Our Principles

Adults can be adults.

We don't police your vocabulary. We don't flag your posts because you said a word that made a filter uncomfortable. You're a paying adult on a platform full of paying adults, and we treat you like one.

What we do moderate is behavior. There's a clear difference between someone venting about a rough day and someone targeting another person with abuse. The first is life. The second is harassment. We act on harassment, bullying, threats, doxxing, and hate speech — the behavior, not the vocabulary. Use whatever words you want. Treat people with respect. That's the line.

For content that's appropriate for adults but not for kids — whether that's language, themes, or subject matter — we have a content rating system. The default is what we call Open Porch: content that'd be fine on a neighborhood porch with all ages around. Mature content lives behind age verification and is invisible to anyone under 18. Not hidden behind a button — invisible at the technical level. Teen accounts on family plans cannot access it, period. The server won't serve it. That's not a setting. It's architecture.

Humans moderate this platform.

We use automated systems for exactly two things: catching spam and detecting child sexual abuse material through hash matching. Everything else is reviewed by a human being who reads the context, understands the nuance, and makes a judgment call.

That's expensive and slow compared to letting an AI flag and auto-remove content. We know. We chose it anyway, because automated moderation is how platforms end up banning breastfeeding photos while leaving actual harassment untouched. Algorithms can't tell the difference between a joke between friends and a threat between strangers. People can.

Our moderation team works in tiers. Straightforward reports get handled quickly by trained moderators following documented guidelines. Gray areas — and there are always gray areas — get escalated to senior staff who have the authority and the context to make harder calls. Every moderation decision can be appealed within 14 days, and appeals are reviewed by someone different than the person who made the original decision.

We also take care of the people doing this work. Content moderation is psychologically demanding. Our moderators have rotation schedules, mental health resources, and limits on time spent reviewing difficult content. We won't burn people out to keep costs down.

Your voice doesn't get turned down.

On most platforms, what you post enters a system that decides who sees it. Maybe 10% of your followers, maybe 40%, maybe none — depending on whether the algorithm thinks your content will drive engagement. Creators know this pain intimately: you build an audience of ten thousand people and your posts reach three hundred of them unless you pay to boost.

Trellis doesn't do this. Your feed is chronological. When you post something, every person who chose to connect with you sees it. When you've seen everything your connections have posted, we tell you: "You're all caught up." And we mean it. There's nothing hiding below that line, no algorithmic drip-feed designed to keep you scrolling.

We will never throttle your reach to sell you a boost. We will never suppress your posts because they don't generate enough engagement. We will never insert content from strangers into your feed because a model predicted you might click on it. Your feed is yours. It contains exactly what the people you chose to follow decided to share, in the order they shared it.

This also means we don't amplify outrage. Algorithms learn quickly that anger drives engagement, so they surface content that makes you angry. A chronological feed doesn't have that incentive. What you see is what your people posted — no editorial judgment from a machine optimizing for time-on-app.

The Porch Test.

When we're not sure whether something belongs on Trellis, we ask one question: would this be okay on a neighborhood porch?

A porch is where adults have real conversations. People share good news and bad news. They gossip a little. They disagree about politics and then borrow each other's lawnmowers. Kids run around. Someone's grandmother is there. Nobody's performing for an audience — they're just being people, together.

The porch test isn't about being sanitized or family-friendly in a corporate way. It's about the energy of a place. A porch has room for strong opinions, colorful language, and the full range of human emotion. What it doesn't have room for is someone standing up and screaming at another person, or following someone home, or making someone feel unsafe.

That's our content standard. Not a list of banned words. Not an algorithm scanning for keywords. A question about the kind of place we want this to be.

A town square, not a campaign rally.

Trellis is a community platform. You can talk about anything here, including politics. You can have strong opinions, share them passionately, and disagree with your neighbors. That's what town squares are for.

What we don't allow is organized political operations. No official campaign accounts. No political advertising. No PAC-created groups. No coordinated messaging campaigns. No voter targeting. Politicians are welcome as private citizens with personal accounts — just like everyone else. But Trellis is not a tool for political campaigns to reach voters, and we will never sell that access.

Group admins can set their own rules about political discussion within their groups. Some groups will welcome it. Some won't. That's their call, and we respect it.

We don't compromise on kids.

Every other principle on this page has nuance. This one doesn't. Child safety is absolute, and we enforce it with zero tolerance and no appeals.

Child sexual abuse material is detected through hash-matching technology, reported immediately to NCMEC, and the account is permanently banned. Credible threats involving minors are escalated to law enforcement. There is no gray area here, and we will never create one.

Beyond the hard legal lines, we've built Trellis so that children on family plans exist in a fundamentally different version of the platform. Their accounts can't access mature content at the API level. Their DMs can be restricted to approved contacts. Parents get weekly activity digests. And when a child turns 18, they're prompted to take full control of their own account. Until then, parents set the rules.

Your children's photos will never be used to train AI models. We strip EXIF data from every upload. We offer invisible watermarking and screenshot detection. We built photo protection because parents shouldn't have to choose between sharing moments with family and protecting their kids' privacy.

We show our work.

Every quarter, we publish a transparency report. How many accounts were banned. How many posts were removed. What categories of violations we saw. How many data requests we received from law enforcement, and how many we complied with. How many appeals were filed, and how many were overturned.

Most platforms publish these reports because regulators require it. We publish them because you deserve to know how we're running the place you're paying for. If we're making mistakes, the data will show it. If we're doing well, the data will show that too. Either way, you get to see it.

We don't have investors to answer to. We answer to you. These reports are how we do it.

Your data is yours. Period.

We don't sell your data. We don't share it with advertisers. We don't use it to build profiles for ad targeting. We don't train AI models on your content. We don't analyze your behavior to optimize engagement. We collect the minimum data necessary to run the platform, and we tell you exactly what that is.

You can export everything you've ever posted, every photo, every journal entry, every message — at any time. You can delete your account and we will actually delete your data, not hide it in a database for seven years "in case you come back."

We charge you $4.99 a month because that's our business model. Your subscription pays for the servers, the moderation team, the development, and the infrastructure. We don't need a second revenue stream from your personal information. The math works without selling you out.