NEW YORK (Reuters) – In January, a few weeks after Twitter Inc. permanently banned former President Donald Trump after the storms of the US Capitol, the social media company began asking US users to help identify misleading tweets and verify facts in a new pilot program. .
But Birdwatch, which has around 2,000 entrants and is currently barred from its own section of the site, is already facing many of the same challenges as Twitter itself – to see facts from biased opinion and to address the potential of harassment or people try to manipulate the system.
“There’s a lot to do to get there, to the point where we’re comfortable putting these things on tweets,” Keith Coleman, Twitter’s vice president of the product, told Reuters.
“Birdwatchers” can record misleading tweets and record them with “notes” to provide more information, which other participants may find useful.
Under pressure to clean up its site, Twitter began labeling misleading tweets for the first time last year, a move that intensified the debate over the role major social media platforms play in public conversation. It has also fueled claims by Republican lawmakers that technology ventures are censoring conservatives.
In urging users to contribute their own checks, Twitter will need to balance the Birdwatch balancing to make it useful without losing the legality of relying on its community.
Public birdwatching data shows notes ranging from balanced fact checks to biased criticism. Some, for example, labeled the unfounded allegation of widespread voter fraud in the November US presidential election as ‘misleading’. Many simply gave opinions – a tweet from SpaceX and Tesla CEO Elon Musk would have to go to Mars. And stay there ”- while other notes added to opinions.
People are investigating things that professional fact-checkers would never do, said Alex Mahadevan, a reporter for the Poynter Institute’s MediaWise project.
Coleman said the next step of the Twitter team is to update the rating algorithm that should highlight the notes to make sure birdwatchers with different views agree that the information is useful.
‘It’s perfectly good that there is a mixture of quality on the input; what is important is the quality of the production, ”he said.
WISDOM OF CROSSES
Lots of knowledge and community moderation are not new models: they support platforms such as the social network Reddit Inc, and Facebook Inc also runs a “community review” program in which users are paid to identify suspicious content to be screened by professional fact checkers. Thomson Reuters Corp. Reuters is one of Facebook’s paid third-party controllers.
One of the most prominent examples of a crowd-based approach is Wikipedia, where volunteers write and edit millions of articles.
Katherine Maher, CEO of the Wikimedia Foundation that runs Wikipedia, said in an interview that the community’s mission to build an encyclopedia – which it calls what it calls a ‘target platform’ rather than an expression platform, define how contributors behave and that Twitter, which has a more diffuse purpose, can be harder to wrestle with.
Borrowing methods of how Wikipedia promotes and rewards credible contributions can help, Maher said. Public editors on Wikipedia gain greater control over other users based on their work.
Coleman of Twitter said the company is compiling reputation scores for Birdwatchers, based on whether a variety of people find their contributions useful.
Maher also said Twitter will need to develop standards and their application for Birdwatch and decide how people can appeal to a note. That should solve the problem, “did she keep an eye on the viewers?”
GUARDIANS
Travis Whitfill, a health care researcher and venture capitalist in biopharm in Dallas, Texas, said he joined the Birdwatch program as a way to correct medical and COVID-19 misinformation.
Wesley Miller, a 47-year-old research analyst, joined the launch after he briefly stepped down from Twitter last year in protest of the lack of action against Trump.
Jeffery Johnson, a 19-year-old conservative first-year student in Bentonville, Arkansas, said he joined in partly as a joke, but that the idea of users rather than that Twitter decides the truth.
Researchers said it was difficult to see if the program would attract mission-driven volunteers, zealots with agendas or bad actors in the future if it was launched more widely.
To help lead the development of Birdwatch, Twitter said it is creating an advisory board of outside experts with backgrounds ranging from crowdsourcing to political science.
The company has also admitted that it will have to work out how to prevent its unpaid birdwatchers from being harassed because of their notes.
Coleman said it is considering options to remove people’s Twitter handles from their notes and find out if there will be extra rules for Birdwatch content. Contributors may also use pseudonym accounts to protect their identity.
“We do not know what will happen and whether people will feel safe,” Coleman said. “It’s really critical.”
Reporting by Elizabeth Culliford; Edited by Kenneth Li and Dan Grebler