> Any idea what our next steps are? [..] try to figure out a less damaging knowledge-based filtering mechanism [..]
It should cost some amount of money to post anything online on any social media platform: pay to post a tweet, article, image, comment, message, reply.
(Incidentally, crypto social networks have this by default simply due to constraints in how blockchains work.)
This is a great idea to prevent bots, but that’s not who posts the bad stuff this thread is talking about. Wherever you set the threshold will determine a point of wealth where someone can no longer afford to speak on these platforms, and that inevitably will prevent change, which tends to come from the people not well-served by the system as it is, i.e. poor people. Is that your goal?
> a point of wealth where someone can no longer afford to speak on these platforms, and that inevitably will prevent change, which tends to come from the people not well-served by the system as it is, i.e. poor people.
"Change" in itself is not a virtue. What I think you want is good or beneficial change? That said, what evidence do you have that poor people specifically are catalysing positive change online?
> This is a great idea to prevent bots, but that’s not who posts the bad stuff this thread is talking about.
There is no difference between a bot and a human as far as a network is concerned. After all, bots are run by humans.
The article specifically says that: "The images and videos including necrophilia, bestiality and self-harm caused some moderators to faint, vomit, scream and run away from their desks, the filings allege."
> Is that your goal?
Simply making it cost something to post online will mean that people who want to post spam can directly pay for the mental healthcare of moderators who remove their content.
If it turns out that you can find a group of people so poor that they simultaneously have valuable things to say online yet can't afford to post, then you can start a non-profit or foundation to subsidize "poor people online". (Hilariously, the crypto-bros do this when they're trying to incentivize use of their products: they set aside funds to "sponsor" thousands of users to the tune of tens of millions of dollars a year in gas refunds, airdrops, rebates and so forth.)
> "Change" in itself is not a virtue. What I think you want is good or beneficial change? That said, what evidence do you have that poor people specifically are catalysing positive change online?
We would probably disagree on what change we think is beneficial, but in terms of catalyzing the changes I find appealing, I see plenty of it myself. I'm not sure how I could dig up a study on something like this, but I'm operating on the assumption that more poor people would advocate the changes I'm interested in than rich, because the changes I want would largely be intended to benefit the former, potentially at the expense of the latter. I see this assumption largely confirmed in the world. That's why I find the prospect of making posting expensive threatening to society's capacity for beneficial change. The effect depends on what model you use to price social media use, how high you set the prices, how you regulate the revenue, etc, but I think the effect needs to be mitigated. In essence, my primary concern with this idea is that it may come from an antidemocratic impulse, not a will to protect moderators. If you don't possess that impulse, then I'm sorry to be accusing you of motives you don't possess, and I'll largely focus on the implementation details that would best protect the moderators while mitigating the suppression of discourse.
>you can start a non-profit or foundation to subsidize "poor people online".
Where are all the foundations helping provide moderator mental health treatment? This is a pretty widely reported issue; I'd expect to see wealthy benefactors trying to solve it, yet the problem remains unsolved. The issue, I think, is that there isn't enough money or awareness to go around to solve all niche financially-addressable problems. Issues have to have certain human-interest characteristics, then be carefully and effectively framed, to attract contributions from regular people. As such, I wouldn't want to artificially create a new problem, where poverty takes away basically the only meaningful voice a regular person has in the modern age, then expect somebody to come along and solve it with a charitable foundation. Again, if charity is this effective, then let's just start a foundation to provide pay and care to moderators. Would it attract contributions?
>the crypto-bros do this when they're trying to incentivize use of their products
The crypto-bros trying to incentivize use of their products have a financial incentive to do so. They're not motivated by the kindness of their own hearts. Where's the financial incentive to pay for poor people to post online?
>There is no difference between a bot and a human as far as a network is concerned. After all, bots are run by humans.
Most implementations of this policy would largely impact bot farms. If posts cost money, then there's a very big difference in the cost of a botnet and a normal account. Costs would be massively higher for a bot farm runner, and relatively insubstantial for a normal user. Such a policy would then most effectively suppress bots, and maybe the most extreme of spammers.
What I don't understand, then, is the association between bots/spammers and the shock garbage harming moderators. From what I know, bots aren't typically trying to post abuse, but to scam or propagandize, since they're run by actors either looking for a financial return or to push an agenda. If the issue is spammers, then I'd question whether that's the cause of moderator harm; I'd figure as soon as a moderator sees a single gore post, the account would get nuked. We should expect then that the harm is proportionate to the number of accounts, not posts.
If the issue is harmful accounts in large quantity, and easy account creation, then to be effective at reducing moderator harm, wouldn't you want to charge a large, one-time fee at account creation? If it costs ten dollars to make an account, bad actors would (theoretically) be very hesitant to get banned (even though in practice this seems inadequate to, e.g., suppress cheating in online games). I'd also be relatively fine with such a policy; nearly anyone could afford a single 5-10 usd fee for indefinite use, but repeat account creators would be suppressed.
>Simply making it cost something to post online will mean that people who want to post spam can directly pay for the mental healthcare of moderators who remove their content.
I don't think that adding a cost to the posts will end up paying for mental healthcare without careful regulation. The current poor treatment of moderators is a supply-demand issue, it's a relatively low-skill job and people are hungry, so you can treat them pretty bad and still have a sufficient workforce. They are also, if I'm correct, largely outsourced from places with worse labor protections. This gives the social media companies very little incentive to pay them more or treat them better.
An approach that might help is something like this: Require companies to charge a very small set amount to make each individual post, such that a normal user may pay in the realm of 5 usd in a month of use, but a spammer or bot farm would have to spend vastly more. Furthermore, but very important, require that this additional revenue be spent directly on the pay or healthcare of the moderation team.
In reality, though, I'd be very worried that this secondary regulation wouldn't enter or make it through a legislature. I'm also concerned that the social media companies would be the ones setting the prices. If such a cost became the norm, I expect that these companies would implement the cost-to-post as a subscription to the platform rather than a per-post price. They would immediately begin to inflate their prices as every subscription-based company currently does to demonstrate growth to shareholders. Finally, they'd pocket the gains rather than paying more to the moderators, since they have absolutely zero incentive to do anything else. I think this would cause the antidemocratic outcomes I'm concerned with.
My question for you, then, is whether you'd be interested in government regulation that implements a flat per-post or per-account-creation fee, not much more than 5usd monthly or 10usd on creation, not adjustable by the companies, and with the requirement that its revenue be spent on healthcare and pay for the moderation team?
Your reply is rather long so I'll only respond to 2 sections to avoid us speculating randomly without actually referring to data or running actual experiments.
To clarify:
> That's why I find the prospect of making posting expensive threatening to society's capacity for beneficial change.
I suggested making it cost something. "Expensive" is a relative term and for some reason you unjustifiably assumed that I'm proposing "expensive", however defined. Incentive design is about the marginal cost of using a resource, as you later observed when you suggested $5.
We often observe in real life (swimming pools, clubs, public toilets, hiking trails, camping grounds) that introducing a trivial marginal cost often deters bad actors and free-loaders[^0]. It's what's referred to in ideas such as "the tragedy of the commons".
> An approach that might help is something like this: Require companies to charge a very small set amount to make each individual post
Yes that's a marginal cost, which is what I suggested. So basically, we agree. The rest is implementation details that will depend on jurisdiction, companies, platforms and so forth.
> I don't think that adding a cost to the posts will end up paying for mental healthcare without careful regulation.
Without data or case studies to reference, I can't speculate about that and other things that are your opinions but thank you for thinking about the proposal and responding.
> Where are all the foundations helping provide moderator mental health treatment? This is a pretty widely reported issue; I'd expect to see wealthy benefactors trying to solve it, yet the problem remains unsolved.
I don't mean to sound rude but have you tried to solve the problem and start a foundation? Why is it some mysterious wealthy benefactor or other people who should solve it rather than you who cares about the problem? Why do you expect to see others and not yourself, solving it?
Raising funds from wealthy people for causes is much easier than people imagine.
But this would necessarily block out the poorest voices. While one might say that it is fine to block neonazi red necks, there are other poor people out their voicing valid claims.
> The images and videos including necrophilia, bestiality and self-harm caused some moderators to faint, vomit, scream and run away from their desks, the filings allege.
You might have heard the saying, common in policy and mechanism design: "Show me the incentives, and I'll show you the outcome."
If you want to reduce spam, you increase the marginal cost of posting spam until it stops. In general if you introduce a small cost to any activity or service, the mere existence of the cost is often a sufficient disincentive to misuse.
But, you can think through implications for yourself, no? You don't need me to explain how to think about cause and effect? You can, say, think about examples in real life, or in your own town or building, where a service is free to use compared to one that has a small fee attached, and look at who uses what and how.
It should cost some amount of money to post anything online on any social media platform: pay to post a tweet, article, image, comment, message, reply.
(Incidentally, crypto social networks have this by default simply due to constraints in how blockchains work.)