As part of my Sample Terms of Service project I am going through the terms clause by clause. This installment covers the core of a terms of service, the User Conduct Clauses that provide the website moderators with their contractual authority to ban users for specific unwanted activity. It also looks at the EU’s DSA and how it impacts moderation for small social media and community based websites.
The third section of the Sample Terms of Service is the longest and most important, covering the various activities the Terms forbid. This is the core of any Terms of Service, because it enables moderation and protects both the user community and the website from bad actors and liability. The set of clauses governing unacceptable user activity is sometimes called “User Conduct Clauses” and this name can create confusion with a “Code of Conduct” which, while similar, serves a different purpose. User Conduct Clauses are a way to give moderators clear guidelines to remove or discipline users who misuse the website, while a Code of Conduct lays out general standards of behavior within a community and the process for responding to violations of them. The core aspect of this distinction is that a Code of Conduct must apply to all members of the community, including its organizers, while User Conduct Clauses must only apply to the users of the site, not its moderators or site owners. Additionally, User Conduct Clauses interact with and are shaped by the regulations and laws of online activity while Codes of Conduct are rarely concerned with external rules and laws.
The set of User Conduct Clauses is long so I’ve split it into two posts. This first part will be the most important, as it covers the goals of these terms and some of the key ways they function. My User Conduct Clauses begin with the most essential term for maintaining a functioning online community, one that’s necessary for any site where users can post material or interact with each other (even a user comment section). Forbidding harassment is essential to creating a space online where people interact, because if one allows harassment, your user base will quickly shrink to include only harassers and those who are willing to tolerate the harassment of others… a community that will not help a business or community grow and will harm its reputation. The first clause jumps right into defining harassment:
3. WHAT YOU CANNOT USE THIS WEBSITE FOR.
You may not use this Website to harass others, including us, other Users or anyone else. Harassment is a broad term that encompasses both legal and illegal activities, and includes any activity aimed at harming, threatening, abusing, stalking, intimidating, or maliciously embarrassing others. It also includes encouraging, coordinating, or planning to engage in harassment.
This should be straightforward: the clause that forbids harassing people, including moderators. It also includes planning harassment or calling for it so that bad actors can’t hide behind “letter of the law” style interpretations of your terms of service. To some it may also seem broad. That’s intentional because the goal of a set of User Conduct Clauses is to give the website freedom to moderate as needed, not to create a fair or general set of principles around online speech. These terms are even somewhat narrower than they could be. For example, a well written Code of Conduct for example would also forbid harassment outside the context of the specific organization, community or event, so that known harassers could be preemptively turned away. This is not the case with these Terms of Service, though one can modify and expand them or include a separate Code of Conduct. With these Model Terms of Service I’ve tried to include a base level of protections for User Conduct that will reduce liability and allow the site to function, rather than the most or least expensive interpretation.
Harassment has many potential meanings and legal definitions, such as the California Penal Code §653.2 definition that requires the victim to have a reasonable fear of their safety because of a statement that is “alarming, seriously annoying, seriously tormenting, or seriously terrorizing […and..] serves no legitimate purpose.” A set of User Conduct Clauses should never include this sort of terminology or degree of legalism. It’s important to leave as little room for questions or demands in one’s terms of service. A website is not connected to the legal system and has no authority to punish anyone beyond banning them from use … it lacks the resources to conduct itself like a court. This means that concepts of “free speech” or “right to a fair trial” are misplaced when it comes to moderation. They are inapplicable and inappropriate, and are raised only in bad faith efforts or as the product of ill-informed idealism.
Instead, terms of service should give a website the authority to remove unwanted users as easily and quickly as possible. This is even more important for small-scale social media and community based websites that often depend on volunteer moderators or otherwise lack the resources for more complex banning processes. Moderation is a psychological taxing and difficult job and in order to retain moderators, a site’s management should provide them with resources and authority to do it as efficiently as possible.
“You may not share or post the personal or private information of others on this Website without their permission. Examples of such information can include: home and employment addresses, sexuality, gender assigned at birth, government identification numbers, financial information, or medical information.”
The following term is similar to the first, an “anti-doxxing” provision, and the only notable parts of it are that again the term doesn’t have typical safe harbor provisions found in legal definitions of doxxing, such as the need for malicious intent, and that the specific set of limitations are designed to protect vulnerable individuals, especially trans people. Trans individuals are more susceptible to doxxing as publicizing their gender or name assigned at birth, even through seemingly innocuous documents related to their pre-transition identity, can be very dangerous if they live in or travel to countries where expression of their identity is criminalized or where there is frequent violence directed at them. Where for most people, someone posting a birth certificate or similar public document might be disturbing, for trans individuals it could be physically dangerous and such doxxing often leads people removing themselves from online or public life.
USER CONDUCT CLAUSES AND THE DIGITAL SERVICES ACT:
You may have heard about all sorts of duties and requirements for social media platforms under the EU’s “DSA”. These requirements exist, but almost never apply to small scale online platforms. The EU Digital Services Act (DSA) came into effect in 2023, and courts across the Union have been adapting to it ever since. This act is designed to provide internet users a set of rights and protections around advertising, illegal content, and website discipline such as user bans. The DSA was written to reign in large social media and commerce platforms like Twitter (“X”) and Amazon.com. These companies are fighting against the regulation, and some even seem to be ignoring it – daring regulators to enforce it against them. Because it is emerging and contested, the DSA shouldn’t be a serious concern for small websites who are not the regulators’ targets.
These Model Terms of Service are an example for a small scale social media website, such as a Mastodon server or hobby forum who should have minimal duties under the DSA because of their size. They are “Small or Micro-Enterprises” (SMEs) under most EU regulations, including the DSA. Under the DSA an SME is defined as having under 50 employees, under 10 million euros in passthrough per year, and under 45 million active users in the EU (the last to avoid also being a “Very Large Online Platform”). Because of their size, SMEs are subject to lesser requirements under the DSA. Larger online platforms, and especially Very Large Online Platforms have significant burdens under the DSA, SME’s, even when they are online platforms, do not.
The primary impositions for an SME online platform under the DSA are: to publicly provide a point of contact for users and regulators within the EU, to remove illegal content quickly when you receive notice of its presence, to respond to regulator requests for information, and to provide clear terms of service. The DSA also requires even SMEs to provide reasons for content removal or bans to the effected user. The level of detail and how these notices should look, while defined in the regulation, is an ongoing subject for litigation, as is what constitutes a ban under the DSA. It is clear though that any website banning a user for a terms of service violation has to give notice to the user that a specific disciplinary action has been taken against the user because of a violation, illegal act, or reported violation/illegal act, and that a specific post or activity is the source of the violation.
The language of the DSA also includes discussion of remedies or appeals for banned users, but these parts of the law don’t apply to SME’s per its Article 19. This means that a very simple notice resembling this should be sufficient:
“Your account, [USER ACCOUNT], has been terminated and you are banned from using [WEBSITE NAME] for [“A violation of our Terms of Service Section [X]”, “posting illegal content”, etc]. Our moderation team reached this decision after [optional: “illegal/violating content was reported to our moderation team and]” they conducted a review [optional: “,including review by automated moderation tools,”]. The material that provided the basis for this ban was posted by [USER ACCOUNT], and includes, but may not be limited to, that of [DATE OF VIOLATION].”
Whatever notice one provides, banning notices shouldn’t offer a great deal of information, and should never be publicized. While the DSA requires some information, it’s good to provide as little as possible while still acting in a way that can be reasonably be interpreted as an effort to comply with the regulation. Most moderation is for clear violations or obviously illegal posts, and detailing one’s decision making process or rationale only invites the banned user or their partisans to contest or complain while giving them a larger amount of information to allege the banning occurred in violation of some law or right. The goal, as always with moderation, is to create a safer space for you users while using the minimum amount of your limited resources.
Leave a Reply