thanks for the interesting proposal and discussion, Adam and others. There have been recent proposals to extend and refine the schema.org/ClaimReview model, see Open Claims [1] by Boland et al. as well as my work on Credibility Reviews [2].
I agree that there's an opportunity for Wikimedia to capture more explicitly information it's already generating: at the moment Wikipedia and Wikidata can already be searched for facts, but a lot of the discussions and deletions of claims which did not make the cut are somewhat "lost". It would be great to be able to search for all these claims which have been removed and to have a record as to why they were not deemed to be "factual enough" (and of course, the same for those claims that do appear: why are these considered "factual"). Similarly, there are discussions about which sources are considered reputable (credible) and which aren't. Again, it's not easy to access this information (especially automatically), but a wikifacts/wikiclaims project could help by providing a way of capturing this more explicitly (a big challenge is how to do this without introducing a lot of overhead to the editors/contributors). [1] http://www.semantic-web-journal.net/content/beyond-facts-survey-and-conceptualisation-claims-online-discourse-analysis *under review* [2] https://arxiv.org/abs/2008.12742 kind regards, Ronald On Fri, 5 Feb 2021 at 23:01, Adam Sobieski <[email protected]> wrote: > Thad, > > > > Thank you for the information about schema.org with respect to ClaimReview > <https://schema.org/ClaimReview>. > > > > A point of disagreement with that schema.org model is the ratings system > concept (x out of N). Instead, for discussion, I prefer a more annotational > approach for fact checking with typed annotations produced and consumed by > both humans and software tools. That is, I prefer the concept of an > informational message, warning, error system for fact checking resembling > software IDE’s. Such a model is intuitive, can be readily formalized, made > machine-utilizable, and typed annotations – informational messages, > warnings, and errors – can be merged from multiple sources or service > providers. > > > > > > Best regards, > > Adam > > > > P.S.: As interesting, a new W3C Community Group is launching on the topic > of *document services*. The group intends to discuss and make new > architecture and API to facilitate: spellchecking, grammar checking, > proofreading, *fact checking*, mathematical proof checking, reasoning > checking, argumentation checking, and narrative checking. If the group > interests you, please do feel free to support the creation of the group: > https://www.w3.org/community/groups/proposed/#services . > > > > *From: *Thad Guidry <[email protected]> > *Sent: *Friday, February 5, 2021 3:12 PM > *To: *Discussion list for the Wikidata project > <[email protected]> > *Cc: *[email protected]; Wikispore experimental project > <[email protected]> > *Subject: *Re: [Wikidata] [Wikimedia-l] Idea of a new project: Wikifacts ? > > > > Oops, the better link for the Schema.org work to support fact checking > (some even still in progress after 3 years) probably should have been this: > http://blog.schema.org/2017/08/schemaorg-33-news-fact-checking.html > > Thad > > https://www.linkedin.com/in/thadguidry/ > > https://calendly.com/thadguidry/ > > > _______________________________________________ > Wikidata mailing list > [email protected] > https://lists.wikimedia.org/mailman/listinfo/wikidata >
_______________________________________________ Wikidata mailing list [email protected] https://lists.wikimedia.org/mailman/listinfo/wikidata
