Birdwatch Overview

https://twitter.github.io/birdwatch/about/overview/

by anigbrowl • 5 years ago

64 68 5 years ago

twitter.github.io

agentdrtran• 5 years ago

I'm cautiously optimistic about this, as long as they can prevent brigading, which allegedly they're working on.

samename• 5 years ago

How long until Twitter revokes their API access?

seanyesmunt• 5 years ago

This is an official Twitter product.

a_diplomat2• 5 years ago

In other words, they want people to do fact checking for free for them.

munificent• 5 years ago

Yes, just like HN wants you to write content in the form of comments for free. I get your point, but this is an overly simplistic economic reduction of the incentives and value propositions.

Yes, Twitter wants you fact check for free. They also want you to write Tweets for free. In return, the hypothetical value proposition is that the tweets you and others read are more likely to be factual. If this is successful at scale, it means that the voters in the hopefully democratic place you live will be more informed and more likely to vote for reality based policies.

So, sure, they don't give you a nickel when you fact check. But that doesn't mean it doesn't (possibly) provide sufficient value to you to be worth doing.

gknoy• 5 years ago

This seems like it would be very vulnerable to people supporting a viewpoint (whether true or not) brigading the heck out of posts about the other viewpoint.

a_diplomat2• 5 years ago

It's also the case that there are a lot more people living in undemocratic countries than in democratic ones. I understand the need to have multiple solutions to this problem, but this can never become an alternative for platforms taking more responsibility for content themselves. And, for democratic countries to create a grand coalition to set norms in this space.

AmericanChopper• 5 years ago

Maybe eventually we’ll circle back to letting people make their own judgements about which “facts” to trust.

bigpumpkin• 5 years ago

Only people in the US can become a birdwatcher, yet they can influence the whole of Twitter. Sounds like more recipe for centralization of discourse.

If they open it up to foreign countries, then Congress will complain about foreign influence in American politics.

Catch 22?

dan-robertson2• 5 years ago

I don’t understand why that’s a catch 22 as it seems to me that the outcomes are different depending on the choice they make.

I agree that the way big American tech companies mostly care about the US and suffer from ethnocentrism when stepping outside the US is a problem but I’m not sure what a solution would even look like.

lhnz• 5 years ago

Back in March 2020, this would have meant getting factchecked or shadowbanned for stating that masks should be protective against coronavirus.

thatguy09003• 5 years ago

Don't worry, every time consensus changes it will be revealed that we were always at war with Eastasia to begin with, I'm sure.

a1369209993• 5 years ago

we were always at war with Eastasia to begin with

I mean, you clearly meant to write Eurasia, but that's a really unfortunate typo.

offby37years3• 5 years ago

Or in January 2020 if it was contagious at all: https://twitter.com/WHO/status/1217043229427761152

ceejayoz6• 5 years ago

People keep mixing up “absence of evidence” and “evidence of absence” with this tweet.

There is no reason to believe the WHO was in possession of secret evidence of human-to-human transmission on that date, and they noted that it was entirely possible said evidence would emerge soon. It did shortly afterwards.

paganel• 5 years ago

As late as early February 2020 being concerned about the coronavirus meant you risked being labeled as a right-wing, Trump-lover, conspiracy theorist.

One of the first Twitter accounts that was getting the straight facts out of Wuhan starting with late January was this lady [1], and because she had written for Epoch Times (among other things) of course meant that everything she was saying was potentially a big, fat lie meant to denigrate the Chinese. I was told as such by at least one user on this very website back then.

[1] https://twitter.com/jenniferatntd

ve55• 5 years ago

This is 100% correct, and for those that doubt this I present the following link of a popular Reddit thread on masks from this time period, where the hundreds of commenters appear to unanimously agree that "masks work" is not only pseudoscience, but an outright conspiracy theory with zero backing, concluding that having your kid wear one to school makes you an asshole that follows misinformation (I wish I was exaggerating, but I am only quoting the comments of the link): https://old.reddit.com/r/AmItheAsshole/comments/fe2oqg/aita_...

hirundo• 5 years ago

While they're not using this data to moderate by consensus at this point that's clearly the direction they're heading. Consensus certainly has its place, particularly in democratic institutions. But IMHO it's an anti-feature for a forum. It's exactly the counter-consensus opinions that I find most valuable. To the extent that they are suppressed the forum simply loses my interest. Yes, it's useful to see where the consensus is heading. But mostly it's a snooze, and the compelling content is the stuff that challenges it.

A better use of this data would be to highlight the counter-consensus speakers so that I can have a listen.

runarberg• 5 years ago

Correct me if I’m wrong but hasn’t Stack Overflow used consensus to moderate with great success? What is the reason that consensus driven moderation works well on Stack Overflow but not other forums like Twitter?

tylersmith• 5 years ago

It's been a very long time since I've seen anyone use SO as an example of good community moderation. Everyone I know and the general sentiment I get online agree that their moderation is far too heavy handed and leads to an unpleasant experience for most participants.

shijie• 5 years ago

I would say that with SO, amortized consensus is fine due to there being a "right" answer to a technical problem or question (more or less). Twitter deals mainly in the market of human opinion, which is necessarily far more nuanced with n sides to a story. This makes Twitter the arbiter of truth for their platform. Perfectly legal, as it's their platform, but I wouldn't expect a free exchange of ideas on such a platform.

CDSlice• 5 years ago

On StackOverflow there is a simple, objective measure for judging answers and posts by. Namely “Does this answer actually solve my specific problem?” If there are syntax errors or if the answer plain just doesn’t solve the problem it won’t get upvoted or selected as the correct answer.

On Twitter what objective measure are you going to use to evaluate “The election was stolen by widespread voter fraud”? The news? Half of the country believes that “liberal” news is full of lies and the other half thinks the same of “conservative” news.

GrumpyNl• 5 years ago

I have to disagree, there are to many down votes on legit questions but they did not rephrase it properly or didn't show the proper code. It should be more help full, specially when you down vote, give a reason and solution or advice.

dan-robertson2• 5 years ago

Don’t they suggest that they don’t want to simply take the majority vote? I’m not really sure whether by consensus you mean the majority opinion or a smaller more shared opinion so maybe I’m misunderstanding.

rrsmtz• 5 years ago

Twitter's owners seem to believe that their platform is an oligarchy of opinion leaders (celebrities, etc.), who influence the malleable minds of their aptly-named followers. And they're probably right that this is the overwhelming use case!

I spend a few months in a corner of Twitter with lots of open-minded and free discussion of politics between non-celebrity nobodies. Many accounts in said corner were brigaded, and then shadowbanned or banned outright, etc.

I'm still sore about the platform's intolerance, but it was a powerful lesson about the intolerance in modern society. Birdwatch seems to be yet another mechanism which enables and furthers the shutdown of free expression.

I'm very happy that HN is still capable of having thoughts that stray outside the Overton window, and that zealots are usually ignored =)

HNfriend234• 5 years ago

yup, I have had the same experience, so much so that I stopped commenting and posting entirely. Waste of time. There are so many people on there that just seem to be outright trolling, posting outright non-sense, and doing other stuff that is just an outright nuisance. It significantly reduces the user experience.

Since HN mainly has startup-minded people on here, I think this actually opens up opportunities for new startups to enter the social media space. There is definitely a large audience who is looking for a more "professional" version of twitter so to speak. Start by blocking anonymous accounts/aliases since this is what every troll hides behind.

um_ya• 5 years ago

I think it comes down to group size and subject matter which drives meaningful debate.

When group size is large, conversation is shallow.

HN is a relatively small group with a relatively confined subject matter.

Same concept applies on reddit, with small niche subreddits.

There's more conversation depth with smaller groups of people.

SEJeff• 5 years ago

A more professional version of twitter is linkedin. I'm not being sarcastic. I'm entirely serious.

brobdingnagians• 5 years ago

Good place to post the "Build a real-time Twitter clone in 15 minutes with LiveView and Phoenix 1.5" ;)

https://www.youtube.com/watch?v=MZvmYaFkNJI

cauthon• 5 years ago

Wow. I was under the impression that Twitter had a fairly high bar for banning accounts. Do you have any examples of the tweets that got members of your community banned?

Consultant32452• 5 years ago

The Unity 2020 group was basically an idea to put a center-left and center-right person on the ticket for POTUS and they had some stuff like they'd split areas of concern and trade Pres & Vice pres for their 2nd term. Super basic, non-threatening stuff. Nuked from orbit as soon as they got any traction.

https://articlesofunity.org/2020/09/press-release-for-our-tw...

at-fates-hands• 5 years ago

Twitter recently clamped down on Qanon accounts, suspending some 70K accounts. Conversely, they also suspended several accounts that supported Antifa which had thousands of followers. I'm not sure the better approach. Suspending the user or an account that has thousands of users attached to it.

I honestly couldn't tell you what got them banned; but I'm sure having politically charged tweets probably didn't help either group. It would seem anything political right now is on the radar of Twitter. Aside from that subject matter, I'm not sure what warrants a suspension these days.

FireflyChild• 5 years ago

I was suspended for 7 days for saying a comedian was "retarded".

Then over a year later I was permanently suspended for saying that a bunch of rich scientists that someone was talking about were "a bunch of insufferable cunts".

If only I'd said those things about Trump, I'd still be on there....

thatguy09003• 5 years ago

Hn is a pretty poor place for unusual opinions, with downvoting comments actually making them physically harder to read and all. It's pretty obnoxious when you have to start manually highlighting text to follow a discussion

dang3• 5 years ago

You can click on the timestamp to go to the comment's page, where the text will be readable. It's an annoying extra step, but not as annoying as downgrading the immune system would be, assuming you find fatal illness annoying. Most (<-- notice I said "most") downvoted comments have been downvoted for good reason. Letting the median downvoted comment have the run of this place would kill it.

mistermann• 5 years ago

I spend a few months in a corner of Twitter with lots of open-minded and free discussion of politics between non-celebrity nobodies. Many accounts in said corner were brigaded, and then shadowbanned or banned outright, etc.
I'm still sore about the platform's intolerance, but it was a powerful lesson about the intolerance in modern society. Birdwatch seems to be yet another mechanism which enables and furthers the shutdown of free expression.

I posted a proposal (basically, government mandated, non-partisan, constant auditing of software and processes of all major public communication platforms) in another thread on this story - do you think this might be a plausible approach to addressing this censorship issue, at least in part?

https://news.ycombinator.com/item?id=25908432

t0astbread• 5 years ago

HN works pretty well for the types of discussion it hosts but I'm not sure if it would be as good if it were a general-purpose politics forum. I suspect techno-politics tend to be less emotional compared to other areas of politics.

lasagnaphil• 5 years ago

Although many people in HN have gripes over political discussion in here, I still think this place is (relatively) the best forum to talk about politics. Subreddits are usually in their own bubbles (which has its benefits, but may be boring in the long run), and Twitter doesn’t allow for any long-form discussion. HN is for the most part centrist, but the standard deviation is big enough that everyone would have some people to agree and some people to argue with.

jamestimmins• 5 years ago

It's notable that when it comes to content moderation, most people are positive that the execs at Twitter are morons for not addressing issues sufficiently, but few people can articulate a clear solution to the problem.

I too have concerns about how this will play out, but we should be open minded about different solutions to an important challenge.

dang3• 5 years ago

There's another active thread here: https://news.ycombinator.com/item?id=25906672. Not sure if we should merge them or what.

zestyping• 5 years ago

Hmm. Brigading is a big concern that's already been mentioned.

A concern I haven't seen mentioned yet is that disputes might end up being won through attrition, which is one of the pitfalls of Wikipedia. I used to contribute to Wikipedia, but I just couldn't keep up with folks who had seemingly infinite time to debate or nitpick.

Eventually some moderators will get tired and decide to do something less emotionally taxing with their time; what will that self-selection mean for the kinds of people who will stay?

beastman82• 5 years ago

How about just stop using Twitter

sargun• 5 years ago

This reminds me of oldschool Slashdot Moderation. Or OKCupid moderation.

https://slashdot.org/moderation.shtml

foolinaround2• 5 years ago

It should not be hard to categorize a tweet into two high level categories - fact and opinion.

If it is a fact, then it could marked as true or false, with further proof.

If it is opinion, then there could be a different grading.

Spivak• 5 years ago

Unfortunately this won’t lead to the outcome you desire unless there’s a real tangible reason you want your statements to be put in the category with higher standards. Because otherwise such a system will lead to everything being labeled as an opinion. The reach and impact will end up being the same and even people reporting facts will want to avoid the scrutiny and red tape.

foolinaround2• 5 years ago

I would expect higher attention to facts - and a system that could point this (and other facts) towards other opinions - just allowing readers to make their own informed opinions as well...

bassman9000• 5 years ago

Through consensus from a broad and diverse set of contributors

https://twitter.github.io/birdwatch/about/faq/

The first phase of this pilot will be limited to a small test group in the US

The pilot will start small and grow over time. Our goal is to admit applicants on a rolling, periodic basis. We will admit all participants who meet the required criteria, but if we have more applicants than pilot slots, we will randomly admit accounts, prioritizing accounts that are likely to participate due to having been recently active on Twitter, and those that tend to follow and engage with different tweets than existing participants do — so as to reduce the likelihood that participants would be predominantly from one ideology, background, or interest space.

Aah, the devil in the details.

we will randomly admit accounts, prioritizing accounts

And there's the catch.

known• 5 years ago

Seeking evidence is better

zuhayeer• 5 years ago

Feel like this will definitely be used to point out sarcasm on tweets too

Birdwatch Overview

Related Stories