Wikipedia talk:Wikipedia Signpost/2023-12-04/In focus

"a free DOI that is unflagged in Wikidata... Sadly, cite q makes it impossible to deal with it here"

No, it does not; for example:



could be changed by the bot to:



and would render as:



However, rather than adding metadata to multiple instances of the same citation, it's far more sensible to hold the data on Wikidata, and to render it as part of each citation from there - which is Cite Q's purpose.

Those of us working on Cite Q, and on citation metadata on Wikidata, would have appreciated being informed of this initiative when it was being developed, in order that the functionality could be rolled out, and metadata updated (by a bot acting on DOI prefixes in exactly the same manner as described above), in parallel. Andy Mabbett ( Pigsonthewing ); Talk to Andy; Andy's edits 16:39, 4 December 2023 (UTC)


 * Well, consider yourself notified. I would have thought Wikidata people monitored CS1/2 talk pages so that cite Q can remain up to date, but that doesn't seem to have been the case. But also cite Q and how it interacts with Wikidata is completely obscure (and we really should not be using it, ever), so no one involved expected it to throw errors like this.


 * Anyway, the current list of registrants can be gotten from the section that starts with

--[[--< B U I L D _ K N O W N _ F R E E _ D O I _ R E G I S T R A N T S _ T A B L E >--
 * in Module:Citation/CS1/Configuration/sandbox
 * &#32; Headbomb {t · c · p · b} 21:18, 4 December 2023 (UTC)
 * Cite Q, and its interactions with Wikidata, are extensively documented. It is not "throwing an error". HTH. Andy Mabbett ( Pigsonthewing ); Talk to Andy; Andy's edits 21:24, 4 December 2023 (UTC)
 * "throwing a maintenance message" then. As for doi-access, cite Q doesn't mention what its equivalent Wikidata property is. &#32; Headbomb {t · c · p · b} 21:31, 4 December 2023 (UTC)
 * The template that Cite Q wraps throws a maintenance message, because it was changed with no notification to the people who maintain Cite Q. Andy Mabbett ( Pigsonthewing ); Talk to Andy; Andy's edits 21:43, 4 December 2023 (UTC)
 * Assuming that all papers with a given DOI prefix are open access is not going to work well in practice, since many publishers have mixed approaches to making papers open access that have been published in their journals. I hope you've been automatically confirming this on a per-paper basis rather than just relying on prefixes? Doing this work on Wikidata to start with (and focusing on data that can be individually checked by a script, rather than blanket automatic assumptions) would have been much better. Thanks. Mike Peel (talk) 21:37, 4 December 2023 (UTC)
 * These are specifically for registrants that have their entire portfolio in open access. MDPI, Frontiers Media, Hindawi, BioMed Central, Athabasca University Press, PeerJ, etc... When they have a mixed portfolio, like IOP Publishing, things don't get flagged. &#32; Headbomb {t · c · p · b} 23:42, 4 December 2023 (UTC)