3 ms·

This is a really interesting project. I've wondered for some time why this hadn't already been done. The obvious problem is politics though. We love to talk abo

This is a really interesting project. I've wondered for some time why this hadn't already been done. The obvious problem is politics though. We love to talk about it, but it makes people goofy. And it's trivial to lie to people, through implication, without ever making a single false statement. For instance: - "Uber does not require drivers to pass any sort of driving examination. Uber vehicles are involved in hundreds of more crashes per year than [insert any domestic cab company.]" Both statements are completely honest and true. But the unstated implication is 100% a lie. The lack of context is everything. Uber fulfills 40 million rides a month - that's 480 million a year. If we assume an average trip length of 5 miles, that's 2.4 billion miles a year. That's orders of magnitude greater than any cab company and so without comparing rates of incidence, I've effectively created fake news without ever directly making a single false statement. This is something that regular mainstream media outlets regularly do today. In fact you can find numerous articles using this exact same intentionally broken logic on this exact issue, at various media outlets. It makes for nice scary headlines and thus gets those clicks. I think the question to be seen is whether 'WikiTribune' editors will step above this - and actually require news be true both in content and implication. Just because something isn't directly stated doesn't mean articles suggesting as much are any less culpable of deceit. If they're able to do this, this could be as revolutionary as Wikipedia was.

nileshtrivedi 9y ago
Articles shouldn't be "finished". Users should be able to annotate them (using Hypothes.is?) which can provide more context that authors might have missed.
ZeroGravitas 9y ago
I've often wondered if that's intentional or not, people seem to have real difficulty with comparing things if it's any more complicated than X being bigger than Y. fivethirtyeight.com seem to have a policy of getting numbers correct rather than chasing a story which seems to help. Another good thing is experts that comment on news stories, to place their numbers in context e.g. the NHS in the news service, that covers all the details and is happy to concede that "we don't know" rather than always wrap things up neatly. https://www.nhs.uk/news/pregnancy-and-child/scarlet-fever-cases-england-highest-50-years/ https://www.nhs.uk/news/pregnancy-and-child/scarlet-fever-ca...
andrewla 9y ago
This is the main reason why I like the idea of a wiki-based news source. That "simple" assertion can be clarified by future editors in exactly that manner, and traditional news sources cannot be. Already Wikipedia is excellent in many ways for exactly this -- during the Catalonia independence event, Wikipedia had way deeper context about the situation than other news sources, including information about previous attempts at independence and other historical information that was relevant in contextualizing the events. That said, I'm not sure that this is a feasible vision. Especially since this seems to be producing more long-form journalism rather than current-event based journalism. The line is obviously a bit blurry, but "Big Read: How fact checking evolved in the internet era" is journalism, for sure, but not really the kind of thing that benefits from a wiki approach. "Why Rohingya are world’s ‘most-persecuted minority’" is a lot closer, but the article itself is not aimed at a single continuous entry point for an ongoing event, but instead attempts to provide a single viewpoint into the situation.
ErikVandeWater 9y ago
The most important question is the algorithm to determine what is important - since anyone can participate in WikiTribune at any level, it must be an algorithm that sets what is important. Most algorithms for this are subject to both political bias and "surprise" bias (how surprising an event is, even if it's not important in the grand scheme of things).

nileshtrivedi 9y ago

Articles shouldn't be "finished". Users should be able to annotate them (using Hypothes.is?) which can provide more context that authors might have missed.

ZeroGravitas 9y ago

I've often wondered if that's intentional or not, people seem to have real difficulty with comparing things if it's any more complicated than X being bigger than Y. fivethirtyeight.com seem to have a policy of getting numbers correct rather than chasing a story which seems to help. Another good thing is experts that comment on news stories, to place their numbers in context e.g. the NHS in the news service, that covers all the details and is happy to concede that "we don't know" rather than always wrap things up neatly. https://www.nhs.uk/news/pregnancy-and-child/scarlet-fever-cases-england-highest-50-years/ https://www.nhs.uk/news/pregnancy-and-child/scarlet-fever-ca...

andrewla 9y ago

This is the main reason why I like the idea of a wiki-based news source. That "simple" assertion can be clarified by future editors in exactly that manner, and traditional news sources cannot be. Already Wikipedia is excellent in many ways for exactly this -- during the Catalonia independence event, Wikipedia had way deeper context about the situation than other news sources, including information about previous attempts at independence and other historical information that was relevant in contextualizing the events. That said, I'm not sure that this is a feasible vision. Especially since this seems to be producing more long-form journalism rather than current-event based journalism. The line is obviously a bit blurry, but "Big Read: How fact checking evolved in the internet era" is journalism, for sure, but not really the kind of thing that benefits from a wiki approach. "Why Rohingya are world’s ‘most-persecuted minority’" is a lot closer, but the article itself is not aimed at a single continuous entry point for an ongoing event, but instead attempts to provide a single viewpoint into the situation.

ErikVandeWater 9y ago

The most important question is the algorithm to determine what is important - since anyone can participate in WikiTribune at any level, it must be an algorithm that sets what is important. Most algorithms for this are subject to both political bias and "surprise" bias (how surprising an event is, even if it's not important in the grand scheme of things).