Live data from Hacker News

Meta censors pro-Palestinian views on a global scale, report claims

theguardian.com

501–510 of 903 posts

Re: Meta censors pro-Palestinian views on a global scale, report claims

#501
post #404

Earlier quoted context omitted.

removal of content is really like 10% of the cases. the most worrisome type of suppression of speech is what they call "shadow banning" which is really "reach limiting". to the extent people now refrain from using the word "Palestine" to avoid metas algirithms. I experienced this myself. We have a draft law in parliament concerning boyocott of israel. I posted a status about the law (notice it's unrelated to war or a…

From my first hand experience designing these algorithms at Meta and running experiments, and listening to feedback from accounts, I would guess that you are not getting reach because your followers do not interact with what you post. In almost every case people call "shadow ban", the cause was occasionally an actual bug (code doing something undesired and unexpected) or far more often people posting bad content. The…

> The ML is good enough to know whether your followers want to engage with political content. If they don't, Facebook will show them less of it.

This is a very simplistic explanation. If the GP is a political activist and "always comments on draft laws", I would expect their followers to be willing to engage with political content.

Re: Meta censors pro-Palestinian views on a global scale, report claims

#502

I have a suspicion that this article is vaguely a response to the previous report that TikTok overwhelmingly censors content in accordance with China's official position.

That entire report was very weird as they compared some uploaded videos with specific hashtags on TikTok vs Instagram. I just don’t understand how that methodology can fly when TikTok is banned in some places, while IG is banned in different ones. So obviously you’ll get stuff anout Hong Kong protests on IG, but not on TikTok since it’s not available in Hong Kong.

There’s also no reliable way of finding where the data is being uploaded from, so huge bot-farms in India wouldn’t push much in TikTok either, cause again, it’s banned there. As much as I hate TT/IG, we shouldn’t try to prove our point with extremely unreliable studies.

Re: Meta censors pro-Palestinian views on a global scale, report claims

#503

I believe its relevant to mention that this doesnt have to be ill intent on part of Meta. Israeli intelligence is spending a lot of resources on influencing narratives. Its not surprising that this would include exploiting and influencing moderation. And its not like social media sites figured out perfect moderation yet, they are an easy target for such determined influences.

Israeli intelligence is likely spending virtually no resources at all on influencing narratives. And if they are, they are failing spectacularly.

More likely, the ministry of the exterior, the IDF spokesperson office, and civilian operations are trying to influence narratives.

From what I can tell these other efforts are not doing so well either.

Re: Meta censors pro-Palestinian views on a global scale, report claims

#504
post #324
post #295

Earlier quoted context omitted.

> get hit by Community Notes How does that work? can anyone post a "community note", or by what process is that decided?

Some related HN discussions: https://news.ycombinator.com/item?id=25906672 https://news.ycombinator.com/item?id=25908439 https://news.ycombinator.com/item?id=33478845

Thanks - it sounds from https://blog.twitter.com/en_us/topics/product/2021/introduci... (via your first link) that they have a rating system whereby community members get to vote on which notes they think are true (a.k.a. which ones they like), and some sort of reputation system to compensate for the weaknesses of relying solely on votes. Did they publish the details of these systems?

Edit: this, from https://communitynotes.twitter.com/guide/en/about/introducti... (via your third link) is interesting:

Community Notes doesn't work by majority rules. To identify notes that are helpful to a wide range of people, notes require agreement between contributors who have sometimes disagreed in their past ratings. This helps prevent one-sided ratings.

Then they link to https://communitynotes.twitter.com/guide/en/contributing/div..., which expands on that. It seems they're basically trying to control for ideological perspective, i.e. to identify signal that doesn't just boil down to "I like / agree with this". I've often wondered if something like that could work.

Re: Meta censors pro-Palestinian views on a global scale, report claims

#505
post #494

Earlier quoted context omitted.

I've seen comments saying "yes please, show me more bombings, I love to see them die" not taken down, because "them" was palestinians. I wonder how long the comment would have been up, if it was referring to the other side instead. edit: I see from the downvotes that my experience doesn't fit your bias. Sorry people -_-

Doesn’t Facebook have a report button? I’m pretty sure comments like that would be taken down if you reported them, it’s really a no brainer whatever “them” refers to.

The report button is rather useless unless it's a delicate issue that Facebook wants to distance themselves from. I find it hard to believe that nobody reports such comments. Probably many people do, but they get the stock reply "this comment does not violate our community standards".

Re: Meta censors pro-Palestinian views on a global scale, report claims

#506
post #216

Earlier quoted context omitted.

> In any case Israel (with the support of the US) has all the power in this conflict and they are the ones actually killing tens of thousands of innocents What happened on Oct 7 then?

What’s happening in Gaza right now? Who is dropping the bombs? Whose army is in Gaza? What is also happening in the West Bank? What about the border in Lebanon. True there was a terroristic raid into Israel in which 1200 people lost their lives but the scale of power does not compare AT ALL

I find this line of reasoning very fascinating. Is your only issue that one side has a lot more weapons? If they were roughly equal would what’s happening in Gaza be ok?

Re: Meta censors pro-Palestinian views on a global scale, report claims

#507
post #314

Earlier quoted context omitted.

Will be free != will be genocided. As I said, it's a creative misinterpretation. If you want to see what a call for genocide looks like, listen to the prime minister of israel: https://www.msn.com/en-us/news/world/netanyahu-openly-calls-... This just scratches the surface of the extreme levels of racism permeating every level of Israeli society and the foreign racists who support it.

I love that you call "from the river to the sea" a creative misinterpretation... then push something that's way more of a creative interpretation of "genocide". Why is "Amalek" a call to genocide (remove the others from our land) but "from the rivers to the sea" (remove the others form our land) isn't? Trying to take you at face value, I looked into what Amalek is and tried to find something more unbiased than you se…

Netanyahu’s governments coalition agreement explicitly states that the Jewish people have an exclusive right on all the land” between the Mediterranean Sea and the Jordan River.

How would you interpret that?

https://www.axios.com/2022/12/29/israel-netanyahu-far-right-...

Re: Meta censors pro-Palestinian views on a global scale, report claims

#508
Meta is as rigged and rotten as it gets. I live in Iran and whenever I have posted something political (not just rants, something serious) or any serious about the gov, it always got taken down. During the Woman,Life,Freedom movement everybody reported how instagram would take down important rebel news and posts and journalisms. you would think meta would work against something like the Islamic republic but it seems like it love all corrupt dictator governments. Let it be Islamic republic or Israel.

Thank god I have left all meta platforms for about 3 years now. But anyone who claims Meta is not pure evil is either evil themself or really really stupid.

Re: Meta censors pro-Palestinian views on a global scale, report claims

#509

It boils down to the algorithmic feed, where Zuckerberg decides what the users see and not the users. Old Facebook didn't have that problem, where there more or less was a chronological order of friends' post. That systems also did not have any problem with "troll farms" (do those even exist, anyway?) since those could not reach you in the first place. On places like Reddit, I get this feeling that some bot network c…

> The atmosphere is way to aligned to be real. I think it’s identical here on HN. Upvoting, downvoting, and using flagging to suppress contrary opinions is rampant and has a huge negative impact toward discussion. I think it’s just the topic du jour on HN is less escalated (usually opinions of tech choices) so it doesn’t seem as extreme.

> using flagging to suppress contrary opinions is rampant and has a huge negative impact toward discussion

I've had showdead turned on for years and have rarely seen contrary opinions flagged, if ever.

The main things I see flagged are personal attacks, open bigotry, flaming, etc. Occasionally I used to see anti-CCP comments flagged, but I haven't seen that in a while. I'd be interested to see some examples where you think something was unjustly flagged.

Re: Meta censors pro-Palestinian views on a global scale, report claims

#510
post #370

This is something people have to relearn over and over again. A lot of the people who are not upset with Meta limiting opinions and content related to the war in Gaza¹. Were gloating and happy when Meta limited opinions and content from right wing politicians² not so long ago. Once you invite censorship for content you dont like, how can you expect it wont bite you in the ass later on. This is why freedom of speech i…

I agree with Supreme Court Justice Clarence Thomas that Congress should consider extending common carrier rules to large social media platforms and force them to carry all legal content. This would in a sense be a form of "compelled speech" in that those companies would have to expend resources to distribute communicates that their management finds objectionable, but at scale the marginal costs would be virtually zer…

At scale the marginal cost would be the platform losing users to smaller platforms with less compelled garbage

And at another scale another cost would be that you're compelled to say something you find objectionable

Post reply on HN