This is a bit like saying "the darkness is bad, so the solution is light, and if the light ever becomes blinding, the solution is more light". Speech is not the point, it's just a medium for useful information. More speech is only good to the extent that it proliferates useful information. If useless spam speech is used to overwhelm useful speech, then that undermines the whole point of the principle of free speech. And with the advent of modern technology, overwhelming the useful speech is easier than ever. Our historic ideas of free speech are wholly unprepared to deal with the overflow attacks that are possible in the modern age. Is it not "censorship" to shout over someone so loudly that they can't be heard?

I see your “The solution to bad speech is more speech” and raise you a “A lie can spread around the world before the truth can get its pants on”. Call or fold?

DDOSing humans rarely benefits the truth. In case of cognitive overload, humans tend to favor the simple answers, not the correct one.

That would only be the solution if the problem was bad speech. The problem is that algorithms optimise for engagement and therefore amplify increasingly extremist views. You cannot talk your way out of that position, it surely must include intervention?

ChatGPT integrations to online forums: Hold my beer.

Cliches aren't an argument, but they're good for clicks. You can do better.

The solution is not as simple as 'more speech' because we are dealing with a historically unprecedented volume of amplified and recorded speech at scale, so things like signal:noise ratio become very important. Simplistic responses like yours suggest an unwillingness or inability to engage with the complexity of the issue.

A simple example of why you're wrong is that it's easy for an actor with access to an echo chamber to launch a viral cascade by modeling outrage over some made-up or minor issue to boost engagement. Fans of the actor share it out of agreement, amusement, or for the pleasure of owning their opponents, while opponents of the actor either deride or argue. The actor gets to appear in the day's trends, gain new followers/subscribers, boosting influence and financial intake.

But there's also a more subtle transfer of wealth, not so obviously measurable by the metrics of the social media platform. The actor spends 1-5 minutes composing their message, like you spend only a short period to select and repost your cliched response. The time that other people spend replying takes at least as long and often longer. So while the actor may have invested 5 minutes out of their 24 hour day in amplifying their own profile, by saying something deliberately controversial, they've caused others to put in significantly more time to responding.

Here, that's not such a big deal because HN discussions are not that complex and people have time to read and consider every comment in a thread they're interested in - so your claim has some validity. But on large social media platforms, the scope is radically different. A provocative tweet can generate thousands of negative replies, so that an investment of 5 minutes by the provocative actor can cause people to collectively put hundreds of hours into their responses, which unconscious social labor serves to amplify the provocateur. There are lots of influencers who have this down to a system and do it on an almost daily basis. It's essentially a theft of time - a little from their fans, a lot from their anti-fans, with that time being converted into engagement and increased reach.

This is why your reflex response is wrong. Bad speech can be designed to yield more speech that amplifies the original bad message for fun and profit. Your concept of reasoned debate only functions in forums where there are some leveling factors like equal time allocations or legal procedures for the allocation of time. Here on HN and on many small forums, that function is done by moderators (who can ban intentionally disruptive users or manage anti-spam filters) and to some extent by users (whose expressed preferences have direct weight like downvotes or flags, and indirect weight by their long-term community participation making them familiar to each other). On big platforms, the calculus is completely different and your model of reasoned debate breaks down.

the low s/n then makes it about who can be loudest and/or most provocative. i don’t disagree that disinformation shouldn’t just be squelched, but it’s not as simple as “education” or “more speech” and as humans we tend to prefer the path of lead resistance.

well, we see where that ends by looking at 4chan. Why are you posting here and not there?

Because their customers/voters want it? Me, I very much prefer censored HN to free speech 4chan. The question is, what are you doing here if you prefer 4chan where the "truth" is not censored?

So true! Now please step into the boxcar. If you resist, you're the one at fault.