaire (talk) 21:49, 24 February 2014 (UTC)
    Yes! I missed it! Thanks! I'd like to correct it, but I don't know what value put in "datatype" element of property documentation! I correct it. --Paperoastro (talk) 21:56, 24 February 2014 (UTC)
      Support Filceolaire (talk) 06:44, 26 February 2014 (UTC)

      On hold TBD when the right datatype is available.--Micru (talk) 13:00, 1 May 2014 (UTC)

    @GZWDer, Paperoastro, Filceolaire, Micru: Quantity datatype is available, with degree (Q28390) (or minute of arc (Q209426) arcsecond (Q829073) etc) as unit here. Can one of you confirm and update the proposal? ArthurPSmith (talk) 15:05, 2 December 2015 (UTC)

    •   Comment The format of the data for the separation of a double star would usually be in seconds of arc specified for a particular date of observation or a future epoch as computed from orbital elements. See WDS SUMMARY CATALOG KEY used in the w:Washington Double Star Catalog for example. For multiple star systems such as w:Epsilon Lyrae the value will change slowly over a millennium. For close binaries this data might prove difficult to keep up to date as it will change over time. For example, w:Sirius#Sirius_B has changed significantly since 1994. See the diagram at How to see Sirius B. There is also the problem of a lack of reliable sources for some star systems. --Mu301 (talk) 19:21, 5 February 2016 (UTC)
    Note: I just noticed that the IAU Style Manual states that "it is becoming common in astronomy to use the milliarcsecond as the unit." However, I would recommend arcsecond as a more useful unit. --Mu301 (talk) 21:58, 5 February 2016 (UTC)

    @GZWDer: Please add an example, using items, and a description. --Srittau (talk) 20:48, 8 May 2016 (UTC)

    @GZWDer:   Not done, orphaned proposal. If there is still interest in this, please create a new proposal. --Srittau (talk) 23:34, 11 June 2016 (UTC)

    probability density function

       Not done
    DescriptionMathematical density function for a probability distribution
    Representsprobability density function (Q207522)
    Data typeMathematical expression
    Template parameter"pdf" in en:template:Probability distribution
    Domainprobability distribution (Q200726)
    Allowed valuesLaTeX
    Examplenormal distribution (Q133871) → "\frac{1}{\sigma\sqrt{2\pi}}\, e^{-\frac{(x - \mu)^2}{2 \sigma^2}}"
    Robot and gadget jobsThis could potentially be filled from the templated in the English Wikipedia.

    To celebrate the arrival of the new data type "Mathematical expression" Finn Årup Nielsen (fnielsen) (talk) 13:30, 10 February 2016 (UTC)

    •   Comment We should convert TeX string (P1993) and keep this property for all mathematical formula. Snipre (talk) 15:43, 10 February 2016 (UTC)
      As noted for another suggestion, the "Mathematical expression" will render the LaTeX. — Finn Årup Nielsen (fnielsen) (talk) 10:28, 11 February 2016 (UTC)
    •   Comment Note there is not "one" formula for a probability distribution (Q200726). There are mean, variance, PDF, CDF. Another approach could be to use a generic 'formula' property and then use qualifiers.
    •   Support also "TeX string" probably should be rendered as a string given its name - and for example should not be converted to MathML as is apparently planned for some renderings of "Mathematical expression". Distinct properties for mathematical expressions on an item are definitely appropriate as this proposal (and others) makes clear. ArthurPSmith (talk) 13:29, 11 February 2016 (UTC)
    •   Support that would be cool. --18:22, 11 February 2016 (UTC)~---
    •   Question @Fnielsen, Snipre, ArthurPSmith: Is this property still needed? I see that normal distribution (Q133871) uses defining formula (P2534).--Micru (talk) 14:33, 4 March 2016 (UTC)
      Good question. :-) The formula shown is for the PDF, while every distribution has multiple formula, CDF, mean, variance, other moments, characteristic function, so what should we do about that? Have a whole series of properties? Or use qualifiers? If the equation is to be used in an infobox I suppose the qualifier approach is not well suited? — Finn Årup Nielsen (fnielsen) (talk) 21:50, 4 March 2016 (UTC)
      (To avoid confusion regarding "The formula shown is for the PDF": I have now added a second notation for a normal distribution (Q133871)) — Finn Årup Nielsen (fnielsen) (talk) 21:56, 4 March 2016 (UTC)
      @Fnielsen, Micru: I'm not sure this is really needed now. What might be needed are formulas for other properties of a distribution - for example if we wanted to create a "mean of probability distribution" property, or "variance", etc. But otherwise I think "defining formula" would be exactly what you would expect to see as a description of the distribution itself for, for example, logistic distribution (Q589603), binomial distribution (Q185547), Poisson distribution (Q205692) etc. ArthurPSmith (talk) 14:50, 11 March 2016 (UTC)
      @Fnielsen: I would have expected this
      ⟨ normal distribution (Q133871)      ⟩ defining formula (P2534)   ⟨ \frac{1}{\sigma\sqrt{2\pi}}\, e^{-\frac{(x - \mu)^2}{2 \sigma^2}} ⟩
      of (P642)   ⟨ probability density function (Q207522)      ⟩
      If we use defining formula (P2534) that way, this proposal (and the next one) is unnecessary? However, this seems to stretch the original intention of defining formula (P2534), where "a theorem or a law" was to be represented. But of course, in broader sense any formula could be seen as representing a theorem or a law. Lymantria (talk) 10:14, 25 April 2016 (UTC)
      As no reply has reached this in a month:   Oppose - unneeded. Lymantria (talk) 15:01, 30 May 2016 (UTC)
    •   Oppose per Lymantria. --Izno (talk) 12:11, 31 May 2016 (UTC)
    • Can be deleted for me. Snipre (talk) 13:54, 31 May 2016 (UTC)
    as I noted above, I don't think this is needed. Feel free to close as not done. ArthurPSmith (talk) 20:36, 9 June 2016 (UTC)

    @Fnielsen:   Not done, per discussion. defining formula (P2534) seems sufficient. --Srittau (talk) 23:41, 11 June 2016 (UTC)

    administrative body

    Descriptiongovernmental body of a country, administrative unit, or organization
    Representsgovernment (Q7188)
    Data typeItem
    Domainpolitical territorial entity (Q1048835) or organization (Q43229)
    Allowed valuesinstances of government (Q7188)
    ExampleGermany (Q183)Cabinet of the Federal Republic of Germany (Q313827)

    Strangely enough, while we have properties legislative body (P194) and office held by head of government (P1313), we have no way to link from a country or state to its administrative body. Srittau (talk) 00:21, 19 January 2016 (UTC)


    "We have executive body (P208) for the cabinets of countries" - Yes, see "P208:Cabinet of Israel" in the item about Israel. But do we have a property for the specific cabinet, in this case "Netanyahou III"? We also have to sort out how we should handle cases when two (or more) persons are cooperating in the leadership. Andorra and San Marino has two princes/capitani reggenti and Swedish municipalities can have several "kommunalråd". We have somewhere agreed to use an item for the group of "kommunalråd" but we have not sorted out how that should be implemented or how it affects for example these two European nations and other nations that has had two or more cooperating chief of states. (Competing chiefs is another question.) The Roman Empire, Russia, ancient Egypt, the biblical Israel and several other nations have at least temporary had cooperating kings. Ok, I went a little out of topic here, but the questions are related. We also have political parties with more than one leader. How the office of the prime minister will look like if such parties ever get the center seat in a government is an interesting question. -- Innocent bystander (talk) 08:55, 4 June 2016 (UTC)

      Withdrawn, executive body (P208) is what I was looking for. --Srittau (talk) 23:47, 11 June 2016 (UTC)

    reserve number

    Descriptionidentifier of a Canadian Indian reserve
    Data typeExternal identifier
    Template parameternone yet, but could be used if an infobox Indian reserve is created (now those Wikipedia articles use the same infobox as for cities, but reserves have specific information that would make having their own infobox useful, I plan to create one on fr.wp)
    DomainIndian reserve (Q155239)
    Allowed values/d+
    ExampleAlexander 134 (Q4718161) -> 06650
    Formatter URL$1

    As for P2865 (Band number) just approved above, Indian reserves in Canada are also all identified by a number. It would be good to have a property to add this number to Wikidata. For example the reserve Alexander 134 (Q4718161) has the number 06650 as we can see on the page for this reserve on the Aboriginal Affairs and Northern Development Canada's website. Amqui (talk) 16:59, 27 May 2016 (UTC)

    •   Support Seem usefull and well structured. --Fralambert (talk) 18:45, 27 May 2016 (UTC)
    •   Support Needed due to First Nation Profiles. Regards, Benoit Rochon (talk) 03:27, 28 May 2016 (UTC)

    @Amqui, Fralambert, Benoit Rochon:  Done now reserve number (Canada) (P2887) --Lymantria (talk) 07:01, 4 June 2016 (UTC)

    Thanks, Amqui (talk) 05:35, 5 June 2016 (UTC)

    reserved seats

       Not done
    DescriptionIn India, some of national (lok sabha)/ state constituencies have reservations for SC/ST (particular caste)
    Data typeItem
    Template parameterReservation (Lua Module en:Module:Infobox lok sabha constituency
    Domainen:elections in India (I am not sure which domain is more suitable)
    Allowed valuesen:Scheduled Castes, en:Scheduled Tribes
    ExampleIn Thiruvallur Lok Sabha constituency (Q7785493), I will use under qualifier in point in time (P585). currently using under classification of race (P2502).
    Proposed byMdmahir (talk) 14:41, 27 March 2016 (UTC)

    currently using under classification of race (P2502) (classification of) eg. Thiruvallur Lok Sabha constituency (Q7785493), but later found that property can be used only in sports Mdmahir (talk) 14:41, 27 March 2016 (UTC)

    •   Question @Mdmahir: Let me try to understand this. In a constituency there are a number of reserved seats for certain minorities and you want to express this using statements, correct? --Srittau (talk) 13:43, 23 April 2016 (UTC)
    • In general I like this property, we have a similar concept here in Germany, but I am not sure the modelling you do in Thiruvallur Lok Sabha constituency (Q7785493) is the right way. Instead of cramming all individual elections into the constituency item, you should in my opinion create items for individual elections, where you then could use this as a regular property. Also, I would suggest to name it reserved seats or something similar. Reservation is a bit too generic for my tastes. --Srittau (talk) 21:21, 5 May 2016 (UTC)
      I would agree the modeling needs a second look. --Izno (talk) 16:27, 3 June 2016 (UTC)
      the whole model for elections/constituencies needs a deep thought. It looks so differently in different times and parts that it is very difficult to see the whole picture. There is also the problem that we do not have very detailed sources everywhere. -- Innocent bystander (talk) 19:26, 3 June 2016 (UTC)

    @Mdmahir:   Not done, no consensus. --Srittau (talk) 23:54, 11 June 2016 (UTC)

    @Mdmahir, Srittau: Do you mind if I open this again? There is maybe not consensus today, but I am hopeful that we will come to one, even if it will take time and need more talk! -- Innocent bystander (talk) 06:42, 12 June 2016 (UTC)
    @Innocent bystander: Personally, I would prefer a new suggestion, based on the discussion above, i.e. not as a qualifier and with proper domain etc. Maybe after a general discussion about how voting and election results are to be modelled. But if you want to reopen this proposal, please go ahead. --Srittau (talk) 12:34, 12 June 2016 (UTC)
    Ok, closed again!  – The preceding unsigned comment was added by Innocent bystander (talk • contribs).
    I think it could gain from the different formulation, maybe with more detailed samples from two or three countries. Obviously it's something we would want to include.
    --- Jura 12:54, 12 June 2016 (UTC)
    sorry to say, Are there any users here to knows Indian elections/politics? for example latest Uttar Pradesh election had 80+ reserved seats in its assembly elections. see en:Uttar_Pradesh_Legislative_Assembly_election,_2017. -- Mdmahir (talk) 08:15, 12 March 2017 (UTC)
    Any update on this? -- Mdmahir (talk) 14:48, 30 November 2017 (UTC)

    I stopped contributing to wikipedia/wikidata after I failed to get consensus because I got disappointed and don't want to waste time. This property is specifically for India elections. Why people confuse, I see people from other regions they able to get it done. -- Mdmahir (talk) 15:19, 7 April 2018 (UTC)

    Molendatabase verdwenen molens ID

    DescriptionIdentifier of former/disappeared mills in the Netherlands at the Molendatabase verdwenen molens.
    Data typeExternal identifier
    Template parameternl:Sjabloon:Infobox molen: "molendatabase-nl-verdwenen" - Template:Infobox windmill (Q13383928)
    Domainmill (Q44494)
    Allowed values[1-9]\d*
    Example't Soepuus (Q2236961)4650
    Akermolen (Q1825312)432
    Formatter URL$1
    Robot and gadget jobsCan be imported by "Harvest templates" gadget

    The Molendatabase verdwenen molens (Database of disappeared mills) is a sister database of the Nederlandse Molendatabase (Dutch Milldatabase) (Nederlandse Molendatabase ID (P2863)) and contains a lot of interesting and detailed information on mills in the Netherlands that disappeared or are not mills anymore. Used in nlwiki infobox, but also e.g. in enwiki. Lymantria (talk) 14:55, 29 May 2016 (UTC)

    •   Support -- T.seppelt (talk) 20:00, 29 May 2016 (UTC)
    • Do you mean "not used as mills anymore"? Or as in, they were demolished completely and something else built there? As for disappeared mills, can you point to such a mill in en.WP? --Izno (talk) 20:45, 6 June 2016 (UTC)
      • Looking at the examples, it includes at least repurposed, but still standing mills. --Srittau (talk) 21:45, 11 June 2016 (UTC)
    •   Support. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:55, 8 June 2016 (UTC)
    •   Question Could you please verify the examples? One leads to, while the other leads to --Srittau (talk) 21:45, 11 June 2016 (UTC)

    @Lymantria, T.seppelt, Pigsonthewing:   Done --Srittau (talk) 12:55, 12 June 2016 (UTC)

    data set

    Descriptionitem has an applicable data set
    Representsdata set (Q1172284)
    Data typeURL
    Domainin principle, anything
    Allowed valuesURLs to data repositories
    ExampleMakita 6408 (Q23775259) → sound pressure → 88 decibels relative to 20 micropascals --> dataset --> [3]
    Sourcedata sharing platforms
    Robot and gadget jobspart of import of NIOSH Power Tools Database (Q17144694)

    While my specific motivation involves linking to time series associated with power tools' sound pressure levels, it has a broad number of applications. We have plenty of Wikidata items on scientific papers, but it would be great if, when possible, we could also link to datasets associated with those papers. My specific example uses this property as a qualifier for a statement, but it could also be a property of a scientific article itself. Where multiple datasets can be attributed to one item, qualifiers such as P794 (P794) could be used. James Hare (NIOSH) (talk) 14:20, 23 May 2016 (UTC)


    Any reason I couldn't just use external data available at (P1325)? James Hare (NIOSH) (talk) 13:22, 6 June 2016 (UTC)

    • On the data item related to the paper? That seems appropriate to me. The property looks little-used too. --Izno (talk) 13:47, 6 June 2016 (UTC)
    •   Oppose external data available at (P1325) seems to be exactly this proposal, actually. --Srittau (talk) 20:33, 11 June 2016 (UTC)
    • @James Hare (NIOSH): seeing as you've got two property creators agreeing with your own-found alternative, will you withdraw this proposal? --Izno (talk) 20:59, 11 June 2016 (UTC)

    full name

    Descriptionfull name of a place, organisation, person etc.
    Data typeString
    Domaininstance of (P31) Wikidata property to indicate a name (Q19643892)
    Allowed valuesstring
    ExampleFC Bayern Munich (Q15789) → Fußball-Club Bayern, München e. V.
    Steffi Jones (Q290127) → Stephanie Ann Jones
    NEC (Q219203) → Nippon Electric Company
    IBM (Q37156) → International Business Machines Corporation
    Sourceexternal reference, Wikipedia list article
    See alsoopposite short name (P1813) / nickname (Q49614)

    Storing the real / full name of an object is useful. I considered using the description or as an alias, but this is nuts. Description should descripe the item. And to catch the right entry in the alias array is impossible.

    I found no property (like long name or real name); especially companies have often a long name registered, sport fans are using nicknames since centuries. So ... where to place this information?
    The multilingual problem (like jap.


    , Nippon Denki Kabushiki-gaisha) can be solved by adding a qualifier P2439 (P2439)

    Sure, we can say: Use the (real) long name as label and drop the familiar name in alias, but then there is a lot of update work to do. Plagiat (talk) 19:40, 12 June 2016 (UTC)


      Question Can't we use official name (P1448)? --Srittau (talk) 22:25, 12 June 2016 (UTC)

      Withdrawn   thank you for →official name (P1448) didn't find this property. --Plagiat (talk) 04:45, 13 June 2016 (UTC)

    syntax clarification

    DescriptionQualifier to provide extra details on the syntax of a statement value, in particular for format as a regular expression (P1793).
    Representssyntax (Q37437)
    Data typeMonolingual text
    DomainProperties with "format as regular expression" property, P1793.
    Allowed valuesAny descriptive text
    ExampleYandex Music artist ID (P1553) : format as a regular expression (P1793) → "[1-9]\d{0,6}"
    qualified with syntax clarification → "numeric string, 1 to 7 digits (English)"

    Proposed as a result of the deletion discussion of generic property "comment", comment (DEPRECATED) (P2315), this qualifier would be used to replace "comment" for the 240 or so occurrences where it is used as qualifier on format as a regular expression (P1793). It might eventually be expanded to include other cases where some cryptic syntax requires an explanation. LaddΩ chat ;) 21:50, 2 June 2016 (UTC)

    •   Support seems very useful. Thryduulf (talk: local | en.wp | en.wikt) 10:41, 3 June 2016 (UTC)
    •   Support it's clear and useful. GAllegre (talk) 13:40, 6 June 2016 (UTC)
    • Domain should be any properties with "format as regular expression" property. --Izno (talk) 12:07, 3 June 2016 (UTC)
      • Indeed. Corrected. -- LaddΩ chat ;) 15:37, 3 June 2016 (UTC)
        • I think this should probably already be "regex clarification" rather than "syntax clarification". The latter has some potential misuse that I'd like to avoid beforehand. --Izno (talk) 13:42, 6 June 2016 (UTC)
          • @Izno: IMHO, using this qualifier to clarify the syntax of any other cryptic statement would not be a misuse; I would prefer to allow its use beyond regexes, since it is still pretty specific. -- LaddΩ chat ;) 13:24, 12 June 2016 (UTC)
            @Laddo: But you just agreed to restrict the domain to properties with the "regex" property. :D --Izno (talk) 15:46, 12 June 2016 (UTC)
    • Pleading guilty. Either both name and domain restricted to regex, or neither. Let me sleep on it. -- LaddΩ chat ;) 16:47, 12 June 2016 (UTC)

    @Laddo, Thryduulf, GAllegre, Lymantria:   Done --Srittau (talk) 01:45, 19 June 2016 (UTC)

    month of the year

    Originally proposed at Wikidata:Property proposal/Generic


    This was originally suggested by ArthurPSmith in the proposal for day of week (P2894) with which it is complementary. It is intended as a generic property and so the uses above are illustrative rather than exhaustive. The allowed values could be constrained to just instances or subclasses of month (Q5151) (or a different single item) if the months of various calendar systems were all instances or subclasses of that item - a few are but there appears to have been no systematic structuring of this data (most just use calendar date (Q205892) which is too broad for a meaningful constraint to this proposed property) Thryduulf (talk: local | en.wp | en.wikt) 13:49, 8 June 2016 (UTC)



       Done: icon (P2910) (Talk and documentation)
    Descriptionpictogram suitable to represent the item, subproperty of "image", parent property of "logo"
    Representscomputer icon (Q138754)
    Data typeCommons media file
    Examplewheelchair accessible (Q24192067)File:Wheelchair-green3.png
    • This proposal is misplaced. It is not about a Commons-related property, just a property that uses a commons link. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:46, 29 May 2016 (UTC)
      Agreed. @Pigsonthewing: Do you have a better location in mind? --Izno (talk) 21:12, 6 June 2016 (UTC)
      I linked it on the general page as well. --Srittau (talk) 00:17, 12 June 2016 (UTC)
    •   Support Sounds good. NMaia (talk) 12:59, 14 June 2016 (UTC)

    @Jura1, Srittau, NMaia:   Done now icon (P2910) --Lymantria (talk) 13:07, 15 June 2016 (UTC)

    plural form of label

    Descriptionplural form of the label of this item
    Representsplural (Q146786)
    Data typeMonolingual text
    Domainany item, units of measure in particular
    Allowed valuesonly plural form of the item label
    Examplemetre (Q11573) → meters (en) → metriä (fi)  ; Mus (Q39275) → mice (en)
    See alsofemale form of label (P2521)

    Sounds necessary, at least for units of measure. See this discussion. LaddΩ chat ;) 10:47, 14 June 2016 (UTC)

    • A property similar to this has been rejected multiple times before. What's different compared to those discussions? --Izno (talk) 11:35, 14 June 2016 (UTC)
    •   Oppose, per numerous earlier discussions. --Yair rand (talk) 18:24, 14 June 2016 (UTC)
    •   Oppose as currently proposed, because it does not address any of the issues raised with the previous proposals. - Nikki (talk) 19:33, 14 June 2016 (UTC)

      Withdrawn, my mistake, I thought I was the first to have that brilliant idea. -- LaddΩ chat ;) 19:36, 14 June 2016 (UTC)

    • @Laddo:: you might want to limit it to units.
      --- Jura 12:50, 19 June 2016 (UTC)
    • Actually, I was about to propose this property myself when I found this here. Transforming a known singular word into plural may be close to trivial in quite some languages, but it is not in my own language, German (at least there's considerably more special cases in German compared to English). I have read some of the related discussions and agree that this is mostly a linguistic problem and may not be a subject of Wikidata. However, does anyone know any machine-accessible public database that provides singular to plural mappings? Machine-accessible Wiktionary doesn't seem to have arrived by now, right? (BTW: Units of measurement were about the least reason why I was goin to propose this property.) Eri523 (talk) 22:25, 3 January 2018 (UTC)
    • Agree with Eri523. It is the same with Hebrew. And the rejected multiple times before is list of discusions that the word plural mention and not discusion about plural form of label. for example HaOlam HaZeh (Q291439) magazin has a frequency of publication (publication interval (P2896)) of 7 days. Geagea (talk) 01:46, 11 June 2019 (UTC)

    time gap

       Done: time gap (P2911) (Talk and documentation)
    Descriptiontime gap
    Representstime (Q11471)
    Data typeNumber (not available yet)
    Template parameterwill be in use in the function generalclassification (then stageclassification...) of Module:Cycling race, used in more than twenty Wikipedias
    ExampleParis-Roubaix espoirs 2016 (Q20017698)... Jenthe Biermans (Q18571962), ranking (P1352) : 2, time gap : 49 s

    Hi everybody. Two months ago, I have had the property race time (P2781). With Molarus we continue the work to centralise the datas on Wikidata for cycling and reuse it in possibly 20 Wikipedia thanks to the program Module:Cycling race. teamroster is now finished, perfectly operational and tried like the three first functions, we now start this day to develop generalclassification, a function that will permit to display in possibly twenty Wikipedias the table of the general classification of a cycling race.

    To do this, I need two other properties. The first, object of this proposal, is a time property, called "time gap" or somethink like that in English and "écart de temps" in French. With race time (P2781), we give the time of the winner, with the time gap, we say the difference between the winners and other cyclists. Sunday, I was at the fr:Paris-Roubaix espoirs 2016. As you can see, the second arrive 49 seconds after the first, the third arrives at the same time than the second. Note that I will appreciate to have the time expressed if necessary in hours minutes and seconds as we do for a date. It seems it is not again possible, I will ask to the development team. (and sorry for my English) Jérémy-Günther-Heinz Jähnick (talk) 12:40, 2 June 2016 (UTC)

    • This is derived information from the times of the participants in the races. In general,   Oppose, unless there's some reason why you cannot derive it from those times. --Izno (talk) 15:21, 2 June 2016 (UTC)
    • I would also prefer using race time (P2781) for the winner, runner-up and so on and then calculate the difference. I can see this not working, if the runner-up etc. are not modelled, though. --Srittau (talk) 21:02, 11 June 2016 (UTC)
    •   Support I understand the objections, but it is a fact that all rankings (not only for cycling) are presented with a time gap on internet and...everywhere actually. Moreover, for long races, like Tour de France, the total time is really huge (like 100 h), so it is quite long to type the general classification time of each rider without a time gap. Eventually in cycling there are a lots of massive sprint. In this case all times are the same, it is once again quicker and safer, to type "0" than making a copy/paste of the winner time. Psemdel (talk) 20:34, 12 June 2016 (UTC)
      • I prefer a good data model over ease of entering data. --Srittau (talk) 22:37, 12 June 2016 (UTC)
        • To give more informations, if you see this example, we have only race time (P2781) for the winner of the race. For all other cyclists, we have only the time gap because it is the only data that is widely used in cycling. This way to do is the good data model. But your idea is in reality not bad, because during the team time trial as this example you can see that we use the race time, the time gap and the average speed. But this particular classification will be another function. Note also that we can't ask to all user to calculate in seconds the time for each cyclist. I prefer datas entered as we do on cycling (and I work regularly on races) as not data because it is too long/too difficult to enter. Note that these datas like all other datas on Wikidata about cycling are entered here just because users are able to reuse them in their Wikis. Without this, Wikidata would have only datas entered by bots for cycling. Users work here because they have a gain to do it, not for the love of free licences or similar ideology. Jérémy-Günther-Heinz Jähnick (talk) 09:35, 13 June 2016 (UTC)
    •   Support Looks really nice this property to general classification.

    date depicted

       Done: date depicted (P2913) (Talk and documentation)
    DescriptionThe date that is represented in a painting, map or even in a literary work.
    Data typePoint in time
    DomainRepresentations of images and literary works: depiction (Q1166770)
    ExampleAddresses and travel map of Helsinki from 1876 (Q17610148) → 1876

    This property is needed to distinguish different dates that are used to describe a map. publication date (P577) tells the time of publication, inception (P571) the time of creation, but neither of them can describe what time is depicted with the map. Susannaanas (talk) 09:17, 7 June 2016 (UTC)

    •   Support. We do have depicts (P180) but that is better I think used for the subject than the date, e.g.
      ⟨ Thryduulf's map ⟩ instance of (P31)   ⟨ map (Q4006)      ⟩
      depicts (P180)   ⟨ Roman Empire (Q2277)      ⟩
      publication date (P577)   ⟨ 2 June 2016 ⟩
      depicts date search ⟨ 23756 ⟩
      . We also have set in period (P2408) but that's intended for fictional works and can be much vaguer than a specific date. Thryduulf (talk: local | en.wp | en.wikt) 10:06, 7 June 2016 (UTC)
    •   Support Seems equivalent to dct:temporal (temporal coverage), so add this as one of the names, and as "equivalent property". This is better than "subject" because it doesn't require to have all possible intervals as items, and because it allows temporal computations (eg give me all Creative Works that cover a given time interval. --Vladimir Alexiev (talk) 19:21, 7 June 2016 (UTC)
    •   Support Valuable for posthumous portraits that have an inscribed date which is not the date the painting was created. - PKM (talk) 00:57, 10 June 2016 (UTC)
    •   Support Also valuable for Commons. --Srittau (talk) 21:10, 11 June 2016 (UTC)

    @Susannaanas, Thryduulf, Vladimir Alexiev, PKM, Srittau:  Done now date depicted (P2913). -- Lymantria (talk) 08:33, 16 June 2016 (UTC)

    month of the year

    Originally proposed at Wikidata:Property proposal/Generic


    This was originally suggested by ArthurPSmith in the proposal for day of week (P2894) with which it is complementary. It is intended as a generic property and so the uses above are illustrative rather than exhaustive. The allowed values could be constrained to just instances or subclasses of month (Q5151) (or a different single item) if the months of various calendar systems were all instances or subclasses of that item - a few are but there appears to have been no systematic structuring of this data (most just use calendar date (Q205892) which is too broad for a meaningful constraint to this proposed property) Thryduulf (talk: local | en.wp | en.wikt) 13:49, 8 June 2016 (UTC)


    distinctive jersey

    Descriptiondistinctive jersey
    Representstime (Q11471)
    Data typeItem
    Template parameterwill be in use in the function generalclassification and stageclassification... of Module:Cycling race, used in more than twenty Wikipedias
    Example2016 Tour of Oman (Q22752188)... Vincenzo Nibali (Q312322), ranking (P1352) : 1, distinctive jersey : red jersey, general classification (Q24257872)

    Hi everybody. As you can see in action on fr:Modèle:Cycling race/generalclassification/Documentation and on fr:Modèle:Cycling race/stageclassification, two functions have been developped by Molarus to permit to display on different Wikipedias the classifications of races. We must be able to make complete classifications, so for this I need a property for adding one or more jerseys. I start to list them on Wikidata:WikiProject Cycling/Kit to translate/Jerseys. Sometimes a cyclist can have three jerseys, so it wasnot possible to use image (P18) and media legend (P2096), more than we need a translated description for the jersey. Jérémy-Günther-Heinz Jähnick (talk) 11:26, 12 June 2016 (UTC)

    Can't we (technically) deduce those information from properties like overall leader at the end of the stage (Q20882763) from present and previous stages? Psemdel (talk) 20:30, 12 June 2016 (UTC)
    J'avais pensé à ça, mais in fine par exemple ça ne pouvait pas donner la couleur du maillot, sachant que cette donnée pouvait aussi varier au fil des éditions. Sous-jacent il y avait aussi le problème de l'ordre de ces maillots, sachant que différents contributeurs remplissent les vainqueurs, et qu'il n'y a pas d'ordre précis. La solution la plus simple consistait à utiliser un système proche de fr:Modèle:Maillot. Les premiers essais, avec une propriété de subsititution sont plutôt bon puisque Molarus a réussi à définir une couleur de fond par la même occasion. Jérémy-Günther-Heinz Jähnick (talk) 09:43, 13 June 2016 (UTC)
    •   Support I've been waiting for this property to all the jerseys. It will be very helpful to stage and general classification.

    phone number type

       Not done
    Descriptionqualifier for phone numbers: fax/toll free, etc.
    Data typeItem
    Template parametertollfree/fax in voy:Template:Listing as documented at voy:Wikivoyage:Listings (possibly others of Template:Listing (Q14330485) )
    Allowed valuesitems for types of phone numbers
    ⟨ Hotel California ⟩ phone number (P1329)   ⟨ +1-213-555-4823 ⟩
    794 search ⟨ fax number ⟩
    Robot and gadget jobsimport from Wikivoyage
    •   Comment using P794 (P794) instead would probably do.
      --- Jura 06:44, 17 May 2016 (UTC)
    •   Comment Please provide an example using real items and values. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:33, 17 May 2016 (UTC)
    •   Oppose use
      ⟨ Canary Riverside Plaza Hotel (Q5475511)      ⟩ phone number (P1329)   ⟨  +44 20 7510 1999 ⟩
      has quality (P1552)   ⟨ standard rate phone number ⟩
      (I can't find an example of a hotel with both an item and a freephone number in a 2 minute search). Thryduulf (talk: local | en.wp | en.wikt) 14:38, 17 May 2016 (UTC)
      • I don't think we should add qualifiers for "standard rate phone numbers" ;). To avoid that someone actually tries to dial it, use a US one with 555.
        --- Jura 14:45, 17 May 2016 (UTC)
      • It's common to have multiple numbers. voy:Manhattan/Theater District has Hotel Pennsylvania (Q1631006), 401 7th Ave (between 32nd/33rd), ☎ +1 212-736-5000, toll-free: +1-800-223-8585. Large hotel with really famous phone number PEnnsylvania 6-5000 (Q7119041) that has its own catchy tune Pennsylvania 6-5000 (Q7163635) but for some reason the number is missing from the Wikidata record for the hotel. (It's on the 78rpm record, though.) K7L (talk) 15:01, 17 May 2016 (UTC)
    •   Weak support P794 (P794) would do, but I prefer a specialized property. --Srittau (talk) 15:21, 17 May 2016 (UTC)
    •   Comment There might be an obvious use for four numbers or so: phone number (P1329), freephone, mobile, fax. That'd be three new properties, which would be manageable. K7L (talk) 15:37, 17 May 2016 (UTC)
      • Also telecommunications device for the deaf (Q354147). Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 20:29, 17 May 2016 (UTC)
      • I would definitely say fax number should be a separate property, regardless. Faxes and voice calls are different things and I can't see any benefit to combining them into a single property. I generally support subproperties over qualifiers if the number of variations is quite low (which seems to be the case here) because qualifiers are often quite broad in meaning and get used very inconsistently (there's already three different suggestions here... and I'd like to add fee (P2555) novalue as another option for free numbers :P) and properties are generally easier to add and easier to query for than qualifiers. - Nikki (talk) 11:04, 19 May 2016 (UTC)
        • International and industry standards such as vCard treat fax numbers as a subtype of telephone numbers; we should follow their lead. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:16, 20 May 2016 (UTC)
          • That's what we have subproperty of (P1647) for. - Nikki (talk) 13:43, 20 May 2016 (UTC)
            • My point is that they treat "fax" equally with "mobile" and "freephone", and not, as you seem to suggest, differently to them. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 16:22, 20 May 2016 (UTC)
          • The current system is that Template:Listing (Q14330485) has separate fields for fax or freephone/tollfree numbers and the template generates the vCard (actually an hCard) in whatever format - without forcing us to change fax=+xxxx to be phone=+xxxx/and, by the way, this is a telecopier. K7L (talk) 15:52, 20 May 2016 (UTC)
      • Once the datatype for phone numbers is sorted out (see Wikidata:Properties for deletion), I think we should do a separate property for the fax numbers. For others, qualifiers should be sufficient.
        --- Jura 07:59, 24 May 2016 (UTC)
      • Change my mind, see proposal below.
        --- Jura 15:57, 24 May 2016 (UTC)
    •   Oppose Use a generic qualifier to represent the type of phone (besides fax; see below): either P794 (P794) or of (P642). --Izno (talk) 14:51, 10 June 2016 (UTC)
      Not done Unsufficient support. Lymantria (talk) 10:16, 12 June 2016 (UTC)


       Done: fax number (P2900) (Talk and documentation)
    Descriptionnumber of fax line
    Data typeString
    Template parameterfax in voy:Template:Listing as documented at voy:Wikivoyage:Listings (possibly others of Template:Listing (Q14330485) )
    Allowed valuesphone numbers with country code
    ⟨ Hotel California ⟩ fax search ⟨ +1-213-555-4823 ⟩
    Robot and gadget jobsimport from Wikivoyage
    •   Comment somewhat legacy, but included in listings template at Wikivoyage.
      --- Jura 15:57, 24 May 2016 (UTC)
    •   Oppose per my comments in the preceding proposal. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 20:01, 25 May 2016 (UTC)
    •   Comment given the support by Nikki and K7L above (#phone number type), I'm marking this as "ready".
      --- Jura 08:21, 4 June 2016 (UTC)
    • I am going back and forth between having this as a separate property and having this as the more generic "phone number". My head says "phone number" and my gut says "fax". This is mostly due to the difference in output: data using a faxline is going to end up being printed rather than heard (except in that lovely set of tones signifying a call :D), which may make it more desirable to have it elsewhere than a 'generic' phone property. I will tentatively   Support this property. --Izno (talk) 14:43, 10 June 2016 (UTC)
      • Yeah, which went me to go from a qualifier solution to a separate property. Given the extensive support, I think this is ready for creation.
        --- Jura 08:07, 11 June 2016 (UTC)
    •   Support, phone = talk to someone, fax = machines talking to each other. (Also, specialized properties top qualifiers, in my opinion.) --Srittau (talk) 00:14, 12 June 2016 (UTC)

    @Jura1, Izno, Srittau:   Done, now fax number (P2900) as subproperty of phone number (P1329) --Lymantria (talk) 11:38, 12 June 2016 (UTC)


       Done: icon (P2910) (Talk and documentation)
    Descriptionpictogram suitable to represent the item, subproperty of "image", parent property of "logo"
    Representscomputer icon (Q138754)
    Data typeCommons media file
    Examplewheelchair accessible (Q24192067)File:Wheelchair-green3.png
    • This proposal is misplaced. It is not about a Commons-related property, just a property that uses a commons link. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:46, 29 May 2016 (UTC)
      Agreed. @Pigsonthewing: Do you have a better location in mind? --Izno (talk) 21:12, 6 June 2016 (UTC)
      I linked it on the general page as well. --Srittau (talk) 00:17, 12 June 2016 (UTC)
    •   Support Sounds good. NMaia (talk) 12:59, 14 June 2016 (UTC)

    @Jura1, Srittau, NMaia:   Done now icon (P2910) --Lymantria (talk) 13:07, 15 June 2016 (UTC)

    focal height

       Done: focal height (P2923) (Talk and documentation)
    Descriptionheight of the lamp of a lighthouse from sea level
    Data typeNumber (not available yet)
    Template parameter"focalheight" in en:Template:Infobox_lighthouse
    Domainlighthouse (Q39715)
    ExampleBengtskär lighthouse (Q3737012) → 51 m

    Properties for lighthouses could be more straightforward, since lighthouses are widely listed. Currently declaring this requires qualifiers. See also: WikiProject Lighthouses Susannaanas (talk) 20:44, 9 June 2016 (UTC)


    SecondHandSongs song ID

    DescriptionIdentifier for a song, in the SecondHandSongs database
    Data typeExternal identifier
    Domainmusical works
    ExampleHere Comes the Sun (Q210401)383
    Formatter URL$1

    Useful source of data on cover versions, and sampling, of songs. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 19:33, 31 May 2016 (UTC)

    Note also: Wikidata:Property proposal/SecondHandSongs artist ID. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits
      Support Looks useful! -- Innocent bystander (talk) 12:47, 1 June 2016 (UTC)

    @Pigsonthewing, Innocent bystander:   Done -- Lymantria (talk) 05:39, 14 June 2016 (UTC)

    @Lymantria: Thank you. Please note Wikidata:Property proposal/SecondHandSongs artist ID, which is ready for creation, too. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 15:10, 14 June 2016 (UTC)

    Other versions

       Not done
    Descriptionother versions or editions of this work
    Data typeItem
    ExampleOs Lusíadas (Q781898)The Lusiads (Q19093144), The Lusiads (Q10342013)

    Useful mostly for Wikiquote. Often a single work has multiple editions or translations, and currently there is no way to bind them together. NMaia (talk) 11:56, 5 June 2016 (UTC)


      Comment I have a vague recollection of a discussion along similar lines to this that proposed having an item about the work and then separate items for different editions and translations, all linking to the item of the work with an "edition of" or "translation of" property. Would that not work better than peer to peer linking? Thryduulf (talk: local | en.wp | en.wikt) 13:47, 5 June 2016 (UTC)

    Yes, perhaps that would be even better in terms of structure :) NMaia (talk) 14:49, 5 June 2016 (UTC)
    edition or translation of (P629) and has edition or translation (P747)Máté (talk) 17:09, 5 June 2016 (UTC)
    •   Oppose, use the properties indicated by Mate. --Izno (talk) 14:39, 6 June 2016 (UTC)
    •   Oppose per my previous comment and the properties Máté has noted. Thryduulf (talk: local | en.wp | en.wikt) 20:54, 6 June 2016 (UTC)

    Not done. Duplicates edition or translation of (P629) etc. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 14:35, 15 June 2016 (UTC)

    date depicted

       Done: date depicted (P2913) (Talk and documentation)
    DescriptionThe date that is represented in a painting, map or even in a literary work.
    Data typePoint in time
    DomainRepresentations of images and literary works: depiction (Q1166770)
    ExampleAddresses and travel map of Helsinki from 1876 (Q17610148) → 1876

    This property is needed to distinguish different dates that are used to describe a map. publication date (P577) tells the time of publication, inception (P571) the time of creation, but neither of them can describe what time is depicted with the map. Susannaanas (talk) 09:17, 7 June 2016 (UTC)

    •   Support. We do have depicts (P180) but that is better I think used for the subject than the date, e.g.
      ⟨ Thryduulf's map ⟩ instance of (P31)   ⟨ map (Q4006)      ⟩
      depicts (P180)   ⟨ Roman Empire (Q2277)      ⟩
      publication date (P577)   ⟨ 2 June 2016 ⟩
      depicts date search ⟨ 23756 ⟩
      . We also have set in period (P2408) but that's intended for fictional works and can be much vaguer than a specific date. Thryduulf (talk: local | en.wp | en.wikt) 10:06, 7 June 2016 (UTC)
    •   Support Seems equivalent to dct:temporal (temporal coverage), so add this as one of the names, and as "equivalent property". This is better than "subject" because it doesn't require to have all possible intervals as items, and because it allows temporal computations (eg give me all Creative Works that cover a given time interval. --Vladimir Alexiev (talk) 19:21, 7 June 2016 (UTC)
    •   Support Valuable for posthumous portraits that have an inscribed date which is not the date the painting was created. - PKM (talk) 00:57, 10 June 2016 (UTC)
    •   Support Also valuable for Commons. --Srittau (talk) 21:10, 11 June 2016 (UTC)

    @Susannaanas, Thryduulf, Vladimir Alexiev, PKM, Srittau:  Done now date depicted (P2913). -- Lymantria (talk) 08:33, 16 June 2016 (UTC)

    memory capacity

    Descriptionmemory storage capacity of some data storage device
    Data typeQuantity
    Domainrecording medium (Q193395)
    Allowed valuesbyte (Q8799) and derivatives
    ⟨ Core2 Quad Q6600 (Q15218754)      ⟩ has part (P527)   ⟨ L2 cache (Q12635161)      ⟩
    <memory capacity> search ⟨ 8 megabyte (Q79735)  ⟩
    ⟨ Commodore 64 (Q99775)      ⟩ has part (P527)   ⟨ primary memory (Q11140433)      ⟩
    <memory capacity> search ⟨ 64 kilobyte (Q79726)  ⟩

    Memory storage capacity is a key figure of several computer related items. Lymantria (talk) 20:08, 6 June 2016 (UTC)


    Definetely there is a conspiracy against me in this project... I'm still thinking what to do about it.--MisterSanderson (talk) 05:09, 23 June 2016 (UTC)

    • Which memory this property is exactly referring to? Using this query:
    SELECT ?item ?itemLabel ?value WHERE {
      ?item wdt:P2928 ?value
      SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en". }
    limit 100

    Try it!

    I found, that some use this property for RAM, other for the persistent storage. Sirexo (talk) 09:20, 20 June 2017 (UTC)


       Done: InterPro ID (P2926) (Talk and documentation)
    DescriptionInterPro unique identifier
    RepresentsInterPro (Q3047275)
    Data typeExternal identifier
    DomainMolecular Biology, Medicine
    Allowed valuesIPRxxxxxx 'IPR' as the prefix, followed by 6 digits
    Example5-Hydroxytryptamine 2A receptor (Q24575295)IPR000455
    Formatter URL$1
    See also

    InterPro is a resource of protein families, domains and functional sites that provides functional analysis of protein sequences by classifying them into families and predicting the presence of domains and important sites. Having InterPro in WikiData would allow linking proteins together by family/function across and within species. Gstupp (talk) 21:08, 12 June 2016 (UTC)


      Support This is a key database, lets get this done. --I9606 (talk) 21:14, 13 June 2016 (UTC)

      Support Very important resource. --Putmantime (talk) 01:46, 14 June 2016 (UTC)

    Doc James
    Daniel Mietchen
    Andrew Su
    Projekt ANA
    Pavel Dušek
    Was a bee
    Chris Mungall
    Dr. Abhijeet Safai
    Sami Mlouhi
    Netha Hussain
    Abhijeet Safai
    Shani Evenstein
    ZI Jony
      Notified participants of WikiProject Medicine Andrew Su
    Marc Robinson-Rechavi
    Pierre Lindenbaum
    Michael Kuhn
    Dan Bolser
    Timo Willemsen
    Salvatore Loguercio
    Daniel Mietchen
    Ben Moore
    Alex Bateman
    Vojtěch Dostál
    Andra Waagmeester
    Elvira Mitraka
    David Bikard
    Dan Lawson
    Francesco Sirocco
    Konrad U. Förstner (talk)
    Chris Mungall (talk)
    Kristina Hettne
    Karima Rafes
    Finn Årup Nielsen
    Jasper Koehorst
    Till Sauerwein
    Amos Bairoch
    Was a bee
    Muhammad Elhossary
    Damian Szklarczyk
    Yusra Haider
    Hannes Röst
    Kritika Dusad
    T.Shafee(evo&evo) (talk)
      Notified participants of WikiProject Molecular biology

      Support Very useful database to incorporate functional proteomics into wikidata. Julialturner (talk) 21:23, 13 June 2016 (UTC)

      Support Important resource for functional annotationn of protein items. Sebotic (talk) 04:31, 14 June 2016 (UTC)

    @Gstupp, I9606, Putmantime, Julialturner, Sebotic:   Done Now InterPro ID (P2926) --Lymantria (talk) 05:46, 21 June 2016 (UTC)

    Timezone offset

    Descriptiondifference between Coordinated Universal Time (Q1536) and the timezone
    RepresentsUTC offset (Q187110)
    Data typeQuantity
    Domain instance of (P31) time zone named for a UTC offset (Q17272482)
    Allowed valuesunit hour (Q25235)
    ExampleUTC±00:00 (Q6574) → 0, UTC+04:00 (Q6779) → +4

    We need a way to determine the value of a timezone. For example in voy:en:Template:Quickbar (see voy:United States of America as an example) we don't need to display every single timezone but just a MIN and a MAX value. There the offset of a timezone is required.

    See also voy:en:Wikivoyage:Travellers' pub#Switching_.7B.7BQuickbar.7D.7D_to_Wikidata. @Matroc, LtPowers, WOSLinker: -- T.seppelt (talk) 05:22, 6 June 2016 (UTC)


    There are a limited set of time zones. I therefore suggest that items are used. I don't know what is best for places spanning over several zones, but to be able to add several values (one for each zone) might be an option. --Averater (talk) 08:02, 6 June 2016 (UTC)

    domain of saint or deity

    Will be very useful for the items of deities in different belief systems. Lingveno (talk) 06:12, 9 June 2016 (UTC)


    @Lingveno, Thryduulf, Isno, Pigsonthewing:   Done Now domain of saint or deity (P2925) --Lymantria (talk) 05:37, 21 June 2016 (UTC)


       Not done
    Descriptionattributes/symbols of this deity
    Data typeItem
    ExamplePoseidon (Q41127)trident (Q271628); Zeus (Q34201) -> lightning (Q33741); Osiris (Q46491) -> crook and flail (Q16931086)

    Will be very useful for the items of deities in different belief systems. Lingveno (talk) 06:19, 9 June 2016 (UTC)


      Not done Preferably use has quality (P1552). Lymantria (talk) 17:40, 1 July 2016 (UTC)

    Pleiades place type

    RepresentsPleiades (Q24423804)
    Data typeExternal identifier
    Template parameternone
    Domainitem that represents place type
    Allowed values[a-z-]+
    Source, (authoritative)
    Formatter URL$1

    A strong effort by the Pleiades people to align place types across vocabs is at I recommended they carry it out on Wikidata. We already have GeoNames feature code (P2452), OpenStreetMap tag or key (P1282), Art & Architecture Thesaurus ID (P1014); and they'd also need a "Pleiades place type" to do it. Vladimir Alexiev (talk) 18:04, 7 June 2016 (UTC)

    •   Support as external-id and as named "Pleiades place type ID". --Izno (talk) 20:08, 7 June 2016 (UTC)
      • Thanks @Izno:! But I think since "river" is a word, "type" is better than "type ID". Also their vocab is called "place-types" not "place-type-IDs"
      • Sorry for the confusion re asking to delete this page, the magic of how the parent page is composed of its headings is new to me. Vladimir Alexiev (talk) 20:15, 7 June 2016 (UTC)

        No worries regarding the magic. It will not take long to get used to and will probably help make it easier to focus on one proposal at a time. (And will show up nicer in recent changes and watchlist.)

        The reason I think it should be "place type ID" is that this will be used to link their vocabulary with ours in some way, not to populate our own ontology, and the name, without the phrase "ID", makes it look like our type is following theirs or theirs ours or something. And if it is the latter, then I'm not sure I support the property... --Izno (talk) 20:17, 7 June 2016 (UTC)

        @Izno: I am hopelessly confusedː these edits do not appear onː I obviously messed something up.
        Yes, the purpose is to sync their own types. But that semantic is IMHO carried by the datatype "external-id", not by the name. Vladimir Alexiev (talk) 20:21, 7 June 2016 (UTC)

        Correct, they will only appear on this page.

        I don't think that's sufficient. I would rather it be clearly obvious that the claim is being used as an ID by calling it one. --Izno (talk) 22:41, 7 June 2016 (UTC)

    @Vladimir Alexiev, Izno:   Done Now Pleiades category identifier (P2938). I included "identifier" in the label, as is usual. Lymantria (talk) 06:50, 28 June 2016 (UTC)

    Pleiades place ID

    DescriptionPlace identifier in the Pleiades gazetteer
    RepresentsPleiades (Q24423804)
    Data typeExternal identifier
    Template parameternone
    Domainplace or archaeological site
    Allowed values[0-9]+
    ExampleFilicudi (Q743908) → 462425
    Formatter URL$1

    An excellent and famous ancient place gazetteer. Describes 35k places. Includes RDF export, alignments to Geonames and TGN. I can't believe we don't have it yet. Vladimir Alexiev (talk) 18:13, 7 June 2016 (UTC)


    MSBI person ID

       Done: MSBI person ID (P2914) (Talk and documentation)
    Descriptionidentifier of a person, in the Mapping the Practice and Profession of Sculpture in Britain and Ireland database
    Data typeExternal identifier
    ExampleHenry Moore (Q151097) -> msib2_1214407085
    Formatter URL$1
    Robot and gadget jobsThe MSBI persons are a mix'n'match catalog.

    The Mapping the Practice and Profession of Sculpture in Britain and Ireland site, for short the Mapping Sculpture database, is a scholarly project on sculpture 1851 to 1951.[4]. There are over 6K person identifiers.

    "The University of Glasgow, in partnership with the Victoria & Albert Museum and Henry Moore Institute, launched the project in September 2007."[5] This is a major resource for British art. Charles Matthews (talk) 12:53, 9 June 2016 (UTC)

    •   Support But I believe the IDs should be prefixed, e.g. msib2_1214407085 rather than 2_1214407085. --Magnus Manske (talk) 12:58, 9 June 2016 (UTC)
    •   Support but I'd prefer the label as "MSBI person ID" to make it clear it's an ID for a person not a sculpture. Thryduulf (talk: local | en.wp | en.wikt) 15:41, 9 June 2016 (UTC)
      • Fair point - was actually my intention to have it that way. Charles Matthews (talk) 17:32, 9 June 2016 (UTC)
    •   Support A very useful database, which I have often used as a source on Wikipedia. Fixes suggested above applied, and description made more standard. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 20:37, 9 June 2016 (UTC)
    •   Support - looks good. Andrew Gray (talk) 11:37, 13 June 2016 (UTC)

    @Charles Matthews, Magnus Manske, Thryduulf, Pigsonthewing, Andrew Gray:   Done Now MSBI person ID (P2914), can be matched to Mix'n'match. Lymantria (talk) 05:40, 17 June 2016 (UTC)

    ECARTICO person ID

    Descriptionidentifier for a person, in the ECARTICO biographical database of the Dutch and Flemish Golden Ages
    Data typeExternal identifier
    ExampleErick van den Weerelt (Q24452339)8194
    Formatter URL$1
    Robot and gadget jobsECARTICO person ID are in a mix'n'match catalog

    ECARTICO is a comprehensive collection of structured biographical data concerning painters, engravers, printers, book sellers, gold- and silversmiths and others involved in the ‘cultural industries’ of the Dutch and Flemish Golden Ages.

    This is very valuable and useful collection of structured biographical data, supported by University of Amsterdam and free licensed under CC-BY-SA. Currently the database contains 24 912 peoples — "Painters: 7 671, Engravers: 1 292, Booksellers, printers and publishers: 865, Gold- and silversmiths: 1 953, Sculptors: 209", — and this is "gold pages" (in my opinion). --Kaganer (talk) 17:40, 9 June 2016 (UTC)


    @Kaganer, Pigsonthewing, Magnus Manske:   Done Now ECARTICO person ID (P2915), it can be matched with mix'n'match. Lymantria (talk) 20:22, 18 June 2016 (UTC)

    Lymantria, very thanks! --Kaganer (talk) 21:16, 19 June 2016 (UTC)

    COAM database ID

    Descriptionidentifier of a relevant building or structure of the city of Madrid in the Official Architects' Association of Madrid (COAM) database
    Data typeExternal identifier
    Template parameterQ24481357
    Domainarchitectural structure (Q811979)
    Allowed values[A-Z]{1}\d{1}.\d{3}[A-Z]?
    ExampleTorre Picasso (Q1364394)F2.398D
    Almudena Cathedral (Q849711)F1.157
    Torres de Colón (Q2919969)L2.685
    SourceOfficial Architects' Association of Madrid (Q5777096)
    Formatter URLhttp://$1
    Robot and gadget jobsCan be imported by "Harvest templates" gadget

    Aside from its value as external identifier, this useful and structured database (around 3500 items) usually offers, address, images, pics, architect(s), bibliographical references and a short description about each item. It may as well prove to be useful in coordinating local Commons contests. A template for this identifier already exists in es:wiki (see: es:Plantilla:COAM), with 227 items (see: es:Categoría:Artículos con plantilla COAM), already tagged and prepared for migration to wikidata.--Asqueladd (talk) 18:27, 9 June 2016 (UTC)

    •   Support. The Official Architects' Association of Madrid has a deep background in heritage protection and in the creation of building and structure catalogues in Madrid. As Asqueladd mentions, the records in their catalogue shows valuable and comprehensive information about the items and are, in themselves, reliable sources. Therefore, I cannot but support and encourage the adoption of this property. --Discasto (talk) 21:42, 9 June 2016 (UTC)
    •   Support I strongly support this property. --Rodelar (talk) 23:44, 9 June 2016 (UTC)
    •   Support Very interesting and useful initiative. --ElBute (talk) 07:20, 10 June 2016 (UTC)
    •   Support. This seems like a useful resource to link to. Thryduulf (talk: local | en.wp | en.wikt) 11:18, 10 June 2016 (UTC)
    •   Support. English description standardised; please do the same for Spanish. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 15:46, 12 June 2016 (UTC)

    @Asqueladd, Discasto, Rodelar, ElBute, Thryduulf, Pigsonthewing:   Done Now COAM structure ID (P2917) --Lymantria (talk) 07:00, 19 June 2016 (UTC)

    Great Russian Encyclopedia Online ID

    DescriptionID for an entry on the official website of the Great Russian Encyclopedia
    RepresentsGreat Russian Encyclopedia (Q1768199)
    Data typeExternal identifier
    Domainthousands of items related to different topics (there are some more than 12 thousand live articles to date, and, according to the publisher, there will be available up to 45 thousand by the end of this year)
    Allowed valuesnumber, probably not more than 7-8 digits
    ExampleJohn Lennon (Q1203)2139826
    SourceGreat Russian Encyclopedia website
    Formatter URL$1

    The Great Russian Encyclopedia is one of the most trusted Russian-language reference sources. The official website was opened to the public just two months ago, but the whole project is funded by the Russian government, so it's not expected to close in the near future. INS Pirat (t | c) 06:02, 4 June 2016 (UTC)

    •   Support NMaia (talk) 15:16, 10 June 2016 (UTC)

    @INS Pirat, NMaia:   Done Now Great Russian Encyclopedia Online ID (P2924) -- Lymantria (talk) 07:39, 20 June 2016 (UTC)

    CageMatch tag team ID

    Descriptionidentifier for a professional wrestling tag team at
    Data typeExternal identifier
    Domaintag team (Q1066670)
    Allowed valuesnumber
    ExampleThe Vaudevillains (Q20900198)5383
    Source (english) or (German)
    Formatter URL$1
    Motivation is a database of professional wrestling (Q131359). We already have CageMatch worker ID (P2728), and I think it would be useful to use the website to track tag teams too. Silverfish (talk) 23:45, 10 June 2016 (UTC)


    @Silverfish:   Done Now CageMatch tag team ID (P2939). --Lymantria (talk) 06:34, 30 June 2016 (UTC)

    INSPIRE-HEP author ID

    Data typeExternal identifier
    ExampleFabiola Gianotti (Q983183)Fabiola.Gianotti.1
    Formatter URL$1
    Robot and gadget jobsPossibly Mix'n'Match. Also, some author profiles have other identifiers we could match, like ORCID iD (P496) and arXiv-ID (property missing).

    INSPIRE-HEP (Q5972440) is a major database for high energy physics, that includes author profiles identified by a BAI (Bib Author Identifier).

    For most scientists working within the HEP field, the profile is connected to a HEPNames record that contain name variants and their professional history. But there are also author profiles not connected to HEPNames (example), that are not as well curated. So we should probably recommend linking mainly to HepNames-connected records. Danmichaelo (talk) 20:38, 14 June 2016 (UTC)

    •   Support - sounds like a useful id to have. However, aren't they using ORCID now at INSPIRE? ArthurPSmith (talk) 20:25, 16 June 2016 (UTC)
    They support it, but at least at the moment there's lots of authors without one. Danmichaelo (talk) 11:33, 21 June 2016 (UTC)

    @Danmichaelo, ArthurPSmith:   Done Now INSPIRE-HEP author ID (P2930) --Lymantria (talk) 07:18, 23 June 2016 (UTC)

    Dailymotion channel ID

    DescriptionID of the Dailymotion channel of a person, or organisation
    RepresentsDailymotion (Q769222)
    Data typeExternal identifier
    Template parameternone
    Domainorganisation or person
    Allowed valuesall
    ExampleUnion of Democrats and Independents (Q82892)TV-udi
    Formatter URL$1

    Usefull like YouTube channel ID (P2397) Tubezlob (🙋) 13:36, 14 June 2016 (UTC)


    @Tubezlob, Pigsonthewing:   Done Not Dailymotion channel ID (P2942). I added a regex. Lymantria (talk) 06:58, 2 July 2016 (UTC)

    Catalogus Professorum Rostochensium ID

    Descriptionidentifier in the Catalogus Professorum Rostochensium database on professors of the Rostock University from 1419 to present
    RepresentsCatalogus Professorum Rostochiensium (Q1050232)
    Data typeExternal identifier
    Domainhuman (Q5)
    Allowed values\d{8} - 8 digits, leading zeroes allowed
    ExampleHans Spemann (Q57568)00003440
    Formatter URL$1
    Robot and gadget jobsImport by de:Vorlage:CPR gives the wrong values, dewiki makes use of the possibility to reach the individual pages by Integrated Authority File (Q36578). A mix'n'match entry will be created by me after the property is active (or perhaps even before).

    Interesting and reliable information on the careers at Rostock University. Used quite a bit by dewiki. Seems a good addition to wikidata. Lymantria (talk) 16:25, 20 June 2016 (UTC)


    Munk's Roll ID

       Done: Munk's Roll ID (P2941) (Talk and documentation)
    Descriptionidentifier of a person, in the biographical website of the Royal College of Physicians, London
    Data typeExternal identifier
    ExampleWilliam Harvey (Q93128) -> [6]
    Formatter URL$1
    Robot and gadget jobsMunk's Roll is a mix'n'match catalog.

    A series going back to the 17th century of biographies of Fellows of the Royal College of Physicians, and some others; started by William Munk and continued by librarians of the College; "biographical resource for those interested in medical and social history".[7] Currently 6.9K persons, and the site is updated. Charles Matthews (talk) 15:08, 21 June 2016 (UTC)


    @Charles Matthews, Magnus Manske, Sic19, Pigsonthewing:   Done Now Munk's Roll ID (P2941) --Lymantria (talk) 12:20, 1 July 2016 (UTC)

    British Book Trade Index ID

    DescriptionID for the individuals and business entities listed in the British Book Trade Index database that is hosted by the University of Oxford's Bodleian Libraries
    Data typeExternal identifier
    Domainhuman (Q5) or organization (Q43229)
    Allowed values[0-9]
    ExampleWynkyn de Worde (Q725733)78064
    Formatter URL$1

    This database is a valuable source of information on those working as printers, publishers and booksellers in England and Wales up to the mid-nineteenth century. It also contains data on those in related trades, such as engravers, stationers, papermakers and printsellers, which compliments existing external identifier properties like National Portrait Gallery (London) person ID (P1816) and British Museum person or institution ID (P1711). Sic19 (talk) 15:24, 25 June 2016 (UTC)

    •   Support. Seems like a useful property. Thryduulf (talk) 08:42, 26 June 2016 (UTC)

    @Sic19, Thryduulf:   Done Now British Book Trade Index ID (P2945) --Lymantria (talk) 05:26, 4 July 2016 (UTC)

    PO Box

       Done: PO Box (P2918) (Talk and documentation)
    DescriptionPO box as part of an address
    Representspost-office box (Q1162282)
    Data typeString
    ExampleInterest Group for Liberal Weapons Law in Austria (Q1666004) headquarters location (P159) Vienna (Q1741), qualifiers: postal code (P281) 1051, this property 108

    To be used as qualifier on address-like properties. I stumbled across this in the example above while looking for remaining uses of comment (DEPRECATED) (P2315). It seems useful that we can specify a PO box when specifying addresses. --Srittau (talk) 17:28, 5 May 2016 (UTC)

    First, thanks for testing this out as a separate page, I think it works well! Second - I guess this is a necessary complement to P969 (P969)? Or would it be better to have a general mailing address property (which would include PO Box)? ArthurPSmith (talk) 15:54, 6 May 2016 (UTC)
    Yes, I would think that this would go with P969 (P969). On my example above, the PO box was attached to headquarters location (P159), which makes a bit of sense, but not really. And then there is also located on street (P669), which works like P969 (P969), but for street items. I think all of this is a bit of a mess, at the moment. We should maybe rethink how we are doing addresses, but I guess we would need a more public place for that than this proposal. --Srittau (talk) 20:09, 6 May 2016 (UTC)
      Support just to indicate this has my support. ArthurPSmith (talk) 19:23, 9 June 2016 (UTC)

    @Srittau:   Done Lymantria (talk) 10:15, 19 June 2016 (UTC)


       Done: vice-county (P1887) (Talk and documentation)
    DescriptionThe Watsonian (British Isles, except Ireland) or Praeger (Ireland) Vice-county in which the place is located, for the purposes of biological recording.
    Data typeItem
    Template parameterna
    Domaingeographic location (Q2221906) (place), in the w:British Isles.
    Allowed valuesinstances of vice-county (Q7925010)
    ExampleSandwell Valley RSPB reserve (Q7417062) => Staffordshire (Q17581852)
    Format and edit filter validationinstances of vice-county (Q7925010)
    Sourcevarious external lists and publications
    Robot and gadget jobsPossibly, based on coordinates, avoiding edges of vice-counties, or from Wikipedia or other lists
    Proposed byOriginally proposed by Pigsonthewing and created with consensus by Josh Baumgartner as P1887, second discussion created by Ajraddatz.

    Discussion re-created as per Wikidata:Administrators' noticeboard#P1887. As commented there, this property is suggested to be a "key issue for biological recording in the UK". Ajraddatz (talk) 21:29, 30 May 2016 (UTC)

    Note: The previous proposal can be read at Wikidata:Property proposal/Archive/31#vice-county. Thryduulf (talk: local | en.wp | en.wikt) 23:35, 30 May 2016 (UTC)
    Thanks for linking that. Ajraddatz (talk) 20:03, 31 May 2016 (UTC)
    • Support - previous incarnation was not widely used, no indication of actual need for this property. Ajraddatz (talk) 21:29, 30 May 2016 (UTC)
      I also support an incarnation that would have wider use than the UK only - maybe a general one about sub-county-level organization? I'm honestly not sure how it would be organized in other countries. Ajraddatz (talk) 20:03, 31 May 2016 (UTC)
      Changing to support. Looks like it has the potential to work well. Ajraddatz (talk) 22:13, 3 June 2016 (UTC)
    •   Support. This is significant information for biological recording that cannot be recorded on Wikidata using existing properties. There is no deadline, and no required stated anywhere that properties have to be immediately widely used. Thryduulf (talk: local | en.wp | en.wikt) 23:35, 30 May 2016 (UTC)
        Vice-county, ceremonial county and administrative county of Derbyshire
        For reference, the image right is an an example showing how the vice-county of Derbyshire differs from the modern ceremonial and administrative counties of the same name. Thryduulf (talk: local | en.wp | en.wikt) 00:02, 1 June 2016 (UTC)
    • I would   Support a more wider approach than one that only can be used in the UK. Until this proposal becomes more generic I   Oppose. If the target-article is a vice-county or not can be seen in the P31:Q7925010-statement in that item. -- Innocent bystander (talk) 07:35, 31 May 2016 (UTC)
    • Close as out-of-process and instead recreate undelete Property:P1887. The latter was created by consensus, and was deleted without consensus, as Ajraddatz himself has acknowledged. "as per Wikidata:Administrators' noticeboard#P1887" is a bogus justification, as the AN discussion, which Ajraddatz himself closed, was ongoing and itself had no consensus for this action. [I also note that P1887 does not apply to organisations] Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:47, 31 May 2016 (UTC)
      • A discussion that lasted over a month on AN failed to produce any real results. I think it's time we focus on the actual merits of having this property, rather than whatever bureaucratic nonsense you want to go through with. I assume you support re-creating the property? Ajraddatz (talk) 20:32, 31 May 2016 (UTC)
        • We have already had three discussions focused on the actual merits of having this property. The first found consensus to create the property; the second (the first PfD, last December) failed to find any consensus to delete it, with not a single supporter; as did - by your own admission - the next. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 20:37, 31 May 2016 (UTC)
      • Andy, can you explain why it matters whether the old property is recreated or a new one is created? What difference does it make? I've been reading your fight on this for a while and I just don't get why you are so worked up about it. As you and others have said - there is no deadline, so if the property is unavailable for some time (or some other way of handling it is eventually arrived at) what harm is done? I just don't see why this justifies getting a handful of admins (and more by the day, it seems) upset with you. ArthurPSmith (talk) 21:08, 31 May 2016 (UTC)
        • I can't speak for Andy, but stability matters on a project like this. If properties with three consensuses for existence are deleted and recreated (after months of protest and a fourth consensus) as a separate property, what stability is there for reusers of our data? If the original property is recreated then any external tools etc can just be reactivated rather than needing to be rewritten. As for getting people pissed off, this is an unfortunate but I consider that the benefits to the project from ensuring that consensus is respected far outweighs the downsides of people getting pissed off at me. Thryduulf (talk: local | en.wp | en.wikt) 22:53, 31 May 2016 (UTC)
          • If the property was widely used, then that would be a fair point. But it wasn't. Maybe let's focus on its actual merits instead? Ajraddatz (talk) 23:21, 31 May 2016 (UTC)
            • Why does it mater how widely used the property was? Do you know whether someone was collating the data to use it or was preparing a larger project that made use of this? The property was created with consensus that it was beneficial to Wikidata to have a property to store this data, for reasons given in just about every discussion about it, and it was ready to be used at any time (rather preferable to telling an enthusiastic new contributor to wait between a week and a year for their property proposal to be decided upon). In unrelated discussions about property use, nobody has ever been able to explain to me (despite repeatedly asking) why lightly used properties are a problem? The only answers given related to vandalism, misuse and constraint violations, but the volume of and propensity for these are all entirely independent of how many uses a property has (indeed a more used property is more likely to attract vandalism and constraint violations are far easier to sort among a few uses than among many uses). Thryduulf (talk: local | en.wp | en.wikt) 23:41, 31 May 2016 (UTC)
              • Maybe let's focus on its actual merits instead?
                Ajraddatz (talk) 23:50, 31 May 2016 (UTC)
                • As I pointed out to you above, and more recently elsewhere: "We have already had three discussions focused on the actual merits of having this property. The first found consensus to create the property; the second (the first PfD, last December) failed to find any consensus to delete it, with not a single supporter; as did - by your own admission - the next.". Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:36, 1 June 2016 (UTC)
        • Technically, it doesn't matter "whether the old property is recreated or a new one is created", because if the latter action is taken, it will have to be merged into P1887 anyway. What matters is that we don't have admins deleting properties with no consensus; that we don't have admins closing discussions while they are in progress (and despite objections from their recent participants); that we don't allow multiple, disruptive deletion nominations; that we don't invent procedure on-the-fly to cover up the failure of admins to act on reasonable requests to oversee the actions of other admins, and that we don't waste the time of volunteers with redundant, duplicate proposals for properties that have already found consensus for their creation, and no consensus for their deletion. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:34, 1 June 2016 (UTC)
    •   Question @Innocent bystander, Ajraddatz: why do you think something broader is required? To my knowledge the concept of a vice-county is unique to Great Britain, Ireland, the Channel Islands and Isle of Man and this method of recording data for areas that are not used in other contexts (e.g. current or former administrative territorial entities, cultural regions, statistical areas, etc) is not used elsewhere. Certainly the Wikipedia article and other references I've read give no hint that there is an equivalent elsewhere.Thryduulf (talk: local | en.wp | en.wikt) 00:02, 1 June 2016 (UTC)
      • @Thryduulf: The fact that Staffordshire (Q17581852) is a vice-county is already stated in that item. location (P276):Staffordshire (Q17581852) then can fulfil exactly the same purpose and you will get the same information. BUT, to state that a nature reserve, a rare flower or a bird can be found in one specific area with maybe very specific living circumstances actually looks like a good idea. We maybe do not have exactly the same kind of entity in the rest of the world, but to state that something can be found in the limestone-islands of the Baltic Sea or the erosion zones of Ångermanälven still looks very useful. -- Innocent bystander (talk) 05:27, 1 June 2016 (UTC)
      • If we're thinking in a comparative sense, many countries have a sub-county administrative zone that is still above the municipal level. In Canada, these are called township zones, which they apparently are in China as well. I just wonder if this property could be modified to fit all sub-county-but-not-municipal administrative zones, something which could make it more usable and remove that concern as well (though I do agree that widespread use shouldn't be mandatory for a property to be made). Or if it is even needed at all, or even if it could be used as a qualifier within the location property; i.e. location --> whatever, with the qualifier of sub-county admin area --> thing. Though this probably duplicates data that can already be figured out by going from the bottom up, i.e. listing the lowest level and then extrapolating the other administrative districts that it is in. Ajraddatz (talk) 07:31, 1 June 2016 (UTC)
        • @Ajraddatz: I do not like pretending like this hierarchy exists in reality. They maybe do so in Canada or China, but here in Sweden there is a chaos of different types of overlapping administrative territorial entities and they follow no hierarchy at all. A Swedish district do not have to follow the same borders as a municipality, neither a parish or a civil parish do. That something here is located in a district, does often not gives you information in which municipality you live. One family in Boteå district lives in Kramfors municipality, while the rest (I think) lives in Sollefteå municipality. "Located in" "Boteå district" then does not give you enough information to tell where they get their municipal service or where they pay municipal tax. -- Innocent bystander (talk) 10:05, 1 June 2016 (UTC)
        • Yeah, vice-counties are not part of any hierarchy of administrative areas. For most you can infer which country they are in, but at least one is party in the UK and party in the Republic of Ireland (they pre-date even Irish Free State (Q31747)) and it is not unlikely that some span the England-Wales border (England-Scotland border is less likely but still possible, e.g. around Berwick-upon-Tweed (Q504678)). It would be possible to add statements noting a location's vice-county but this would require (a) statements on every location in the British Isles and (b) a property to record the location's vice-county, so I don't recommend that approach. I suppose the vice-county could be added as a second location (P276) statement but that is going to be less clear, possibly harder to query (I'm not sure) and likely to see the entries removed by well-meaning users who don't know what a vice-county is. Thryduulf (talk: local | en.wp | en.wikt) 10:54, 1 June 2016 (UTC)
          • I understand that vice-counties are not used for administrative purposes, but they still represent an organizational unit above the municipal and below the county. So even if there is no explicit administration based on the vice-county that a city / town / etc is in, it should be possible to figure out by the software by listing the lowest possible administrative division. So you take something like Anglian Tower (Q4763423), list the lowest possible administrative division (York (Q42462)), and then by extension the software can figure out that it is part of whatever vice-county it is a part of - 10 minutes of Googling and I can't figure that out. Basically, it isn't an administrative division, but it could be figured out as one in the software. Would it also be possible to add as a second location (P276) statement, with a qualifier that it is a vice-county? Though I suppose there wouldn't be much different between that and creating this property. Ajraddatz (talk) 20:18, 1 June 2016 (UTC)
            • No, they don't represent any organisational unit at all, and it is not conceptually possible to say that they are "above the municipal and below the county" as they exist as entirely independent concepts that have exactly no relation to each other (it's as nonsensical as saying that area control center (Q854272) regions are "above the state and below the country" - they exit independently for different, unrelated purposes). Most are larger than municipal but this cannot be guaranteed, some are larger than counties, others are smaller than counties, some are about the same size but with different borders - at least one is in two different countries. Also, the borders of vice-counties do not change - a given point in the Vice County of e.g. Derbyshire will always be in that vice-county but which administrative and ceremonial county/counties it is in can change at any time - see the image above. location (P276) has been brought up several times, including in the first property proposal - see for example the comment of mine immediately preceeding yours for a couple of reasons why it would be inferior to a dedicated property. Thryduulf (talk: local | en.wp | en.wikt) 23:07, 1 June 2016 (UTC)
              • My own understanding of this was based on Google searches, so thanks for explaining this better to me. Looking at a map, it seemed like the boundaries for the vice-counties intersected with those of counties, but if they are completely different then perhaps this property does make sense. Though, they still represent organizational units by definition - it's just an issue of them not being comparable with the administrative districts. My only other concern is the use for this - it was proposed as a necessary part of bio recording, but went largely unused. Is this actually an important area? Are these vice-counties widely used for that purpose? And is enough data involving them available to promote future use? Ajraddatz (talk) 18:21, 3 June 2016 (UTC)
                • The boundaries of vice-counties are generally similar to ceremonial counties with the same name but they are not identical and some are very different (e.g. Oxfordshire (Q17581832)), there are vice-counties without direct equivalent (e.g. Mid-west Yorkshire (Q17582085)) and others that sound more similar than they are (e.g. West Kent (Q17581823) does cover the western part of present day Kent (Q21694674) but also most of south-east London, at least as far west as Greenwich (Q179385)). Further, the boundaries of ceremonial and administrative counties can and do change, but vice-counties are static by design. As for data availability - there is 160+ years of biological recording and other data that uses the vice-county system, including contemporary recording with tools available to convert grid references and GPS points to vice-county location. As there are many independent organisations using vice-counties the format and license of that data will vary though, but much will be in the public domain by now. There is plenty of potential use, but I cannot promise how quickly that will occur - but as I've still never seen any valid reason why low use of properties is actually a problem (I've seen a couple of attempts, but none that actually stood up to scrutiny - they are no more or less likely to be subject to random vandalism and probably less likely to suffer targeted vandalism; constraint violations are less likely to happen with low-use properties and easier to resolve when they do). Thryduulf (talk: local | en.wp | en.wikt) 19:54, 3 June 2016 (UTC)
                  • Thanks Thryduulf, that makes sense. I've changed my vote to support. I appreciate your time spent walking me through it; perhaps some more information could be included on the talk page or something to make this more clear next time. Ajraddatz (talk) 22:13, 3 June 2016 (UTC)
        • The reason why located in the administrative territorial entity (P131) is not suitable for a vice county was explained in the original property proposal. A suggestion to use location (P276) was also raised there, but was deemed not a barrier to the creation by consensus of the property. There being no consensus for its deletion, this fatuous, duplicate proposal should be speedily closed, and the deletion of P1887 reverted forthwith. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:45, 1 June 2016 (UTC)
      • This is not an external id property for which we do have some very localized properties; it is intended as a property indicating the location of an item within another item. That seems a very generic thing to me. If location (P276) is deemed unsuitable or ambiguous as Thryduulf suggests above, how about creating a new property such as "located within" or "located within the nonadministrative territorial entity" to distinguish it from the administrative version? Another option perhaps is the simple part of (P361) ArthurPSmith (talk) 14:16, 2 June 2016 (UTC)
        • Why is there so much scrabbling around trying to find a broader property - are there any usecases that we are missing? part of (P361) isn't a good fit as e.g. Sandwell Valley RSPB reserve (Q7417062) is no more part of a vice-county than it is part of an administrative county (perhaps even less so). "Located in non-administrative territorial entity" is less problematic on the surface, but what would be the domain and allowed values other than vice-counties? Everything else I can come up that could possibly fit is either not a territorial entity (e.g. Flight information region (Q1433404)) or part of some other administrative structure (e.g. police area (Q7209560)). Thryduulf (talk: local | en.wp | en.wikt) 17:10, 2 June 2016 (UTC)
          • because wikidata, structurally, is benefited by having fewer more general item-valued properties if they are sufficient to capture the information in question. The generic nature of instance of (P31), subclass of (P279) and part of (P361) in particular allow some powerful querying to be done that wouldn't be nearly as easy if there were a collection of disparate properties representing the same kinds of relationships across different domains. String-valued (or external-id valued) and numeric-valued properties are a different matter, as the meaning of the values is very much tied into the specific property. But the power of wikidata is that items (Q-identifiers) are in themselves already highly specific - they don't need further interpretation, in general. We know Staffordshire (Q17581852) is a vice county because it is specified that way, and differs from Staffordshire (Q21694786) and all the other Staffordshire's. We don't need to have the property telling us that the value is a vice county, the item already tells us that. So what is it that this proposed property would be telling us? If it's covered by an already existing property, we really should use that. If we do need a new property, we certainly don't need it to be specific to vice counties. And I would have thought "territorial entity" includes anything that specifies a geographic region in some way, so Flight information region (Q1433404) for example ought to count (though the region specified is above the ground). ArthurPSmith (talk) 14:16, 3 June 2016 (UTC)
    • Still not about organisations. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:33, 3 June 2016 (UTC)
      • I've moved this to its own page (per the new suggested process) and linked it under both Organizations and Places. ArthurPSmith (talk) 14:23, 3 June 2016 (UTC)

    @Thryduulf, Pigsonthewing, Ajraddatz, Srittau, Jura1, ArthurPSmith:  Done The conclusion of this discussion is that there is majority support and substantial forseeable use of this property. I have considered both recreation of the property as a new one and undeletion of vice-county (P1887). For stability reasons I have thought the latter to be the most appropriate. Make good use of it. Lymantria (talk) 11:31, 20 June 2016 (UTC)

    @Thryduulf, Pigsonthewing, Ajraddatz, Srittau, Jura1, ArthurPSmith: (as Ping didn't work) Lymantria (talk) 04:51, 21 June 2016 (UTC)

    percent of water surface

    Descriptionwhich percentage of the territory of this country is water
    Data typeQuantity
    ExampleSingapore (Q334) -> 1,444

    I have seen such thing in Infobox Country in ruwiki and I think this one would be useful. Lingveno (talk) 15:49, 23 May 2016 (UTC)

    • Support, but I think "percent of area water" or "water as percent of area" would be better English labels. "Percent of water surface" suggests that "of the area that is water, this percent is <x>" rather than the intended "of the surface area of <x>, this percent is water". Thryduulf (talk: local | en.wp | en.wikt) 22:39, 23 May 2016 (UTC)
    •   Comment I wonder if this should be a qualifier to "area". Countries may have different views on what parts of the sea are included.
      --- Jura 07:25, 27 May 2016 (UTC)
    •   Support This is a figure I have heard before. --Srittau (talk) 21:35, 11 June 2016 (UTC)

    @Lingveno, Thryduulf, Srittau:   Done Now water as percent of area (P2927). --Lymantria (talk) 05:44, 22 June 2016 (UTC)

    Encyclopedia of Triangle Centers identifier/ETC ID/Kimberling number

    DescriptionIdentifier for triangle centers used in the Encyclopedia of Triangle Centers, founded by Clark Kimberling
    RepresentsEncyclopedia of Triangle Centers (Q1340177)
    Data typeExternal identifier
    Domaintriangle center (Q778806)
    Allowed valuesX\([1-9]\d*\)
    Examplede Longchamps point (Q1807042) → X(20)

    Arthur Rubin
    Nomen ad hoc
    The Anome
    Daniel Mietchen
      Notified participants of WikiProject Mathematics


    The Encyclopedia of Triangle Centers is a comprehensive encyclopedia giving a lot of useful information on triangle centers. Used as a source in several wikipedia's. Lymantria (talk) 08:20, 23 April 2016 (UTC)

    Oh my goodness, I had no idea there were so many different "triangle centers"! I   Support this proposal - however, I don't think "Kimberling number" is a standard name, and what you're proposing for the value is not a number anyway (i.e. has an X and parentheses). Maybe "Kimberling center identifier"? ArthurPSmith (talk) 18:12, 18 May 2016 (UTC)
    You are right about the naming as "number" being less apporpriate (I was hinted by the French and Dutch wikipedia articles linked to by Encyclopedia of Triangle Centers (Q1340177)). I changed the name (added aliases). In publications in stead of parentheses often the number is proposed as subscript. I follow the way it is presented in ETC itself.Lymantria (talk) 21:05, 21 May 2016 (UTC)
      Support. author  TomT0m / talk page 10:50, 8 June 2016 (UTC)

    @Lymantria, ArthurPSmith, TomT0m: done --Pasleim (talk) 09:15, 23 June 2016 (UTC)

    BacDive ID

       Done: BacDive ID (P2946) (Talk and documentation)
    Descriptionidentifier for a microorganism, in the BacDive database
    Data typeExternal identifier
    Allowed valuesValid IDs
    ExampleMycoplasma hominis (Q2616475)8607
    Formatter URL$1

    Notable database. Tobias1984 (talk) 19:30, 11 June 2016 (UTC)

    This is necessary to include links in species articles of bacteria and archaea to the database [BacDive]. In a discussion in [WP Microbiology] a consensus was found to integrate these links. L.C.Reimer (talk) 07:03, 4 July 2016 (UTC)

    output method

    Originally proposed at Wikidata:Property proposal/Creative work

       Done: output method (P5196) (Talk and documentation)
    Descriptionoutput method or device used to interact with a software product
    Representsoutput device (Q778637)
    Data typeItem
    Domainsoftware (Q7397) and video game console (Q8076)
    ExampleAdrift (Q18693102)Oculus Rift (Q3274429)
    See alsoinput method (P479)


    Needed to claim what software supports certain peripherals such as VR headsets (these are not always platforms). Danrok (talk) 14:03, 25 March 2018 (UTC)


    ΛΧΣ21 Vacation9 John F. Lewis (talk) Bene* talk #Reaper (talk) Josve05a (talk) Chris Mason (talk) FunPika Arthena (talk) Wangxuan8331800 (talk) Sjoerd de Bruin (talk) Nicereddy (talk) Syum90 (talk) DrakeCaiman (talk) --George (Talk · Contribs · CentralAuth · Log) Andreasburmeister (talk) Danrok (talk) 18:20, 30 October 2015 (UTC) Macrike (talk) Dispenser (talk) 16:56, 7 July 2017 (UTC) --Zache (talk) 13:34, 12 July 2017 (UTC) SharkD  Talk  06:41, 9 November 2017 (UTC) ZebaX2010 (talk) 00:49, 21 November 2017 (UTC) Sight Contamination (talk) Lewis Hulbert (talk) 20:26, 13 December 2017 (UTC) Jean-Fred (talk) 10:48, 28 February 2018 (UTC) Santer (talk) Cloaker416 (talk) 22:18, 12 June 2018 (UTC) Rampagingcarrot (talk) 19:57, 28 June 2018 (UTC) Diggr (talk) 08:07, 3 July 2018 (UTC) Harsh Rathod Poke me! 09:42, 7 July 2018 (UTC) Kirilloparma (talk) 00:30, 5 August 2018 (UTC) Sir Lothar (talk) 10:10, 10 August 2018 (UTC) Cwf97 (talk) 14:33, 22 October 2018 (UTC) Peterchanws Brasig Le Yota de Mars YotaMoteuchi (talk) 08:09, 22 May 2019 (UTC) Coloradohusky CptViraj BugWarp ʂɤɲ User:Nw520 Cynde Moya Dexxor Floyd-out CadetPatrick AntisocialRyan ihaveahax LotsofTheories (Talk · Contribs · CentralAuth · Log) LotsofTheories (talk) 08:01, 23 March 2021 (UTC)  Notified participants of WikiProject Video games Danrok (talk) 14:06, 25 March 2018 (UTC) Tobias1984
    Ruud Koot
    Daniel Mietchen
    Tinker Bell
    Jasc PL
    Tris T7
    Peb Aryan
    FWVH (passionné d'informatique et d'électronique)
    Sylvain Leroux
      Notified participants of WikiProject Informatics Danrok (talk) 14:55, 25 March 2018 (UTC)

    • Question How do this property work with other devices. Force Feedback controls, Joystick, Wheels, iFeel Mice, Novint Falcon, flight sticks, and motion control platforms. Rumble motors. Light up keyboards, mice, and headphones. What about arcade games like Dance Dance Revolution with the lighting controls? What aspect of the Dreamcast's VMU or the Wii U's GamePad does this cover? Motion Tennis output to the Apple TV from the iPhone[8]. How about all apps with Chromecast support? Dispenser (talk) 14:46, 25 March 2018 (UTC)
      I'd suggest that you would use either input method (P479) or this proposed property based on the primary function of the device, i.e. a haptic control pad is primarily an input device, whereas a printer is primarily an output device (even though most now have displays) Danrok (talk) 15:04, 25 March 2018 (UTC)
      Grand Theft Auto V (Q17452) puts on a light show dependent on game state. On the other side, the Oculus Rift's 9DOF IMU is occasionally used as the controller method and there are some demos using the built-in microphone. Dispenser (talk) 18:30, 31 March 2018 (UTC)
    • I'm not totally sure this is really the best possible property to do this but I failed to come up with anything better, so maybe my feeling is wrong. Therefore   Support as long no one has a better idea. -- MichaelSchoenitzer (talk) 15:39, 25 March 2018 (UTC)
    •   Support David (talk) 17:14, 25 March 2018 (UTC)
    • @Danrok:_What difference do you make between a 3D game and a game that « supports » VR headset ? author  TomT0m / talk page 19:50, 25 March 2018 (UTC)
      @TomT0m: Some games (and software) specifically support VR headsets and their respective controllers, and some may not be playable without. This is because the user interface is designed to work with VR hardware instead of the usual keyboard/mouse. Some software is designed and coded from scratch for use with VR only, e.g. Rec Room. Other "standard" games have had VR support added on, e.g. Elite Dangerous. Some games have a specific VR release version, e.g. Fallout 4 VR Danrok (talk) 20:55, 25 March 2018 (UTC)
    •   Comment Something like this does seem to be needed. However, given the above comments what about a more general "supported device" (and possibly "required device") property? ArthurPSmith (talk) 14:25, 26 March 2018 (UTC)
      • Do "supported device" and "required device" already exist? My problem with "output method" is that the VR controllers are _input_ devices. SharkD  Talk  19:30, 2 April 2018 (UTC)
        • No, those don't exist - but as the proposal noted, input method (P479) is there, so I suppose a matching output property makes sense... ArthurPSmith (talk) 15:12, 4 April 2018 (UTC)
    •   Support Macrike (talk) 19:14, 18 May 2018 (UTC)