Hacker News Reader

Comment by pjc50 | Hacker News Reader

Unfortunately as we've seen from email, if there's no system for silencing posters you get drowned in spam. The history evolved as follows, and this is how I'd expect it to go for any "decentralized" comms system:

- various private blocklists pop up identifying alleged spammers. These have a false positive rate, but they're useful enough to become popular

- deliverability starts to become an issue; some people can't deliver because they're on a blocklist

- people start to notice that a big, popular service has both good antispam and good deliverability

- after a while, everyone's back on big, popular services which occasionally false-positive ban people.

View on HN

That's why there should be laws requiring large services to provide due process for users who have invested significant time and energy into the platform.

Its just common sense that a spam bot should be nuked on sight but a legitimate user should get a fair hearing before you ban them. Ideally the social media platforms would do something akin to Wikipedia's arbitration committee (basically the Supreme Court of Wikipedia) but scalable by hiring a suitably diverse pool of "unskilled" people for a $15 an hour or so remote job where they decide if a longtime legitimate user should be banned. I think this system would do a much better job of avoiding false positives. A properly functioning spam bot detector shouldn't be detecting any legitimate longtime users whatsoever.

Banning email addresses without sufficient notice to allow accounts tied to that email to be moved to another email should be illegal. You should also have a legal right to access your digital purchases even if your account is banned. There's nothing wrong with not wanting a customer any longer but you shouldn't be able to cause irreparable harm to your customer on their way out the door and there should be safeguards in place to ensure you don't fire any customers by mistake.

Email is a different beast though, as it is set up to replicate snail mail. You need to be able to receive mail from anyone at any time.

For private, personal use at least, you don’t need this for a service that competes with discord. I’m not allowing random strangers into my chats, having an allowlist basically makes spam a non issue.

Now, that doesn’t fix the problem for public channels, but spam/trolling is always an issue on those anyways, and frankly, I don’t understand how people can use discord for that purpose in the first place.

If the filter is in every participant’s individual hands, then:

1. if a person I am interesting in delivery to blocks me by mistake - I have an option to ask him to unblock me by other communication means.

2. if my opponent in a public discussion blocks me, other people can still see my arguments.

Both these benefits are absent in a centralized blocking system, no? And I do not see any drawbacks, are there some?

However, email is easily portable, you can download your messages, and even own your domain, so you can switch email providers without changing emails.

Spam doesn't have to managed by an opinionated central banlist. For email the solution is aliases like anonaddy or SimpleLogin.

Reply on news.ycombinator.com