@pch_xyz @waifu CC @p @PCH_XYZ

EN // TRANSLATION (Computer translated and Corrected by Nekobit):

Free speech online…

Since the development of the Internet and social networking sites, free speech is largely influenced in US technology companies. It is necessary to understand the background of the First Amendment to the US Constitution and Section 230 of the Communications Decency Act before discussing freedom of speech in the Fediverse (ED note: Federated Universe is the original translation).

When the First Amendment of the US Constitution was enacted, the situation was a binary opposition between the government and the citizens. The purpose was to limit the power of the government to protect citizens’ freedoms, including freedom of speech. However, the channel through which most people speak today is the platform controlled by Internet giants such as Facebook, Google and Twitter. Although the First Amendment prohibits the government from censoring private websites, it also allows private companies to censor user-generated content and take certain political positions, as this is freedom of speech for private companies. The First Amendment does not apply to relationships between private companies and individual users.

The role of Section 230 of the Communications Decency Act In a nutshell, ISPs are not responsible for speech posted by users of their platforms, and as long as ISPs act in good faith, they will not be responsible for removing or censoring third parties. Civil liability is up to them for the content provided. During the 2020 U.S. presidential election, Trump’s controversial remarks were not favored by the liberal media. Due to the subsequent attack on the U.S. Capitol in 2021, Trump was successively banned from Facebook and Twitter, and related issues were censored as well. Trump therefore believes that Section 230 of the Communications Decency Act, which allows Internet giants to selectively censor speech, should be repealed; while Democrats believe that Section 230 allows Internet giants to evade the responsibility of regulation, and they also want to repeal Section 230. Facebook, Twitter and other companies are opposed to the repeal, because without the umbrella of Section 230, social networking companies will strictly prohibit any controversial issues to users’ posts, so as not to cause troubl. This is detrimental to the development of free speech. Article 230 will be abolished or amended in the future, and it will have a significant impact on the development of the Internet ecology. Not only users of mainstream social networks are affect by this, but users of the Fediverse should pay attention to the development of this matter as well.

State of the Fediverse

In Fediverse, to register to join an instance, you must agree to the terms of service (Terms of Service) of the instance. The terms of service generally include that the content of the post must not violate the laws of the location (host/webmaster/NIC) and other unwanted content. The content that appears in this instance can be classified as harmful to most people, such as false information, advertisements and harassment; culturally inappropriate and may cause discomfort, such as bloody violence, nudity, and the level of permission for Lolita pictures; reflections of certain values, such as gender equality, freedom of speech, etc. According to my observation, instances that tend to favor progressive values ​​tend to block more users and instances, have more banned speech, and have less chance of bickering between users and admins; instances that favor free speech tend to block fewer users. There will often be disputes between users and admins including disputes over freedom of speech itself.

Moderation refers to the avoidance of excessive behavior or political opinion, and here I will call it moderation. Most of the small and medium-sized instances are managed by the admin. In the large-scale instances, there will be a full-time moderator, which is equivalent to the board owner in the BBS era, with the power to control speech, such as deleting statuses, banning or even freezing users (ED note: Shadowbanning is possible in Misskey). If the above behaviors are performed on the instance where you are located, the content will be deleted. Some instances will not accept the deletion activity; if it is a user of the remote instance, the operation will only occurs on the other instance.

This means that ordinary users can operate on bad behavior, including

• Mute the discussion thread, that is, it will not appear in notifications but can still be viewed. It is often used in the case of being tagged in a relatively lively discussion, such as a large thread (hellthread) with multiple tags.
• Mute a user
• Block a user
• Report a user
• Instance level blocking

Reports can optionally be sent to the other instance administrator, and possibly to the local administrator as well. Due to policy differences in instances, user reports may be processed and responded to or ignored. If it is spam, harassment, or clear violations, you should first report the post to give the other party’s administrator a chance to deal with it, not jump straight to the block; it is impossible for someone to monitor their own instance 24 hours a day.

The means that admin / moderator can operate on remote users, including:

• Silence - posts will not federate: it is only on the homepage of users who are concerned
• Freeze (suspend) - the user can no longer create statuses and cannot interact with the user
• Delete the status - local only (ED: It may federate though if enabled?)
• Suspending the instance entirely (ED: “Report this instance” I believe was not intended by translator)

The admin / moderator can perform instance-level operations, such as not displaying all media from the instance, or muting or even blocking the instance. Mastodon’s instance-level operations are as follows:

• Media removal - Media files from these servers will not be processed or stored, and no thumbnails will be displayed, requiring manual click-through to the original file
• Silent servers (silenced) - Posts from these servers will be hidden in public timelines and conversations, and no notifications will be generated from their users interactions, unless you are following them
• Suspended servers (suspended) - No data from these servers will be processed, stored or exchanged, making any interaction or communication with users from these servers impossible

Pleroma’s instance-level operations are at least the following:

• Reject (Reject) - This instance will not receive messages from these instances
• Timeline removal (Removal from “Known Network” Timeline) - Remove statuses from these instances from the known network
• Media force-set as sensitive - This instance forces the post media of these instances to be set as sensitive

Misskey’s instance-level operations are only one of blocking instances.

If a remote user’s posts and their instance’s policies are clearly not in line with my values, then most of the users on that instance should be too, and blocking the entire instance seems to save a lot of unnecessary hassle. But is it really so? In a study by Hassan et al. (2021), surveying 1298 Pleroma instances, 110,000 users, 2.45 million posts, and 46 different policies, they found that rejecting an instance made 86.2 % of users and 88.5% of posts are rejected (ED: “affected by pond fish” lol); and there is a weak correlation between instance size and rejection by other instances. For example, if a user in one instance posts a misogynistic post that is uncomfortable and tiring to read, there may be more than 80% of other users in this instance who consider the post “harmful”, blocking this instance directly cuts off the chance to contact the “harmful user”. Of course, from the user’s point of view, it is understandable to use it as a signal to filter the source of the information, but at the instance management level, I tend to leave it to the user to decide. So, I also don’t think FediBlock is a good idea, why should I use a list made by someone else based on the principle of not knowing what to decide for my instance users which content they should not see on the instance? However, as instances get larger, it becomes increasingly difficult to continue to follow this principle.

Revolver

For the admin, not only does it cost money and time to maintain the instance, but moderation is also an additional problem, and the “people” problem is often more difficult. There is a project, the solution to the problem of moderation is to not use moderation; since the admin is a chore, then there is no need to have an admin at all. As the proverb often goes, “Instead of trying to solve the problem, it is better to make the problem not a problem”.

Author: waifu

The goal of Revolver is:

• No admin is required on the architecture
• Cannot be removed by the domain name provider
• Not subject to centralized DNS, LetsEncrypt and other institutions
• Federation through the IPFS protocol
• Uses the LitePub protocol and is compatible with other applications in the Fediverse
• Lightweight, so lightweight that it can be ran on mobile phones
• Spam-resistant
• Support for end-to-end encrypted private messages

The developer of the Revolver project is also the webmaster of freespeechextremist.com. What he wants to solve is the problem of censorship, decentralization with P2P, and the promotion of freedom of speech. The issue of false information and incitement is not within the scope of this project’s consideration, nor should it be considered by this project. Everyone should be responsible for their own speech, and everyone should build up the ability to read information, instead of relying on a moderator or someone to judge for themselves.

Reference (Translated links offered by Google)

• The law of the jungle for freedom of speech (Part 1)
• Should social media be responsible for user speech? Reviewing the 🇺🇸 Section 230 of the U.S. Communications Decency Act that sparked the repeal Controversy
• Hassan, AI, Raman, A., Castro, I., Zia, H, Cristofaro, E., Sastry, N., & Tyson, G. ( 2021). Exploring Content Moderation in the Decentralised Web: The Pleroma Case. Proceedings of the 17th International Conference on emerging Networking Experiments and Technologies (ACM CoNext 2021) https://arxiv.org/abs/2110.13500
• g0v.social server Rules (requires login to see list) (ED: Not a useful source, mastodon.social or rage.love is a better example)
• Example rules for shitposter.club

:dogwalk: and im done.