Live data from Hacker News

“Negative/hate tweets will be max deboosted and demonetized”

twitter.com

21–30 of 412 posts

Re: “Negative/hate tweets will be max deboosted and demonetized”

#21
post #10

Earlier quoted context omitted.

It's impossible to detect negative sentiment purely by text. It's an unsolvable problem.

It's more fundamental than that. Some forms of speech can indirectly or directly limit the speech of others. Free speech absolutism in an unachievable panacea; it is a logical fallacy.

The people who oppose free speech always try to redefine it this way.

Freedom of speech is freedom from censorship or moderation. Not freedom from discomfort

Re: “Negative/hate tweets will be max deboosted and demonetized”

#22
I'm certainly not a fan of how Musk has handled the Twitter takeover, but this sentiment, "New Twitter policy is freedom of speech, but not freedom of reach", seems highly reasonable to me, no?

That is, what I find so awful about most social media (and places like YouTube) is how they focus on some of the most anger-inducing content for "engagement". That is, I think the problem is much more their algorithms than what people actually post.

That said, I think that, realistically, being at the bottom of the heap won't actually be much different from being banned - which I actually think is a good thing.

Re: “Negative/hate tweets will be max deboosted and demonetized”

#24
post #5

There's still a question of how "negative/hate" is defined and whether this will be transparent. I'm skeptical, but the general principle of discouraging and not profiting from negativity is, uh, positive.

We have laws that define this. What changed was that a small group of technocrats took it upon themselves that any questioning of their ideology was “hate speech.”

We have bad laws that badly define this. The US has done no better than Musk at defining "hate."

Re: “Negative/hate tweets will be max deboosted and demonetized”

#25
post #3

Further in the thread he names some banned accounts that will be reinstated. A couple weeks ago, Elon said that Twitter’s content policy (and account reinstatement) would be decided by a moderation council composed of people with diverse views. There have been no reports of this council being formed yet. Edit: why was this post flagged?

From: https://twitter.com/elonmusk/status/1593673844996288512

Kathie Griffin, Jorden Peterson & Babylon Bee have been reinstated.

Trump decision has not yet been made.

Re: “Negative/hate tweets will be max deboosted and demonetized”

#26
post #5

There's still a question of how "negative/hate" is defined and whether this will be transparent. I'm skeptical, but the general principle of discouraging and not profiting from negativity is, uh, positive.

Negativity is an extremely broad concept. There is no chance that “negative” tweets will be uniformly deboosted. Will Elon deboost tweets from Ukrainians condemning Russian war crimes? Will he deboost his own negative tweets toward US politicians? Will he deboost US politicians’ tweets criticizing the other party? Will he deboost tweets critical of his management of Twitter? Discouraging “negativity” necessarily mean…

Want to make some money? Users can pay to boost their own tweets. But I think that's just advertising...

Re: “Negative/hate tweets will be max deboosted and demonetized”

#27

Note that this is the dictionary definition of shadowbanning, which is the primary thing the free speech advocates on Twitter were hoping Elon would stop Twitter from doing.

The difference between hiding a particular problematic tweet and all tweets of an account is important though. It seems Elon refers to the former, while shadow banning refers to the latter. Let's hope it is so.

Re: “Negative/hate tweets will be max deboosted and demonetized”

#28
post #10

Earlier quoted context omitted.

It's impossible to detect negative sentiment purely by text. It's an unsolvable problem.

It's more fundamental than that. Some forms of speech can indirectly or directly limit the speech of others. Free speech absolutism in an unachievable panacea; it is a logical fallacy.

I don't think that's true. The only way to limit speech on an algorithmicially mediated platform is via algorithms and there is no algorithm that can encode non-trivial properties of language regardless of what the AI maximalists might say to the contrary.

Re: “Negative/hate tweets will be max deboosted and demonetized”

#30

Elon “free speech” Musk. >Negative/hate tweets will be max deboosted & demonetized. Guess this guy finally realized that a free for all model is not feasible.

It's impossible to detect negative sentiment purely by text. It's an unsolvable problem.

As someone who specialized in this area for years in social media specifically, I have never seen evidence you can get past 80% accuracy with any known sentiment analysis method.

Brute force stats like Naive Bayes is still as good as it gets on platforms like Twitter where speech patterns are too short and irregular for traditional NLP to be very useful.

Post reply on HN