Wikidata:Requests for comment/Improving handling of labels of deleted entities in archived discussions
An editor has requested the community to provide input on "Improving handling of labels of deleted entities in archived discussions" via the Requests for comment (RFC) process. This is the discussion page regarding the issue.
If you have an opinion regarding this issue, feel free to comment below. Thank you! |
THIS RFC IS CLOSED. Please do NOT vote nor add comments.
- The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.
- The latter discussion is already a planned feature for implementation. The former discussion has been forwarded to the development team. John F. Lewis (talk) 16:47, 27 April 2014 (UTC)[reply]
Note: this RFC was previously titled "Instances of human". The motivation for the RFC was not about a matter of classification, as that title suggested, but rather about how we should handle references to deleted items. At the time of this writing, if an item is deleted, then it appears with a meaningless red link reading something like '(no label) (Q14897290)'. This makes many historical discussions infeasible to understand. This RFC is about how to fix that in the immediate and longer terms.
Since the RFC elicited many off-topic comments while labeled "Instances of human", I have changed it to "Improving handling of labels of deleted entities in archived discussions". Emw (talk) 03:02, 22 November 2013 (UTC)[reply]
The instances classes Q14896454, Q14914342 and Q14870023 were nominated for deletion, twice, but they are not empty. Because these aren't normal items but items intended for use in specifications (P-numbers) I don't think this can simply be decided on Wikidata:Requests for deletions.
Nominations and discussions:
- Wikidata:Requests for deletions/Archive/2013/10/24#Q14870023
- Wikidata:Requests for deletions/Archive/2013/11/15#Instances of human
I will point nominators and commentators to this RfC.
With regards, -Cycn (talk) 13:42, 15 November 2013 (UTC)[reply]
- Items intended for specs? What is a non-human person, or a non-person human? Please direct me to a citable reference that supported their creation. With regard to their being used they are both defined as opposites of human person (yet another tautology), and where you can only have opposite, though honestly I am not certain that they are opposites. It is all nonsensical, confusing and unsupported. I was under the impression that there was that requirement for citable, evidence-based properties, and these have no such supporting evidence beyond someone's opinion. — billinghurst sDrewth 14:16, 15 November 2013 (UTC)[reply]
- I stand by my earlier comments. All people are humans and all humans are people. I don't see how there could be a person that's not a human or a human that's not a person. --Jakob (Scream about the things I've broken) 14:45, 15 November 2013 (UTC)[reply]
- It was decided that human (Q5) would be used for instance of (P31) claims simply because "person" is too ambiguous. There is no reason to have these items if all humans use human (Q5), and all non-humans use something else. The Anonymouse (talk) 17:47, 15 November 2013 (UTC)[reply]
- I can maybe see a use for something like Q14914342 for fiction, myths etc. But I am not sure it is the right path to take... -- Lavallen (talk) 18:59, 15 November 2013 (UTC)[reply]
- For fiction, we can use instance of (P31) --> fictional character (Q95074). --Jakob (Scream about the things I've broken) 19:05, 15 November 2013 (UTC)[reply]
- Isn't Q14914342 the same as juridical person (Q155076) and Q14870023 the same as natural person (Q154954)? Legal personality and natural person are concepts specifically in the field of law, but I can't think of any other fields that would have to distinguish between human and non-human persons (other than fiction, which can use "fictional character" alongside "taling animal", or whatever). Human nonpersons is related to personhood (Q7170675), which is also a legal concept, but is really broad and ill-defined; it covers slavery, women's sufferage, abortion, and other big issues that probably shouldn't be grouped under one umbrella. --Arctic.gnome (talk) 01:06, 16 November 2013 (UTC)[reply]
- Delete: No translation given. No explanation given (discussion page empty). Even if these inconsistent sounding items would make sense in English, WD should also be usable for non English users. --Kolja21 (talk) 02:46, 16 November 2013 (UTC)[reply]
- Comment I created the items in question. They were created as a way to explore the implications of classifying all subjects like Coco Chanel (Q45661) as human person. That discussion was interesting and you should check it out, but I want to emphasize that this RFC should not be about whether we should classify things as human nonperson or nonhuman person. (FWIW, we should not do that, it's a horrible idea; these items should be deleted; but first see below.)
- This RFC should be about a procedural matter for concluded RFCs, and perhaps more generally about how we improve labels for deleted items and properties so that archived discussions can be feasibly understood. For example, let's say we delete the three items that triggered this RFC -- Q14870023, Q14914342 and Q14896454. If we delete those items without changing their usages in the archived RFC in question from using Template:Q, then instead of seeing intelligible labels for those items we would see red '(no label)' links, like Terri Schiavo (Q14897290). It would make that conversation very difficult to comprehend.
- The solution to this problem is simple. Let's get consensus that changing links to deleted items in archived discussions, including concluded RFCs, is OK. In other words, let's establish that it's not a problem to edit archived discussions like Wikidata:Requests_for_comment/Migrating_away_from_GND_main_type (even though it says in bold red "Please do not modify it") and change links to soon-to-be-deleted items like Q14896454 to human nonperson. This way we can make sense of past conversations. Emw (talk) 03:18, 16 November 2013 (UTC)[reply]
- We have to define 3 cases: an human which is a real preson, a fictional character which is human and a fictional character which is not human.For the real beings which are not human like famous animals, we can directly use the specie. 141.6.11.16 17:03, 20 November 2013 (UTC)[reply]
- This RFC is not about whether we should classify things as human nonperson or nonhuman person, etc. The title of this RFC should be improved. The motivation for the RFC does not come directly from a question of how we should classify humans, etc.; it comes from a question about community opinion on fixing / preventing broken references to deleted items in archived discussions. Emw (talk) 02:48, 22 November 2013 (UTC)[reply]
Changing links to deleted items in archived discussions
editThis proposal is to allow labels to deleted properties like Terri Schiavo (Q14897290) to be changed to have a readable label like Terri Schiavo (Q14897290) in archived discussions, including concluded RFCs.
Support This would help us understand past conversations. Emw (talk) 03:18, 16 November 2013 (UTC)[reply]
- I agree that I think this is a big problem. I have sometimes considered starting a RFP for sysop, only to understand what dead links are about. But I'm afraid I would be desysoped for inactivity as soon as that is applied. -- Lavallen (talk) 12:32, 16 November 2013 (UTC)[reply]
Support per Emw --Pasleim (talk) 08:04, 19 November 2013 (UTC)[reply]
Support--GZWDer (talk) 15:57, 22 November 2013 (UTC)[reply]
Support My recommendation would be to keep the label and Q or P number: Terri Schiavo (Q14897290)--Micru (talk) 09:48, 23 November 2013 (UTC)[reply]
- I agree the Q or P number is needed. I've updated the proposal. Emw (talk) 13:36, 4 December 2013 (UTC)[reply]
Redirect items instead of deleting
editI believe (but I cannot find a citation) that the developers are developing a redirect for items so that, when items are merged, the 'merge from' item is redirected instead of being deleted. This is needed so item Qreferences are stable and persistent and don't disappear. This will be important when external websites start to use data from Wikidata. When redirects happen we need to develop a template to link to redirected Qitems.
'human person' should redirect to 'human'; 'non-human person' should redirect to 'person' or 'legal personality'; I have no idea what a 'human non-person' is or what it should redirect to - if we had redirects. Filceolaire (talk) 07:40, 16 November 2013 (UTC)[reply]
Support Filceolaire (talk) 07:40, 16 November 2013 (UTC)[reply]
- I would be certainly open to using this when the functionality is enabled - for now, instead of waiting for something which may or may not ever come to be, we should use what we have IMO. Practices should change after the tech is released to us, not before. Ajraddatz (Talk) 05:28, 21 November 2013 (UTC)[reply]
- Same opinion. Let's wait for it, then we can discuss it.--Micru (talk) 09:50, 23 November 2013 (UTC)[reply]
Support--GZWDer (talk) 15:57, 22 November 2013 (UTC)[reply]