The problem with this idea (aside from the gratuitous use of bitcoin, and the pricing out of people who can't pay) is that punishing people isn't the hard part of online moderation. For most simple cases, a system of warnings, temporary suspensions, and permabans mostly works. Yes, you've got to deal with sockpuppets, but my understanding is that they're usually fairly easy to recognise, and the extreme cases are usually rare and notorious figures. Nothing in this process is so complex that it requires a stake system to be fixed.

No, the problem is that determining bad behaviour (or at least, doing so in a way that is fair, just, and broadly accepted) is very difficult. There's no objective naughtiness meter that detects when someone's behaving in bad faith. There's no set of words that always, regardless of context indicate undue rudeness. There's no perfect demarcation between just asking questions, and harassment.

So basically any system is going to have to make a series of judgement calls, and those judgement calls are going to be specific to the context of the person (or people) making the judgement. And most importantly: other people will disagree with that judgement call! Not least the poor victim of your unfair banhammer (or the poor victim of the abusive spammer you choose to do nothing about).

Money's not going to fix that problem - if anything, it'll make it worse when people have more on the line.

It's not trying to solve the problem of assessing whether rules were broken, it's attempting to add friction to bad behaviour. If you get banned and lose your deposit, it's going to cost you another $N to try again. Right now there's practically zero cost to signing up to a website again with a new account to continue the bad behaviour. At least this way, only the richest and most determined trolls will continue to have an impact on services.

Saylor discussed this mainly in the solution of DDOS mitigation. Perhaps he didn't intend for it to extend as far as content moderation and I'm taking his idea too far, but I think he probably did.

That may work.

Though it will also exclude people that can't afford it. I'm thinking of poor people and children. The kind of people who use computers at the local library, because they can't afford their own device (and more expensively) the Internet connection to make it useful.

And yes, I realize that some of the above groups are also absolute shits, who I would keep off the Internet (if I could) until they decided to grow up.

It's not $10 per site, it's $10 one off. You would get the money back almost immediately if your request is not abusive. I'd be interested to know how many twitter users couldn't scrape together a once off $10.

So who's going to fund this adjudicatory system? Because lots of little $10 deposits, that are fully refundable, won't do it.

You should ask Michael Saylor to be sure how he proposes this will all work, but he did mention his company (MicroStrategy) spends $1M/year on DDOS mitigation. I expect if Lightning ends up as cheap and fast as he and others are saying it will be, it could be funded by companies as a cost of doing business in the same way they fund the electricity and hardware for serving their websites.

It's actually the sort of thing I could see CloudFlare etc. providing, so each website doesn't need to implement it themselves.

Who decides whether your request is abusive?

The site. They stake their reputation on being fair. Unfair sites will lose users. This is basically what happens already with moderation, it's just adding a small monetary penalty to bad actors to discourage them.

And then there's the perfect exit scam when you collect a bunch of abuse bonds from your users on the way out and run with the money.

That sounds like a social credit dystopia, only that the credit is actual money.

It's not linked to your identity. You can make a new wallet any time, but it'll cost another $10. In that sense it makes a pretty awful social credit system.

So rich people can be obnoxious online with impunity while poor people have to watch their every move? Sounds awesome.

We're still at the stage of trying to apply the crypto solution to more non-problems when the existing solutions would completely suffice. Crypto adds nothing to the original thought outside of shilling.

What existing solution? The point here is that no one found a good solution yet...

You're talking about a 'universal' ban system predicated on crypto. Whatever system you think is a good or bad solution, I don't think I want a universal social credit score system on the internet that lets all sites ban me on the credibility on individual sites with different moderation abilities. Whatever it is you think it solves, it creates huge problems on societal level.

> Bad behaviour (such as causing abuse) results in a penalty being applied to the collatoral.

Cool. Who decides when I visit a random website that my request was abusive?

In my experience: some security appliance installed somewhere.

So false positives cost me money?

Yes. If that happens, you shouldn't give that website your business.

Won't I only know that after it happened?

What if they deem me abusive after several months of using the website?

This is actually pretty common. Won’t be long before we see another HN post about someone getting banned from a platform with no idea why and no recourse forward.

Actually similar happened to me recently. I woke up to find my PayPal account close and no one can tell me why.

It's neither a completely crazy nor completely new idea. E.g. Metafilter charges (charged? I haven't checked in a while) a $5 one-time sign-up fee that's less for revenue generation and more intended to make it sting a bit if you're banned (which Metafilter is notoriously hesitant to do).

Problem being that Metafilter is unpopular compared to many of the free alternatives. Correlation is not causation, but there is a plausible causation mechanism here.

Somethingawful did the same. I think it probably led to it's success and also kept it going a few extra years.

That is my literal nightmare for the future of the web.

It’s brilliant for people who think Twitter bans too much to … give them a monetary reason to ban more?

Crypto won’t solve the problem only make it worse.