Wikidata:Property proposal/merger
mergee
[edit]Originally proposed at Wikidata:Property proposal/Organization
Description | object that merged into the subject, that existed previously, so as to form an indistinguishable whole. Inverse of merged into (P7888). Use has subsidiary (P355) if the object stays distinguishable as a separate entity aliases merger of, incorporated, integrated, assimilated, acquisition |
---|---|
Represents | mergers and acquisitions (Q731112) |
Data type | Item |
Template parameter | "merger" in en:Infobox organization, "merger" in en:Infobox political party, ... |
Domain | allowed values of the inverse property merged into (P7888) |
Allowed values | domain of the inverse property merged into (P7888) |
Example 1 | ExxonMobil (Q156238) → Exxon (Q4781944) and Mobil (Q3088656) |
Example 2 | Sun Microsystems (Q14647) → Star Division (Q316875) |
Example 3 | CDA (Q273749) → KVP (Q1548365) and CHU (Q143058) and ARP (Q574747) |
Example 4 | Friedrich Krupp AG (Q679201) → Hoesch AG (Q883662) |
Robot and gadget jobs | solve inverse constraint violations |
See also | merged into (P7888) (the inverse), has subsidiary (P355), business division (P199) |
Motivatie
[edit]A, B, C and D being organizations:
A -- \ C ----- / \ B -- D
Currently merged into (P7888) and separated from (P807) exist, which can link from A (and B) to C, and from D to C respectively. I feel like we should be able to traverse back or forth between A and D, but that is currently not possible. The reverse properties "incorporated" (C "incorporated" A and B) and "split-off" (C "has split-off" D) could make that possible.
This property probably needs an inverse constraint with merged into (P7888), which leaves us with quite some constraint violations. I hope a bot could help fix them.
See also: proposal for split-off (permalink). --Strepulah (talk) 19:48, 9 October 2021 (UTC)
Notified participants of WikiProject Companies
Discussion
[edit]- Comment please make the label more explicit. Not that it's confused with merged into (P7888)--- Jura 11:07, 10 October 2021 (UTC)
- @ديفيد عادل وهبة خليل 2, Moebeus, Jklamo, ArthurPSmith, FogueraC, Jheald: @Arbnos: as participants of Wikidata:Property proposal/merged into. --- Jura 11:07, 10 October 2021 (UTC)
- Comment Changed the label from "merger" to "incorporated". --Strepulah (talk) 12:47, 10 October 2021 (UTC)
- Comment In fact, the situation of the motivation can be descrived with A, B replaced by (P1366) C and C replaces (P1365) A, B. The property merged into (P7888) was proposed for situations that can not be modeled with those properties, like this other one, where B merged into (P7888) C:
C -------------- / \ B -- D --
- --FogueraC (talk) 14:10, 10 October 2021 (UTC)
- The same goes for the first figure I think, as they depict the same process. The thing is that replacing and merging are two very distinct concepts, the difference becoming more apperent in the following comparison: Bush replaced by (P1366) Obama < and > Bush merged into (P7888) Obama, the latter obviously being as wrong as it sounds gross. --Strepulah (talk) 16:32, 10 October 2021 (UTC)
- Yes, replaced by (P1366) can be used for other things apart from organizations. But "B merged into (P7888) C" models the situation where C already existed before the merging. If your proposal is that "incorporated" is the inverse property of merged into (P7888), then it does not accept the example in the motivation. If your proposal is that "incorporated" models the example, then it is not the inverse property of merged into (P7888), but a subproperty of replaces (P1365) restricted to organizations. FogueraC (talk) 18:20, 10 October 2021 (UTC)
- It is meant as the reverse of merged into (P7888). I figured it should work for the given example as well, since merged into (P7888) is also being used for that situation (see query). But if it is for entities that existed previously only, then I agree that the inverse should meet that requirement as well. I changed the description to reflect that. --Strepulah (talk) 13:19, 12 October 2021 (UTC)
- Yes, replaced by (P1366) can be used for other things apart from organizations. But "B merged into (P7888) C" models the situation where C already existed before the merging. If your proposal is that "incorporated" is the inverse property of merged into (P7888), then it does not accept the example in the motivation. If your proposal is that "incorporated" models the example, then it is not the inverse property of merged into (P7888), but a subproperty of replaces (P1365) restricted to organizations. FogueraC (talk) 18:20, 10 October 2021 (UTC)
- The same goes for the first figure I think, as they depict the same process. The thing is that replacing and merging are two very distinct concepts, the difference becoming more apperent in the following comparison: Bush replaced by (P1366) Obama < and > Bush merged into (P7888) Obama, the latter obviously being as wrong as it sounds gross. --Strepulah (talk) 16:32, 10 October 2021 (UTC)
- Comment Aside from the fact that inverses generally aren't necessary or wise here, I think the proposed "incorporated" label would be confusing - could easily be confused with inception (P571). "Acquired", "subsumed" maybe? ArthurPSmith (talk) 17:16, 11 October 2021 (UTC)
- Comment Resource Description and Access (Q1519318) has three relationship properties for these relationships: has product of merger: "Relates a corporate body to a corporate body who resulted from a merger of two or more other corporate bodies."; has component of merger: "Relates a corporate body to a corporate body who forms another corporate body by merging with one or more other corporate bodies."; and has mergee: "Relates a corporate body to a corporate body who merged with another corporate body to form a third." --UWashPrincipalCataloger (talk) 16:13, 18 October 2021 (UTC)
- Comment I would assume "incorporated" = "date of incorporation" so I would not support that label. - PKM (talk) 23:14, 18 October 2021 (UTC)
- ArthurPSmith mentioned that too. I changed the label, although you could argue that it makes more sense to change the label of inception (P571) into "date of inception", if that is where the confusion comes from. --Strepulah (talk) 10:45, 19 October 2021 (UTC)
- Comment Changed the label from "incorporated" to "has incorporated". --Strepulah (talk) 10:46, 19 October 2021 (UTC)
- Or "mergee" maybe? --Strepulah (talk) 10:51, 19 October 2021 (UTC)
- Yes, I think "mergee" would be a good option. UWashPrincipalCataloger (talk) 08:11, 22 October 2021 (UTC)
- Changed. --Strepulah (talk) 12:07, 22 October 2021 (UTC)
- Yes, I think "mergee" would be a good option. UWashPrincipalCataloger (talk) 08:11, 22 October 2021 (UTC)
- Or "mergee" maybe? --Strepulah (talk) 10:51, 19 October 2021 (UTC)
- Comment Not sure if "mergee" is better understood. How about "absorbed entity"? Can this also be used on administrative territorial entities? --- Jura 12:47, 25 October 2021 (UTC)
- I liked "mergee" better as it is the inverse of "merged into". Yes, it could be used on administrative territorial entities, and any other of the value-type constraints of its inverse really. --Strepulah (talk) 19:46, 25 October 2021 (UTC)
withdrawn
[edit]So inverse properties are not ideal. Neither do links that only go one way I think, but something will be worked out in the future I assume. Until then the relateditems plugin will suffice. --Strepulah (talk) 15:36, 28 October 2021 (UTC)
- I think it's actually better to create this property than use replaces/replaced by instead. --- Jura 13:45, 29 October 2021 (UTC)
- I also prefer the use of merging to replacing here, I am just not sure about how to best implement inverse relations. --Strepulah (talk) 22:30, 29 October 2021 (UTC)