Wikipedia talk:Wikidata/2018 Infobox RfC/Archive 3

Archive 1

Archive 2

Archive 3

Infoboxes should summarize information already in an article

So... there really should be nothing in an infobox that is not in the article, as all the infobox does is provide a summary of key facts. Right?

It seems to me that pulling content of infoboxes out of Wikidata kind of breaks that ...paradigm. Don't know what folks think about that. Should the RfC say something about this? Jytdog (talk) 22:14, 2 March 2018 (UTC)

My view there is that if the infobox contains information not yet in the article, then the article writers need to catch up. ;-) Thanks. Mike Peel (talk) 22:37, 2 March 2018 (UTC)

That is kind of the cart driving the horse. Jytdog (talk) 22:45, 2 March 2018 (UTC)

Yay, maybe we can switch to automobiles rather than horse-drawn transport soon then? ;-) Thanks. Mike Peel (talk) 22:46, 2 March 2018 (UTC)

WP:INFOBOXPURPOSE refers. · · · Peter (Southwood) ^(talk): 12:03, 8 March 2018 (UTC)

Yes, it says "As with any guideline, there will be exceptions where a piece of key specialised information is difficult to integrate into the body text, but where that information may be placed in the infobox.". Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:26, 7 April 2018 (UTC)

Yes, the RfC should say things about this. This is a precedent that needs to be set and discussed. --Dirk Beetstra ^{T C} 12:19, 8 March 2018 (UTC)

It's not a precedent. For years, we have commonly placed images, maps, chemical formulae, specialised info like DSM-5, etc. in infoboxes with no requirement for them to appear in the body of the article. However, there may be occasions when a piece of information exists on Wikidata, but is not (yet?) in our Wikipedia article. There are multiple reasons for that, and I agree that some cognisance needs to be made of the fact. Here are a few examples of different kinds of situations where Wikidata values are not in Wikipedia. There are clearly very different in context, and require different considerations.

Examples of Wikidata values not in Wikipedia

Type	Topic	Wikidata	Wikipedia	Reason	Conclusion
Unimplemented	Birmingham	Scheduled Monuments in Birmingham (Q17600855): list of monuments (P1456) = Scheduled Monuments in Birmingham		Not yet included, possibly	May be implemented by drawing on Wikidata value
Not accepted/controversial	Night (book)	Night (Q592503): genre (P136) = autobiographical novel, memoir		Genre is considered too controversial to include	Should not be imported
Disputed/wrong nuance	Ludwig van Beethoven	Ludwig van Beethoven (Q255): date of birth (P569) = 17 December 1770	Baptised: 17 December 1770	Date of birth not known, only date of baptism	Should not be imported

A discussion that doesn't recognise a spectrum of reasons for these sort of differences isn#t worth having. --RexxS (talk) 18:45, 7 April 2018 (UTC)

Well put. Btw, Rexx ... is there anything offensive in my vote rationale? These RfCs are frustrating for me, and I don't know what to say that's short and sweet (or at least not bitter). - Dank (push to talk) 19:28, 7 April 2018 (UTC)

I think I originally wrote this because of the argument that I sometimes get removing data - 'but it is in WikiData, so it must be good, and therefore we should transclude it' .. I know that my response is, in line with what you write, 'no, it does not belong as our policies/guidelines argue against it'. --Dirk Beetstra ^{T C} 19:36, 7 April 2018 (UTC)

For me, there is a bit of a risk that just slapping an {{Infobox with only data from WikiData}} on a page could transclude data that is referenced on WikiData and technically merits transclusion, but which is, even for just that article, not supposed to be in the infobox. --Dirk Beetstra ^{T C} 19:38, 7 April 2018 (UTC)

@Dank: I think your !vote rationale is to the point, and I'm impressed by this: "When we understand the difference between what works and what doesn't, then we can craft the right RfC", which puts my feelings better than I could have hoped to. We should spend time looking at why Template:Infobox telescope work well even as an opt-out infobox and Template:Infobox gene works well even without local input. We need to catalogue the problems of trying to implement Template:Infobox person/Wikidata and Template:Infobox book/Wikidata even as opt-in infoboxes. Is it the number of transclusions? Is it that Wikidata just doesn't have sufficient references yet? Or what? It's no good relying on old wives' tales like "some topics just aren't suitable for getting information from a database", that gets trotted out by the Luddites without a scrap of evidence. We need some real solid research to present to the community, otherwise folks will simply !vote according to their own prejudices. If we only want to know the relative proportions of !voters with each prejudice, this RfC will serve the purpose. If we want to work out what course of action is best for Wikipedia (and the other Wikimedia projects, including the other language Wikipedias), then I don't believe the community has sufficient background yet to reach an informed decision. --RexxS (talk) 21:12, 7 April 2018 (UTC)

I agree that Template:Infobox gene works well, from what I've seen, and I'm going to try to figure out why. - Dank (push to talk) 21:40, 7 April 2018 (UTC)

Numbers

In case anyone else is interested, I've put together a chart of # of votes per option at [1] (dividing votes into 2 or 3 where 2 or 3 options for a question have been specified; ignoring ones like 'support', 'oppose' or '54367QRTE$#$$$^'). Of course, that's the crudest possible measure of this RfC, and doesn't take into account people's comments, but it might be a useful first-look. I'll keep it updated as we go; if anyone wants to help then I'm happy to give edit access to the spreadsheet (it's currently set to allow anyone to comment on it). Thanks. Mike Peel (talk) 17:07, 9 April 2018 (UTC)

@Mike Peel: You should also include "total number of !votes cast" (for each question and total). 198.84.253.202 (talk) 21:58, 19 April 2018 (UTC)

Added at the bottom of the first sheet (A32:B36). Thanks. Mike Peel (talk) 22:34, 19 April 2018 (UTC)