Wikidata:Property proposal/KIT Linked Open Numbers ID

KIT Linked Open Numbers ID

Originally proposed at Wikidata:Property proposal/Authority control

   Done: KIT Linked Open Numbers ID (P5176) (Talk and documentation)

Description	identifier for a natural number in the KIT
Data type	External identifier
Domain	natural number (Q21199)
Allowed values	[1-9]\d*
Example	42 (Q812996) → 42 2922 (Q19243120) → 2922
Source	http://km.aifb.kit.edu/projects/numbers/
External links	Use in sister projects: [ar] • [de] • [en] • [es] • [fr] • [he] • [it] • [ja] • [ko] • [nl] • [pl] • [pt] • [ru] • [sv] • [vi] • [zh] • [commons] • [species] • [wd] • [en.wikt] • [fr.wikt].
Number of IDs in source	1 trillion
Expected completeness	always incomplete (Q21873886)
Formatter URL	http://km.aifb.kit.edu/projects/numbers/web/n$1

Motivation

Karlsruhe Institute of Technology's Linked Open Numbers database contains entries for the first trillion natural numbers, as linked, open data under a CC0 licence. These already include links to Wikidata, where available. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 20:50, 1 April 2018 (UTC)

Discussion

•  Comment I am wondering whether you are planning to import all the trillion numbers or just some of them. John Samuel 21:04, 1 April 2018 (UTC)
  • Only those where there is already a Wikidata item; hence "Expected completeness: always incomplete". Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 21:07, 1 April 2018 (UTC)
•  Support --Denny (talk) 21:56, 1 April 2018 (UTC)
•  Support David (talk) 14:37, 2 April 2018 (UTC)
• {{Notdone}} April 1 is over. Funny one.
  --- Jura 11:52, 3 April 2018 (UTC)
  • @Jura1: I don't think it was intended as a joke. @Pigsonthewing: can you please confirm? Micru (talk) 07:28, 5 April 2018 (UTC)
    • I've reverted Jura's disruptive closure of this proposal. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:22, 5 April 2018 (UTC)
      • Thanks for the confirmation. John Samuel 17:53, 5 April 2018 (UTC)
•  Support Micru (talk) 11:40, 5 April 2018 (UTC)
•  Oppose It's a good proposal for April 1, but it doesn't really add anything to Wikidata that couldn't be generate here directly.
  --- Jura 12:09, 5 April 2018 (UTC)
  • False; since it adds an external identifier which - by definition - can't be done internally. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 14:40, 5 April 2018 (UTC)
    • funny. An identifier for a list of Wikidata items and hex conversions referenced elsewhere
      --- Jura 16:56, 5 April 2018 (UTC)
      • No: an identifier for a concept; as most people here will know. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:45, 8 April 2018 (UTC)
        • Maybe you could explain what the resource provides that we don't?
          --- Jura 13:12, 8 April 2018 (UTC)
          • I did, at 14:40, 5 April 2018 (UTC). Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:22, 8 April 2018 (UTC)
•  Comment How about instead of this proposal, establishing a mechanism for a third-party formatter URL based either on the numeric value or some other property (or label?) of an item? It does seem useful to have a routine way to generate links to external sites like this. ArthurPSmith (talk) 16:44, 5 April 2018 (UTC)
  • This property would actually be a good proxy for querying integers at Wikidata.
    --- Jura 16:56, 5 April 2018 (UTC)
  • @ArthurPSmith: It will be useful to have a separate track for discussion of external identifiers. John Samuel 17:55, 5 April 2018 (UTC)
  • {{Support}} Arthur's proposal. Mahir256 (talk) 03:09, 6 April 2018 (UTC)
  •  Support I adjusted the formatter url to make this easier. We could just name this property external integer identifier.
    --- Jura 07:39, 6 April 2018 (UTC)
    • Restored. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:44, 8 April 2018 (UTC)
      • Why did you make the formatter url less convenient?
        --- Jura 13:07, 8 April 2018 (UTC)
•  Comment There is more discussion about this at Wikidata:Project_chat#Prime_factors? and another unanswered question. @ديفيد عادل وهبة خليل 2, Denny: seems you got pranked on April 1.
  --- Jura 17:08, 9 April 2018 (UTC)

  Do you know that I am a co-creator of Linked Open Numbers? I am rather well aware of how and why it was created :) --Denny (talk) 17:31, 9 April 2018 (UTC)

  Oh .. no. I'm aware of your great achievements at Wikidata. So why abandon us to do it elsewhere? :)
  --- Jura 17:36, 9 April 2018 (UTC)

  Thank you for the kind words. Because it was created in 2010, before Wikidata was started :) --Denny (talk) 19:23, 9 April 2018 (UTC)

• @Jura1: you have both a support and an oppose vote here. Also April fools? And can Denny comment on whether this website is actually likely to be around and useful to link to (i.e. do you support this proposal?) ArthurPSmith (talk) 21:36, 10 April 2018 (UTC)
  • @ArthurPSmith: He already did; and he already confirmed that the site has been around longer than Wikidata. Why have you marked this as not ready? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 23:32, 10 April 2018 (UTC)
  • Based on your suggestion, I think it would be useful if made with a slightly different format. It might not even need a formatter url. Given Denny's comment, it seems that the website is stale and hasn't evolved in many years. As Pigsonthewing notes implicitly, there isn't really any content beyond it being an identifier. I'd restore this version.
    --- Jura 04:36, 11 April 2018 (UTC)
    • "Given Denny's comment, it seems that the website is stale and hasn't evolved in many years." Poppycock - Denny said no such thing; because to do so would be false. "As Pigsonthewing notes implicitly, there isn't really any content beyond it being an identifier. And that is not what I either said or implied. Do not attempt to speak for me. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 14:53, 11 April 2018 (UTC)
•  Comment I didn't expect this to be so heated. The website sure is stale-ish - there are rare bursts of activity on it. Then again, that's partially in the nature of the beast - the kind of knowledge covered about natural numbers doesn't exactly change particularly frequently. I wish that this discussion was far more light-hearted. Personally I think such a link would be cute, but certainly not essential - it is kinda predictable - and there is hardly any chance for data getting bad or something. But I am obviously biased regarding the dataset. --Denny (talk) 16:08, 11 April 2018 (UTC)
  • I think the general idea is nice. I hope eventually one day we get https://numbers.wikidata.org/ to do more of that. If it just wouldn't be so empty. Even the proposer can't really describe what it would have beyond links to Wikidata, his traditional invective and a selfquote. The discussion on project chat suggested another resource instead and even here we have suggestions how to make this more useful. Unfortunately, this hasn't really led to the proposal to evolve. Has anything changed on that site since the start of Wikidata?
    --- Jura 16:33, 11 April 2018 (UTC)
    • Your arguments against this property having been refuted, you're now resorting to dishonesty {"can't really describe what it would have beyond links to Wikidata") and ad hominem. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:46, 11 April 2018 (UTC)
  • Regarding has anything changed on that site since the launch of Wikidata: in fact, yes. --Denny (talk) 15:54, 12 April 2018 (UTC)
•  Strong support It's really shame to miss a stable list of IDs. --Liuxinyu970226 (talk) 14:40, 26 April 2018 (UTC)
•  Comment the unclear value added by the website (other than having trillions of pages and providing the sum of digits), the general staleness of the website, and the format proposed that makes it unpractical for use by others (e.g. insisting on "n42" instead of "42"), suggests that there isn't really suitable for Wikidata. Not sure what to conclude from the lack of civility of the proposer and the fact that they haven't attempted to address these issues.
  --- Jura 09:39, 30 April 2018 (UTC)
•  Weak oppose I think it would be helpful that the ID would be the number itself. This could be realized by using the formatter http://km.aifb.kit.edu/projects/numbers/index.php?number=$1 or change the formatter in a way the proposer seems to dislike into http://km.aifb.kit.edu/projects/numbers/web/n$1. For me the contents would be just enough to consider supporting the proposal if the prime factorizations were to stretch to larger numbers (the largest number factorized is 999,999). Lymantria (talk) 20:59, 4 May 2018 (UTC)
  • @Denny: You supported the proposal as drafted; do you see any merits in modifying it as Lymantria proposes? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 22:16, 4 May 2018 (UTC)
•  Support only with the n removed. At least 3 editors prefer this, so I will reinstate this change. --99of9 (talk) 02:09, 7 May 2018 (UTC)

Formatter URL

Unfortunately, Mircu insists on edit warring this proposal, claiming a "majority" support his version. This is false. I (as proposer), Denny, David, and Liuxinyu970226 all explicitly supported the version proposed. As, ironically, did Micru himself (on 5 April). That's a count of five. Only three people have actually expressed support for the other version. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 19:25, 7 May 2018 (UTC)

That people (including me) supported the creation of the property, doesn't mean that the same support applies to the kind of formatter. From my side I consider the discussion for the creation of the property resolved (majority supports the creation of this property). Now the only issue left to clarify is the formatter URL. I am pinging everyone that participated in this discussion so we can have a clear overview of who supports each option. --Micru (talk) 20:04, 7 May 2018 (UTC)

It beggars belief that you would suggest that people who supported the proposal did not support what was in the proposal. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 21:33, 7 May 2018 (UTC)

@Pigsonthewing, Jsamwrites, Denny, ديفيد عادل وهبة خليل 2, Jura1:@ArthurPSmith, Mahir256, Liuxinyu970226, Lymantria, 99of9: Please state your preference.  – The preceding unsigned comment was added by Micru (talk • contribs) at 20:05, 7 May 2018‎ (UTC).

I don't need to restate my prefence; I have aready stated it in the proposal I submitted. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 21:33, 7 May 2018 (UTC)

Option A: The "n" stays in the formatter, not in the value

With this option the formatter would look like this http://km.aifb.kit.edu/projects/numbers/web/n$1 and the value would look like this 2922.

 Support--Micru (talk) 20:04, 7 May 2018 (UTC)

 Support --Lymantria (talk) 20:12, 7 May 2018 (UTC)

 Support ArthurPSmith (talk) 20:24, 7 May 2018 (UTC)

 Support (to clarify, this does not change my support for Arthur’s suggestion above) Mahir256 (talk) 21:09, 7 May 2018 (UTC)

 Support 99of9 (talk) 22:24, 7 May 2018 (UTC)

Option B: The "n" is in the value

With this option the formatter would look like this http://km.aifb.kit.edu/projects/numbers/web/$1 and the value would look like this n2922.

Outcome

@Pigsonthewing, Jsamwrites, Denny, ديفيد عادل وهبة خليل 2, Jura1:@ArthurPSmith, Mahir256, Liuxinyu970226, Lymantria, 99of9: Property created as KIT Linked Open Numbers ID (P5176). Formatter URL as per Option A. If someone can create the statements for all the natural numbers, please do so.--Micru (talk) 00:03, 20 May 2018 (UTC)