Announcing the Twitter Trust and Safety Council | Hacker News Reader

Announcing the Twitter Trust and Safety Council

blog.twitter.com

41 pointsprotomyth64 comments

I'm sorry, but a council with more than 40 member organisations sounds a lot like lip service rather than anything substantive. How will this council do anything? The post explains why they need one, and says who they are, but doesn't say what they'll do apart from "tapping into their expertise".

I don't see the ACLU or any other organization dedicated to freedom of expression on that council. This doesn't bode well for Twitter's support of free speech.

Free speech isn't aligned with their business model and their biggest share holders are not so keen on it either. Not saying I agree with it, just stating that it's on par.

For years, Twitter promoted itself as paragons of free speech even calling itself the "free speech wing of the free speech party." http://www.theguardian.com/media/2012/mar/22/twitter-tony-wa...

It looks like they have turned their backs on that. I don't know how to see it otherwise.

Twitter has shown for quite a while that they don't care for freedom of expression (which they might be legally allowed to do, that doesn't make arbitrary censorship of users a morally acceptable choice).

And what happens when "more than 40 organizations and experts from 13 regions" aren't in agreement with each other?

This seems like hand-waving; and if it isn't, look what happened when reddit tried to move away from being an anything goes environment

I assume some of those organizations and experts are more equal than others.

For some perspective of how Twitter has changed over the years, particularly their relationship to free speech, here is Vice's History of Twitter's rules: http://motherboard.vice.com/read/the-history-of-twitters-rul...

It's a worthy goal, I wish the name didn't sound so... I dunno, Orwellian?

> On Twitter, every voice has the power to shape the world.

> Grassroots advocacy organizations that rely on Twitter to build movements and momentum

> ensure that we are enabling everyone, everywhere to express themselves with confidence

Then why are they so eager to remove accounts of people they don't like or who speech they don't like. Weren't they bragging just last week about removing 125000 "terrorist" accounts.

Detecting and hiding abusive tweets algorithmically is NOT THAT DIFFICULT. These methods are employed widely by Facebook and Google. It's baffling to me that Twitter doesn't do the same thing, and is willing to burn money on pointless bureaucracy like this.

Remember the "Other" inbox in FB? Twitter abuse could have been a footnote on par with that. Instead it's the only thing anyone talks about (deservedly so).

When you say something "algorithmic" is "NOT THAT DIFFICULT" on HN, I feel like you shoulder some obligation to explain some of the algorithmic approaches. You know, if it really is "not that difficult".

That sure as shit sounds like a more interesting discussion to have than a semantic debate about free speech.

A regression model trained from existing spam and abuse report data to twitter would cover ~99% of terrible at-replies on twitter.

Look how effectively a hacked-together script that just uses a couple of heuristics like GGAutoBlocker works. [1]

Something a little more principled, trained with unfettered access to user data, would effectively kill the problem. IMO.

[1] https://github.com/freebsdgirl/ggautoblocker

Your understanding of what is difficult is wrong because you're looking at it as a detection problem when it's actually a false positive problem.

See: http://laughingmeme.org/2011/07/23/cost-of-false-positives/

That is an overly simplistic problem description.

It's trying to identify spam/abuse in a universal way, which is a much harder (and barely coherent) problem, and then delete those tweets which makes the cost of a false positive much higher.

What’s far more feasible is to compute the probability a given user would like to see a specific tweet, and then have those below a certain threshold hidden (IE, don’t ping the notification tab for it). This is actually a very well-studied problem, and the massively parallel ML infrastructure it would take exists. Eg, http://www.datanami.com/2014/07/17/inside-sibyl-googles-mass...

It's amazing that Twitter seems to think pouring effort and bureaucracy into "gentrifying" its userbase is somehow going to turn around the service. Twitter has a major problem, but it's not cyberbullying or radical activists; it's that the product is super stale and they haven't innovated in years (ever?).

Not killing their ecosystem would have returned far more value to Twitter than becoming a platform with a hardcore political bias. Their inaugural members go so far as to include Feminist Frequency.

> Their inaugural members go so far as to include Feminist Frequency.

Why is this going "so far"? Anita Sarkeesian has been the source of a bunch of online (and offline) harassment, and has been outspoken against it, especially on Twitter.

> Why is this going "so far"? Anita Sarkeesian has been the source of a bunch of online (and offline) harassment, and has been outspoken against it, especially on Twitter.

Except of course that she's cried wolf on many occasions (going so far as to /invent/ threats and then pretend that she called the police about it and they advised her to tweet about it) and in general has incredibly extreme views (which IMO don't hold up to even cursory scrutiny, but that could be my bias showing). The point is that a "trust and safety" council should also protect people from lynch mobs created by people like Feminist Frequency, right? So maybe we should get thunderf00t (not going to happen) to be part of this council?

I find it hilarious / tragic that the top comment on "twitter attempts to fix their social problems" is that they need to stop doing that and fix their product problems; and the top comment on "twitter attempts to fix their product problems" is that they need to stop doing that and fix their social problems.

Reply on news.ycombinator.com