Hacker News Reader

> how do you balance speech, censorship and moderation?

Decentralization does help a lot here. The X, Y and Z server owners don’t have to agree or coordinate. The fact that they will disagree on their policies is part of the answer. There won’t be a one-size-fits-all policy. Different servers will try different things and ideas will be shared about what works and what doesn’t.

Z server can choose to allow saying mean things and X and Y server can block and ignore posts from Z.

Some servers may have a political bias one way or the other and individuals can choose which to join and pay attention to.

Sounds like it'd just be another Mastodon -- and heck, likely get federated (or in most cases, not federated...the whole issue with Mastodon) in with the rest.

Back in the Usenet days we had killfiles to do user based filtering -- everything would get sent out, and users would decide what sort of content they didn't want to see. Obviously it was harder back then to just spin up a new identity, so it's not simply a matter of taking the old tech and slapping it on the new, but the approach avoided the splinter effect we get trying to decide for users what will and won't be available to them (whether they like it or not). Snowden actually tweeted a week or two about how trivial this would be to implement on Twitter if they ever actually wanted to (with a working one liner to prove it). The issue thus far has usually been that advertising funded social media THRIVES on the controversy, and only moderates what it needs to to keep advertisers happy. They don't WANT you to make it a pleasant experience, because controversy and fights get more ad views.

Perhaps at some point bandwidth becomes cheap enough that we can cut back on how much of a role advertisers play in steering the ship. Until then I don't see a great solution to any of this.

> It seems hard for me to believe that deliberately choosing the less efficient solution is a good idea for a start up.

Many crypto scam startups have made bank with this less efficient strategy. Have you seen all the companies selling banks on internal proof of work blockchains and stuff?

> Centralization came out of pragmatism

Sort of? Centralization is more profitable and easier than building new open internet protocols and formats.

Those are certainly more pragmatic than ideological concerns, which is the point GP was making.

And I agree. I just think monetization is the larger factor.

Reliable servers ain't free, at least that's what I think one of the larger issues is.

No centralization, no server.

"This, and other things CEOs say while smoking crack."

Computer code still runs on CPUs and not the aether. If its accepting connections from other computers it is the server in the client/server model. If it's on your desktop, it just typically means it doesn't have ECC memory and a battery backup dropping the reliability even further. Decentralized architectures have all kinds of fun and complicated failure modes, and especially for consumer use, high speed and low latency typically are not their strong points.

How much does Tim Berners Lee pay for hosting the world wide web?

High speed and low latency are overvalued at the moment, because they are so important to the functioning of the capitalist internet.

Usenet, IRC, email - discussion, conversation, mail - all worked fine for decades on the old, low speed internet. What exactly are we getting in exchange for all this high-availability nonsense anyways?

Usenet was pretty low latency at the time. But that time is long past. This is the problem you have where you look at the past and said "All these things worked great then", but the moment your users use something with higher bandwidth and lower latency they are leaving your site. If everything is slow that is the expectation, if you're slow, you're losing.

So yea, we're the old men yelling at clouds, and the kids have gone on to a ADHD fueled media filled flashing light fest and they don't plan on coming back.

Usenet also enjoyed ISP's typically being willing to host them as part of providing basic Internet service at the time. Good luck asking Comcast and company to host the latest iteration of social media this decade.

I think that the desirable long term outcome involves the vast majority of people embracing the idea that we should pay up front for information services, rather than accepting "free" things from capitalists. TANSTAAFL.

The good part is that operating a usenet server with no alt.binaries.* content is extremely low demand; it was designed to run on the 70s internet, so...

If I'm not mistaken, though, irc still works pretty well, and still provides what it always has, and could still function as the live chat aspect of a small community's technology stack. Moreover, in the event that the high-availability internet goes away, it will probably still work on a 90s-tech internet. If your packets are just conversational text, you really don't need a massive warp-speed pipe to make it work.

Likewise Usenet - yes, we abused it to get our alt.binaries (read: boobies), but that was strictly desperation. It still works fine on the fast internet, but again, will continue to work on a 90s internet, in the event that we lose the fast one.

Fast itself is really what's warping our perceptions of what is normal and acceptable. When was the last time you watched an old Connery Bond movie? My grandpa took me to see every Bond movie that existed at the time back in the 80s, the downtown cineplex threw a festival. I remember enjoying all of them at the time, nothing wrong with em.

I threw on Dr No a couple months ago and I swear, I could not get past about 30 minutes, the pacing was hellishly slow, to the point that wanted to climb into the screen and lob a bomb at his car myself out of sheer boredom. Had a similar experience trying to watch Three Days of the Condor.

Interestingly, I decided to rewatch the Lone Wolf and Cub movies after watching the Mandalorian, and they were just as fun as I remember. Some slow scenes, but good pacing with a decent body count:frame ratio.

I don't think we're so much yelling at clouds as wishing the clouds would slow down long enough for us to yell at em. I have my doubts that the pace can be maintained indefinitely, and they are not trying to bring it under control at all. What I expect next is a painful correction in which the laws of nature will come into play in a big way. Whee!

Not sure I get the problem. Yes an expensive server in expensive data center has more CPU, Ram, disk I/O, and bandwidth than a normal desktop, laptop, or phone.

But say you track 1000 peers, find 50 online, and send a 280 character tweet to ten of them. Sure it's going to take seconds for that tweet to get to whatever fraction of millions of people that are interested. Sure your client should ask the 40 peers if they saw the message to make sure it went out.

The beautiful thing is that even on a smart phone paying 10x the bandwidth of a tweet is still effectively zero.

If the goal was a commodities trading platform, sure distributed wouldn't make sense. But receiving tweets of interest send in the last few minutes seems rather insensitive to a peers bandwidth, latency, CPU, or ram.

Moxie's position on this is pretty strong, and on the side of platforms(centralization) vs standards/protocols(de-centralization), for product reasons.

blog: https://signal.org/blog/the-ecosystem-is-moving/

talk: https://www.youtube.com/watch?v=Nj3YFprqAr8

As someone who is frequently rated limited on HN and occasionally scolded by HNs beloved religion-apologist moderator, I find decentralization to be not only pragmatic but truly essential for the next generation of social platforms.

If the speech is illegal in whatever jurisdictions apply, then remove it; otherwise, it should stay. Yes, even trolling and slurs. If a platform emulating the public square goes beyond what is legal, it ceases to be a neutral platform and inevitably becomes a vehicle for propaganda. I realize that means propaganda could be spread by the users, but I'd rather than than the platform and users spreading it.

The real question here is defining "public square" legally.

Spam is 100% legal in all jurisdictions, so your legal free speech plan sounds like a total hellscape.

Probably not remotely practical, but I wonder what an online equivalent to the "do not call" list would look like.

After it went through congress? Something like this: https://en.wikipedia.org/wiki/CAN-SPAM_Act_of_2003

Domestic network protocols make SPAM, sim jacking (sim swapping, sim hacking), malware and ransomeware limited to TCP/IP (aka the Internet).

Mobile phones utilize IP networks (VoLTE or voice over data networks), so dialtone 2.0 becomes an easy option with domestic network protocols as well.

ESIM is about to change dynamics as well, so any use of a phone number/SMS/2FA for identity verification is rendered null with the current system in place.

https://consumer.ftc.gov/consumer-alerts/2019/10/sim-swap-sc...

Domestic networking protocols and dialtone 2.0 easily address most ransomware, malware, SPAM, etc much easier than trying to police 3rd world countries.

That severe limitation of regulating 3rd world countries is also the same argument with climate change. The can continue to regulate their own soverign borders.

There are definitely laws against spam

Are there laws against spam that aren't narrowly defined as email or phone spam? I'm not aware of laws against forum/wiki/social media spam. The US postal service is largely spam-funded, as is Canada Post.

> are definitely laws against spam

Almost none of them in competent jurisdictions even attempt to define spam with an eye towards balancing against free speech.

What laws are there against posting twitter spam?

No it certainly is not.

Here's the Canadian jurisdiction against SPAM: https://laws-lois.justice.gc.ca/eng/acts/E-1.6/index.html

> An Act to promote the efficiency and adaptability of the Canadian economy by regulating certain activities that discourage reliance on electronic means of carrying out commercial activities, and to amend the Canadian Radio-television and Telecommunications Commission Act, the Competition Act, the Personal Information Protection and Electronic Documents Act and the Telecommunications Act (S.C. 2010, c. 23)

> Unsolicited electronic messages 6 (1) It is prohibited to send or cause or permit to be sent to an electronic address a commercial electronic message unless

> (a) the person to whom the message is sent has consented to receiving it, whether the consent is express or implied; and

> (b) the message complies with subsection (2).

And here's the European jurisdiction against SPAM: https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A...

> Directive 2002/58/EC of the European Parliament and of the Council of 12 July 2002 concerning the processing of personal data and the protection of privacy in the electronic communications sector (Directive on privacy and electronic communications)

> Unsolicited communications 1. The use of automated calling systems without human intervention (automatic calling machines), facsimile machines (fax) or electronic mail for the purposes of direct marketing may only be allowed in respect of subscribers who have given their prior consent.

This wouldn't apply to spam you would read in a public forum. This act is about unwanted messages sent to you. Parent poster is still correct

Bleh, I find your position is boring because it's so easily shown to be unworkable.

Imagine there have a platform run as you suggest and I don't like what you're saying on it: I flood you and anyone that supports you with not-quite-illegal abuse, misinformation, non-sense, spam, false flag support from people with deplorable views. Copy-cats of you that take your positions but weakly argued or with a touch of added genocide. At the end of the day almost no one will see what you wrote, and most of those that do will think you're probably a kook with deplorable views (since that's what you're surrounded with).

Censorship that works by outright blocking speech is a largely outmoded concept in a world where communications networks connect every one to everyone else. In such a world flooding with noise that undermines the conversations people want to have and undermines their willingness to attempt the conversations is a more effective tool to censor communications. Outright blocking isn't very effective, unless it's done secretly, because people will route around it and plugging leaks is extremely hard (requiring essentially AI complete filtering).

As such, censorship is now essentially its own opposite.

This means there really are no magical silver bullets. We ought to stop pretending that private spaces are public squares though-- though this is a problem offline too: with eminent domain being used to seize land and create giant privately owned open air shopping malls in many places... but at least offline doesn't have the same scaling properties that make it so easy to censor by inundating with noise.

There are quite a few websites with little or no moderation working just fine, like 4chan.

But I honestly don't understand this whole argument. Twitter is a social network, you just need to only receive messages from your network or "friends of a friend".

You can't have both a global network and get offended given the multitude of opinions and laws in the world.

The problem seems to be less that people are offended by something, but that some people don't want other people to even be exposed to some thoughts. Some paternalistic higher-than-though attitude.

> There are quite a few websites with little or no moderation working just fine, like 4chan.

Even taking at face value what you just said, it doesn't logically follow that if having one or more places like 4chan is just fine, having online venues be like 4chan would be just fine as well.

> There are quite a few websites with little or no moderation working just fine, like 4chan.

4chan is actually heavily moderated. It may just not be legible to you because they're not removing the stuff you're expecting but different stuff instead.

There are factors like topicality, being small sized, or being highly ephemeral which make moderation easier or lower amounts more viable-- but twitter is pretty weak in those respects, sadly-- I think that's how it manages to both have a reputation for being toxic and overly moderated.

Oh 4chan? Great! Everyone is going to love that.

> There are quite a few websites with little or no moderation working just fine, like 4chan.

Working fine at doing what? Emulating a thousand typing monkeys? Arguing that 4chan has some form of working discourse and rherefore everything should be like it is like arguing there is still some message at the end in the childrens game "silent mail" and therefore it should replace all methods of communication.

Don't get me wrong, there probably even is some sort of use in the existence of something like 4chan (I bet some sociology doctorate has written about the function of the platform for certain demographic parts of the population for example).

But there is probably a reason why you are discussion this topic here on HN instead of over there. And if I guess, this reason has all to do with you being able to have better, cleverer and more informed discussion here, than over there and is not at all hindered by the fact that the HN moderation would cut your voice off if you just pressed the right (wrong) buttons.

Some form of moderation is always better for discussion than none at all, unless you are making some sort of social experiment.

Elon?

Laws aren't static. Also, when someone thinks they've been violated, it takes a lot of time and humans to settle it.

It's already hard enough to keep up with the laws in software.