Hacker News Reader

Comment by shadowgovt | Hacker News Reader

It's going to get into some very thick first-amendment weeds before section 230 protections are even introduced. Where do we draw the culpability line? Should Facebook have been legally obligated to ban those posts? If so, then if next week a failed coup attempt occurs in Burma to try and overthrow the military power that took control, should Facebook be held accountable for aiding and abetting the coup? We rapidly converge on a scenario where Facebook is culpable for existing, and then we may as well start shutting off forums in general.

With broad fringes, we generally draw the line at "no, the media isn't constrained from propagating a message; the responsibility rests on the speaker and the listener" in the US.

Not to say the suit has no merit. Merely that it faces a stiff headwind and a long uphill battle against a very laissez-faire freedom-of-information-flow precedent.

View on HN

In general I think I agree with you, but on forums this is subtly different.

In a general internet forum, it's up to you decide what you read. On FB, there's an algorithm that determines your feed, and therefore the profit of Facebook itself.

If FB were connecting people, making the world a better place, and making profit by doing that, I think I'd be fine with that.

But it looks like FB can trade the general good will of people, replace it with extremism and make a profit doing so. And if that's true, I think we need to seriously look at section 230.

There's a series of commons that we all share (e.g. water, air). I'm not quite sure if this fits in this category, but for democracy to work, people have to agree to not resort to violence to solve their problems. Trading the common good will for your fellow country men for profit, while watching the world burn, is I think not what we want as a society.

Every forum has an algorithm (including this one, which sorts by upvotes and hides dead content unless the user asks for it).

Hacker News doesn't have an algorithm to decides what's the best posts and should appear in a person's feed. It has editors/moderators.

The contents of the main page aren't decided by editors and moderators. It's an algorithm... granted, a simple one, but an algorithm nonetheless. Weighted factor of upvotes and age.

It's even game-able, if a person has a large enough posse to dump upvotes on a topic after creating it.

So I think there's two big differences.

1. Most people use facebook to keep track of their friends and family rather than email these days. Part of it is that you can announce the new baby in the family and it goes to everyone. But the FB algorithm decides what you view. So the most engaging a post is, the more likely it will show up in your feed.

The problem is that not all posts go to everyone, and if you actually want to see what's happening with Uncle Ed's cancer treatment, you have to physically go to his facebook page. Why? See point #2:

2. There was a time before Twitter/FB where ads paid for the server the forum was on. Social media realized that "targeted engagement" could be profitable. So by tweaking their algorithms to keep people on the platform, they would make more in ad revenue as they can sell more ads.

Of course this led to stories where good friends had died a year ago, and the facebook algorithm "neglected" to prioritize it on feeds, because it wasn't as engaging as the Bills/Jets game the day before.

Hacker News in absolutely no way shape or form has an algorithmic system on the same level Facebook has, it is manifestly a bad analogy

It's not an analogy, it's just a literal truth. HN has an algorithm (of far less complexity than Facebook's) for determining what shows up at the top of the news page and the comments on a topic.

A possible question of interest, were this lawsuit to find Facebook guilty, is "If someone somehow gamed HN's algorithm (which is relatively simple, so would in practice be pretty easy to game) to encourage a genocide, should HN be liable?" How complicated the algorithm is likely does not enter into the reasoning for culpability.

So HN is largely not making money off their website, since they're aren't selling targeted advertising -- at least not to the extent FB makes.

If your targeted algorithm makes the company money off planning a genocide, and if you're aware of it and do nothing about it (because lawyers say blah blah section 230 blah blah), that's a real problem. If you're making money from people whining about a bad call in a football game, so what? But if your company actively promotes criminal content and profits from it, then section 230 probably isn't going to save you. Either it will change, or the courts will force you to.

> Should Facebook have been legally obligated to ban those posts?

Did Facebook promoted that posts? That seems to be the problem. High hate seems to equal high engagement that equals more money.

If Facebook profited of promoting genocide posts, that is clearly morally wrong. It can easily be also illegal and a felony.

Facebook is not a chat but a curated feed of news. Back when it started it was not like that, but when the timeline became the curated feed then everything changed and responsibility is on the curators side.

Personally, I think Facebook's largest problem in this situation is that they were overly reliant on "AI" systems to flag hate speech, which is against their terms of service anyways.

The problem is that Myanmar up until recently (until after the genocide) wasn't even using Unicode, and Facebook only did the bare minimum to get Zawgyi-text processing up.

They also did not hire any Burmese language moderators until it was too late, in a country where Facebook more or less became synonymous with the internet. Meanwhile, Facebook had worked with local phone companies to preinstall the app on phones.

So again, I think this was growth at all costs, without thinking about the consequences, until they got enough egg on their face.