Fake News

[wseltzer]

Discourse, accountability, blocking in social protocols.
Potential connections to provenance, annotation, identity, privacy. Conditional anonymity. Scaling problems. cf. spam, bullying.

[liamquin]

Connections to provenance in copy/paste, in image metadata, being the default and not requiring work, would go a long way to helping fight false memes. Web annotations may have a role in fighting fake news, but have to avoid tyranny of the majority.

[markselt]

Perhaps a WC3 role could be to:
(1)facilitate a (peer?) vetting feature for reviewing news sites or (2) A centralized repository for reputable news sites. here is an example: http://www.usnews.com/news/national-news/articles/2016-11-14/avoid-these-fake-news-sites-at-all-costs (3) Develop a rating system for identifying probable fake news based on number of advertising vectors (4) Plagiarism vs. satire checker. Many fake news sites appear to copy text verbatim and only change choice phrases.

https://www.washingtonpost.com/news/fact-checker/wp/2016/11/22/the-fact-checkers-guide-for-detecting-fake-news/?utm_term=.af4133ed614c

[wseltzer]

https://www.knightfoundation.org/articles/knight-prototype-fund-seeks-early-stage-ideas-to-improve-the-flow-of-accurate-information

[philarcher]

Note this from Google today. It leverages schema.org markup.
https://blog.google/products/search/fact-check-now-available-google-search-and-news-around-world/

[markselt]

Unclear what database Facebook is harnessing but, they too have a tagging mechanism: http://money.cnn.com/2017/04/06/media/facebook-fake-news-tool/index.html

[ghost]

Just found this. I've been looking at how to leverage #5 and RDF / structured data sources, to support an array of different ways to address different issues pertaining overall to trust with media. It's important that if the world thinks the world is flat, and someone says its round; that these sorts of things in the future are not categorised as fakenews. But what's the role of W3C in this area?

[huchunming]

Is there some ways on calculating the creditability? based on rating feedbacks collected from crowds?

[ylafon]

Some TAG discussions here (raw minutes).

[wseltzer]

Credweb CG report https://credweb.org/report/20181011
suggests potential standards around vocabularies for reporting on disinformation.