Thread by @VitalikButerin

Temporarily breaking out of my Twitter-minimization for a short thread on issues around free speech and the mass deplatformings of the last week.

Obviously the riots were terrible, people still supporting DT are crazy, so moving on to some things that have not yet been said...

1. Every "two buttons" meme has a not-necessarily-equal and opposite two buttons meme.

The two-buttons meme most people are talking about is "private corporations can do what they want" vs "censorship is bad" on the right. But there's also a challenge on the left...

The "outside view" explanation is simple. When factions fight, they typically disagree on principles and have divergent interests. But eventually by random chance, an issue appears where taking some action X satisfies the principles of side A and interests of side B.

Side B wants to support X but feels discomfort about violating its principles (interests usually beat principles), and side A wants to oppose X but similarly feels discomfort. I claim that this pattern is very common in tribal conflicts.

(Once again I'm not claiming equivalence)

But back to this situation, the fact that so many people who would normally never support such corporate power are now cheering tech CEOs running roughshod over democratically elected officials deserves some introspection.

I'm not saying they're wrong... just some introspection.

2. For me, Jack banning Trump violates one principle of liberal-democraticness (Jack was not elected by a vote), but the event was a huge win for another principle of liberal-democraticness: separation of powers.

There's a lot of benefit to be gained from having control over social media and control over the vast levers of governments be under somewhat independent hands. Two centers of power ensures that when one breaks the other can pick up the slack.

3. One important point: the set of Twitter users is not the USA. Using @glenweyl's lingo, the "natural polity" of Twitter does not map well to any single country. So any US-centric governance model for Twitter is likely to have many problems.

radicalxchange.org/kiosk/blog/201…

4. But at the same time, there are huge flaws in the Jacktatorship of today. A big one is: it's not "lawful" enough. The @TwitterSafety high court's opinion blog.twitter.com/en_us/topics/c… is honestly poorly argued and reads like after-the-fact justification.

There's plenty of arbitrariness and inconsistency in Twitter moderation. This tweet is still up:

https://twitter.com/khamenei_ir/status/1003332853525110784

And this one, for that matter (no, I'm not claiming any "equivalences"):

https://twitter.com/Slate/status/1268415955937513473

There's no clear system of rules, appeals process, etc...

5. I see two interesting paths for @jack, radical in different directions. One path is to make Twitter itself democratic and lawful: create a structured process with appeals, and a moderation panel at the top based in Switzerland (NOT USA!). Allow users to vote on panel members.

The other is to make Twitter more email-like, a protocol that one can have many interfaces to. This punts the problem of moderation to the interface layer, and there could be many different interfaces.

e*2. A dumb math joke intermission.

What's the difference between US politics and mathematics?

In mathematics, Q is a symbol of rationality.

6. I think a global conversation medium is a valuable thing to have. I worry that "default" political philosophy (both liberal-democratic and otherwise) will push social media governance in a very state-centric direction, which risks rupturing any semblance of such a thing.

Crypto may actually have a role to play here: it's a non-state-centric ecosystem that has had to grapple with tough political philosophy questions already. But to achieve its potential it needs to get its head out of its sand and admit that certain challenges exist.

I credit @VladZamfir for trying to strike out a productive path in this direction, even if I disagree on some of the details of his proposals specifically.

7. The attempts at bringing down Parler are very worrying. Apple, Google, AWS are much more like "common infrastructure providers" than a social media site is. Parler has a right to exist, full stop.

8. Twitter banning @sci_hub is sad. You should definitely NOT click this link to SciHub's telegram bot telegram.me/scihubot and violate the publishing cartels' god-given right to massive profits.

9. There are ways to improve the quality of online discourse without content-specific restrictions (which are hard to make credibly neutral: nakamoto.com/credible-neutr…)

It's a well-known phenomenon that people just act *worse* on the internet than they do in person; it even has a name: Greater Internet Fuckwad Theory. knowyourmeme.com/memes/greater-…

But clearly GIFT is not so simple. Plenty of online interactions are civil, and plenty of non-anonymous interactions are not. Are there ways to tweak the ways that we interact with the internet to make us more pro-social? I think so!

I credit @jack for experimenting here already: tweets with limits on who can reply to them were a great idea. The "quote-tweet-by-default" experiment was a nice try. I wonder what else we can do. Or maybe we just need less tweeting and more long-form (as I am doing in 2021).

The crypto space can also participate here. @peepethApp is a great experiment. Another cool experiment would be to try decentralized forums moderated by Kleros. Kleros has already shown its sanity by correctly deciding that Biden has won the election: klerosboard.com/dispute/?id=532

Another is to try some cool ideas at combining anonymity with persistent reputations, eg. see here: ethresear.ch/t/anonymous-re…

Basically get the benefits of privacy while still keeping a way to hold people accountable, in zero knowledge.

This is a dangerous moment in a lot of ways, but there's a lot of work to be done!