Wikidata:Property proposal/identifier values as

has similar identifier values as

Originally proposed at Wikidata:Property proposal/Generic

   Not done

Description	property sharing identifier values with an identical scheme, or at least in a similar scheme to be convertible in one way for most values when either is being used. Properties may overlap in domain or not.
Data type	Property
Domain	external-id properties
Allowed values	external-id properties
Example 1	ISBN-13 (P212) → Global Trade Item Number (P3962)
Example 2	Directory of Open Access Journals ID (P5115) → ISSN (P236)
Example 3	Google Arts & Culture entity ID (P10297) → Google Knowledge Graph ID (P2671)
Example 4	OpenAlex ID (P10283) → Microsoft Academic ID (P6366)
Planned use	add where applicable
Expected completeness	eventually complete (Q21873974)
See also	said to be the same as (P460): this item is said to be the same as that item, though this may be uncertain or disputed subproperty of (P1647): all resources related by this property are also related by that property related property (P1659): used to indicate another property that might provide additional information about the subject exact match (P2888): URL used to link two items or properties, indicating a high degree of confidence that the concepts can be used interchangeably complementary property (P8882): property expressing opposing information (e.g. start date and end date). For inverses, use P1696. For properties that negate each other, use P11317. For relations among items, use P461.

Motivation

We keep getting more of these properties, but I don't think have a good way of expressing the relationship. (Add your motivation for this property here.) --- Jura 12:44, 11 February 2022 (UTC)

Discussion

•  Oppose Nopey Nopey Nopey Nope. If you want to find similar things, figure out how to do it with classes or defining properties. Lectrician1 (talk) 06:07, 12 February 2022 (UTC)
•  Comment Interesting idea, though I don't quite see how.
  • Values are currently mostly described by regexes and I don't think I have seen any classes for them.
  • Can you do explain what the classes would be for the samples above? Simiilarly for the other alternative you mention, "defining properties"?
  • For items, there is said to be the same as (P460), but this doesn't work with properties as values due the datatype.

--- Jura 09:59, 12 February 2022 (UTC)

@Jura1 Wait, I notice that ISBN-13 (P212) uses subproperty of (P1647) to say that it's similar to Global Trade Item Number (P3962). Why not just use that? Lectrician1 (talk) 20:39, 13 February 2022 (UTC)

I don't think it works for all four samples, nor does a sub-property relationship imply this. --- Jura 20:57, 13 February 2022 (UTC)

It works for all 4. What does subproperty imply then? See if it were up to me, I'd make items for all identifiers and then use a property between those items to say that some are derived from others. Lectrician1 (talk) 05:25, 14 February 2022 (UTC)

Maybe you can add P1647 in the way you say it works for samples #3 and #4.
We don't have items for all properties and when people create items for some, they tend to get deleted for being in the wrong namespace. --- Jura 07:02, 14 February 2022 (UTC)

Well, they'll likely not be deleted if they're linked to the property with Wikidata item of this property (P1629) Lectrician1 (talk) 12:03, 14 February 2022 (UTC)

Less likely, but I think it still happened. --- Jura 14:27, 14 February 2022 (UTC)

•  Oppose I don't understand what "scheme" is referring to in the proposed description. "I think" you are probably trying to simplify the concept of patterns(forms) or perhaps even encoding ??? where indeed a GTIN-13 number can be encoded in an EAN-13 barcode, for example. GTIN is merely a "set of numbers registered by an authority for identification" and where the encoding mappings and GTIN forms are already available to the industry and not something that a single property in Wikidata will especially help navigate for casual users, or global trade experts. Having said that, you might go to each concept International Standard Book Number (Q33057), European Article Number (Q357404)) and then apply properties on both that show the similarities (number_has = Group, Publisher, Title, Check Digit). Looks like someone did make concepts for some of them specifically ISBN registration group element (Q1654546) and ISBN publisher prefix (Q1654550). But I think using Wikidata's properties to try to re-describe standards or highlight subtle differences or similarities between identification standards is not efficient or useful. That is better done in a blog or industry trade article or book - which the industry already has many and often cited and used. --Thadguidry (talk) 16:49, 14 February 2022 (UTC)
•  Comment More useful would be able to replicate the facts in this sentence "The most commonly used EAN standard is the thirteen-digit EAN-13, a superset of the original 12-digit Universal Product Code (UPC-A) standard developed in 1970 by George J. Laurer." And I'm not sure on the strictness of subclass of (P279) and the term "subset", but maybe that discussion has already occurred and so P279 can be directly used to encapsulate all concepts of "subset of". Don't know. I.E. Could you add EAN-13 (Q3045807) in the subclass of (P279) statement on UPC-A (Q10831949) and is that logical and useful. --Thadguidry (talk) 16:57, 14 February 2022 (UTC)

  Thanks for your input, interesting thoughts. I hadn't worked on our properties (and items) for ISBN parts in while. Indeed these might be of use as well. That being said, the four samples above aren't all the same and to some extent, they include information that should be available in a structured way at Wikidata somehow. For ISBN & GTIN, it's probably simplest, as we probably don't really want both on the same item (see Property_talk:P3962 to discuss further). --- Jura 17:07, 14 February 2022 (UTC)

•  Oppose The correct way to model is: OpenAlex ID (P10283) Wikidata item of this property (P1629) <OpenAlex ID>; Microsoft Academic ID (P6366) Wikidata item of this property (P1629) <Microsoft Academic ID>; <OpenAlex ID> based on (P144) <Microsoft Academic ID>.--GZWDer (talk) 12:23, 16 February 2022 (UTC)

  @GZWDer and then? Somehow this seems incomplete, as it's doesn't seem to say anything on the values. BTW, isn't it the opposite for the first sample? --- Jura 09:27, 18 February 2022 (UTC)

• @Jura1, Lectrician1, Thadguidry, GZWDer:  Not done —MasterRus21thCentury (talk) 05:38, 8 April 2022 (UTC)