Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Ask HN: Anyone interested in building tools for showing bias in news?

news.ycombinator.com

Post reply on HN
Print view View original

151–160 of 163 posts

Re: Ask HN: Anyone interested in building tools for showing bias in news?

#151

Post by 551199 »

splawn wrote: post #149
551199 wrote: post #148

Earlier quoted context omitted.

Exactly, we could do some much better. Religious attitude is really hurting science.

Sure, there will always be room for improvement. However, I disagree with your latter point, putting ignorant, anti-science, religious, magical thinking people into power is really hurting science. Unlike religion, philosophically, Science is always open for improvement. Edit: I had a comma party.

Repeat of study should be the bare minimum. We are in a situation where lot of underlining assumptions are false. It's hard for common people to experiment science but lets examine something that everybody can experiment daily. Nutritional science. It turns out most of the findings are complete bs or paid by special interests. Probably its not as bad elsewhere but faces similar issues presented in The Economist article.

Re: Ask HN: Anyone interested in building tools for showing bias in news?

#152

Post by cdvonstinkpot »

I was on an Assembly project that tried to do this, think it was called "Flipside". It never went anywhere, IIRC required coders didn't materialize.

Having been inspired by sama's dialogue regarding the downside of unfriending those with opposing views (on the election), I've militantly kept up on opposing Facebook friends' perspectives, giving conscious effort to see their point(s). I see the value in adding opposing news sources to my feed, but the rancor I see (on both sides) is a turn off. Haven't found reliable opposing sources that don't require that I, at least at some level, apply a sort of what I've come to refer to as 'normalizing' their points. So much emotionally charged rhetoric- I guess the 'sizzle' factor sells, but requires additional calories burnt to see through & try not to be disproportionally influenced by.

Maybe a sub-Reddit or sub-Voat -type thing could be built which includes meta-rating elements to allow for rating bias leanings. Dunno what kind of software might already exist that could do this kind of thing for cheap.

tl;dr: A failed Assembly project tried this recently It's hard to create a fair Facebook feed of opposing views A Reddit/Voat -type board with meta-elements to track bias might exist cheap

Re: Ask HN: Anyone interested in building tools for showing bias in news?

#153

Post by pizza »

Chris2048 wrote: post #139
pizza wrote: post #118

Earlier quoted context omitted.

This is what falsifiability is all about in science. Mathematics relies upon proofs, but there is no such thing as a scientific proof.. You can find sufficient conditions for a phenomenon and have a wholly water-tight argument as to why it occurs, but you can't use use observation to find necessary condition. This is because observations are evidence, and if a future observation shows problems with your theory, you s…

I'm not sure this is True, as you can use probabilistic arguments.

Probabilistic models are what I was referring to; all probabilistic models should produce a prediction, and also a measure of confidence. That measure of confidence can never be 100%. Similarly, you can never have an incontrovertible belief that is not a mathematically/logically-constructed fact 'outside' of the messy real world in some abstract chamber walled off from anything that is like a real life observation, no matter how certain. Basically, you can't have P(A) = 0 or 1 as your prior probability, because then you break Bayes' rule in the sense that new evidence is meaningless - https://en.wikipedia.org/wiki/Cromwell%27s_rule

Re: Ask HN: Anyone interested in building tools for showing bias in news?

#154

Post by masondixon »

judahmeek wrote: post #138
hga wrote: post #136

Earlier quoted context omitted.

Except that many generations of immigrants have successfully integrated. Echoing masondixon's reply, we on the Alt Right disagree, starting possibly as far back as the Irish immigrants who played a variety of significant roles in the Civil War, e.g. the NYC draft riots. For a stark example of that quite a bit later, in the 20th Century, see the Curley effect: https://www.google.com/search?q=the+curley+effect It's any…

But where are those Irish who lived under Curley now? Have they not integrated? In America, manifestations of the Curley Effect are far more likely to come from class warfare or political ideology (such as abortion, LBGTQ rights, etc) than from ethnic diversity. I agree that white use of identity politics does concern the Left... because identity politics developed to promote minorities and underrepresented communiti…

> That whites feel the need to act like a minority group is disturbing considering the economic and political power that we obviously weld.

I don't see it being about whites though. I think its everyone except-the-activist-element-of-the-minority-group. People think Trump's victory is a victory for whites, but then cannot understand why some blacks, latinos, muslims, women, gays would vote for him.

> However, if even the most privileged demographic groups engage in identity politics, then economic and political inequality will likely either be maintained or be exacerbated through tyranny of the majority.

The minority always needs the support of the people wielding the power to make changes. The prominence of these movements comes from the whites joining their cause and campaigning for them. So it is in these movements best interests for the privileged demographic to engage in identity politics....or so you would think.

But what is happening now is that the privileged are peer pressured to go along with everything, but are told not to engage or criticise.

Naively, you would think that this social justice machine that has been created is a positive force.

But the backlash from the election of Trump shows that it is moving to fast and is actually counter-productive to the objectives of these movements.

Re: Ask HN: Anyone interested in building tools for showing bias in news?

#155

Post by jquip »

harigov wrote: post #7
noname123 wrote: post #5

This is a pretty good website and they also have a proprietary algorithm to determine the political spectrum of the source: http://www.allsides.com/ NYTimes is ranked moderately liberal while Fox News is ranked right ( http://www.allsides.com/bias/bias-ratings )

Thanks! It seems interesting. The problem I see is that - it's hard to pull people away from their news sources to something unbiased. We need to take unbiased articles to the people. Also, journals shouldn't be the only source. Every side needs to be heard and someone's voice may come from their own blog. I am not sure how it would work but we should at least give it a try.

When you say its hard to pull people away, I think its because they have their belief/bias tuned to that source, not only toward the facts in the article. Then you are dealing with a conflicting thought that goes against the cumulative statements that have been asserted by that source and agreed/verified by the listener. Thus further strengthening the belief/bias of the listener to that source.

Re: Ask HN: Anyone interested in building tools for showing bias in news?

#156

Post by jquip »

htwillie wrote: post #131
harigov wrote: post #66

Earlier quoted context omitted.

Good catch regarding the biased/unbiased. Parent to that post pointed out that there is a news source that tries to show unbiased news. I was implying that we need to take such news sources to people, rather than expecting them to get to that on their own. As I mentioned elsewhere, the idea is not to show unbiased news but rather show other side of the news (in the bias spectrum). I am not entirely clear on what make…

The best "tool" available is the No Agenda Podcast. Hosts Adam and John present national and global issues as portrayed by mass media, and systematically DECONSTRUCT the stories. They not only highlight biases, but explore propagandistic elements and discuss how the media intentionally affects the consumers. Most importantly, they work to identify the MOTIVATIONS for medias' biases in the first place, and demonstrate…

In all aspects, they would be brilliant, but what about spectrum of content coverage. What about readership?

But this dissemination of information by identifying biases of the entities involved is a tactic that can be used by anyone. Objective thinking belongs to everyone, doesn't it?

Re: Ask HN: Anyone interested in building tools for showing bias in news?

#157

Post by jquip »

cdvonstinkpot wrote: post #152

I was on an Assembly project that tried to do this, think it was called "Flipside". It never went anywhere, IIRC required coders didn't materialize. Having been inspired by sama's dialogue regarding the downside of unfriending those with opposing views (on the election), I've militantly kept up on opposing Facebook friends' perspectives, giving conscious effort to see their point(s). I see the value in adding opposin…

I'm thinking on the same lines as you! :)

Re: Ask HN: Anyone interested in building tools for showing bias in news?

#158

Post by jquip »

pinetop wrote: post #65

I recently began working on a (somewhat) related NLP project looking at the shift in sentiment in Trump-focused articles published pre- and post- election. The motivation for the project is the observation that many of the left-leaning news media outlets - who consistently lambasted Trump in the lead-up to the US election - have begun to dial back their criticism post election, presumably in an attempt to (re)build b…

Hallo, any links? I'd be interested.

Re: Ask HN: Anyone interested in building tools for showing bias in news?

#159

Post by jquip »

larubbio wrote: post #64

I had thought about something like this a couple of years ago. My idea wasn't really about tracking bias, but providing reputation for authors and news outlets. Users (via a browser plugin or some other mechanism) could declare an article or statement as biased in a certain direction. That vote would give me information about the article, the news outlet and the reader. I could then present that information back to u…

Very insightful.

Re: Ask HN: Anyone interested in building tools for showing bias in news?

#160

Post by Chris2048 »

ihinsdale wrote: post #147
Chris2048 wrote: post #141

Earlier quoted context omitted.

Now we just need to go through HN an incorporate comment-knowledge into this format :-) Is there good search across questions?

That's an idea! Motivating people to use the argument format has proven a challenge, since it often takes more time to break an argument down into this format than to string sentences together into a paragraph, but I'm confident that is a solvable problem. You can search argument content at https://sequiturs.com/search?q=&page=0&limit=10 . The search indexing is pretty decent considering how easy it was to implement-…

Here's another idea - have a index from news-source article to a page that breaks down it's assertions, then links each one to a discussion of it :-)
Post reply on HN