Gradients of intimacy and creating shelter: An interview with Erin Kissane

by Marisa Morby

A clown fish moves tentatively forward, peeking, as much as a fish can peek, out from behind the safety of its hiding spot in the coral. It has a decision to make. Does it stay here and watch the world go by? Does it venture out among the rest of the fish in the reef? There’s a lot to assess: potential threats, potential food, existing safety.

We open the doors of our homes to let people in and allow ourselves the space to go out into the world. We open our windows. We look outside. We, too, are assessing our environment for our next decision. Do we stay in or do we venture out? Will I invite this person into my home or decline them entry?

These are the questions every animal asks itself when deciding how it wants to engage with the world around it, and the answers to these questions are how communities form. Every animal must have some type of shelter in order to feel safe and in order to thrive. Or, as Erin Kissane says, there are “gradients of intimacy” and “gradients of privacy.” And while we can quickly grasp these real world examples of shelter and safety, they also need to be applied to our communities, both online and offline. How communities form, work, or don’t work has been a focus of Erin’s work throughout her career.

I spoke with Erin about how nature impacts her thinking on community and her hopes for the future of our online lives.

Community work that helped the world

Erin Kissane’s work is likely known to you, even if you’re unaware of the people behind it. Starting in 2020, she volunteered with a team of researchers and scientists to start The COVID Tracking Project, which was used by government and news outlets as an up to date and reliable source for national Covid data.

During that project, Kissane said she “wanted to become a good science communicator and had to do it quickly.” The team, including Jessica Malaty Rivera, a vaccine educator and epidemiologist, provided guidance to gut-check how they communicated this information. Erin says the goal was to find a balance in the sharing of information. “How do we explain this in a way that’s intelligible but has enough integrity that if [people] look further they won’t feel that we clarified through misrepresentation?”

Before and beyond that important work, Erin has been a constant voice discussing how online communities work, or, in many cases, why they don’t. Her goal is to work “toward better networks for human sociability and collective survival.”

Much of her work examines why online communities break down, and one of the core reasons seems to be a lack of guaranteed shelter—a place to protect yourself, and a lack of gradients of intimacy—getting to choose who comes in and having permissions or rules on how people can engage and who they can engage with.

Our lives are a blend of online and offline spaces

Our online spaces today often fail our social needs. We see that through the dysfunction in Facebook community groups, X (formerly Twitter). We also see this in smaller communities that seem to have trouble really taking off. On average, a person spends 5 and 7 hours online each day, and about half of that time is with social media consumption. This constant consumption makes us ultimately feel lonelier and more disconnected from the people around us.

But there are still opportunities where we have hope that these spaces can improve and be better tomorrow than they are today.

Bluesky is a social network that uses the ATprotocol (ATproto) ecosystem. ATproto is a decentralized network protocol. Imagine that you are a boat traveling down a river. Along the river you can make any number of stops into different ports. Each port is different, but you take your boat with you wherever you go. That’s a decentralized network. You can take your identity with you to any place on the internet. In a mycelial network, the mycelium blanket the forest floor. Each strand is connected to a different tree and a different plant. There isn’t one, centralized hub, it’s many individual pieces that form a large network.

Emerging technologies may enable better community structures

I recently saw Erin give a talk at ATmosphereConf 2026 where she used bull kelp as an analogy to talk through how we can investigate and improve online communities. Holdfasts of the kelp anchor it to its substrate and in so doing create areas of shelter for all types of tiny creatures.

We talked about how nature impacts her thinking on community, and she said, “I rely on figures and information from the natural world because nature has solved for so many things. It seems so goofy to me to leave that stuff just lying on the floor and not make use of it intentionally.” She particularly likes to investigate how living things negotiate with each other and do or don’t get along.

In our conversation, Kissane was optimistic about developments in the ATprotocol (ATproto) ecosystem. Permissioned data systems could enable more nuanced privacy controls and community boundaries. That means you get to control your “gradients of intimacy.” What this means is that, as a group, you can decide who comes in and who doesn’t, and as an individual you can determine who gets access to you and who can see what you do. That individual control is the shelter that protects you. This structure can serve as the holdfast that protects you.

In thinking about the future of community, she said that hearing people talk in the Science Track at the ATmosphere Conf really inspired her about the future of online community. “So much of what I’d hoped to see within a year were things that people were actually building but for their labs or for their colleagues.” With permissioned data, in the ATproto ecosystems, we’ll have the ability to create shelters and those gradients of privacy and intimacy.

Kissane says we “can’t thrive in one big flat room.” People need shelter to live safely, but if you’re completely closed off, that also doesn’t work. We have windows and doors to our houses so that we can look out and others can come in. In online communities, we have to give people the benefits of a big world but the additional necessary benefits of privacy and intimacy.

And this is with the knowledge that shelter for communities won’t only produce pro-social groups. As Erin notes, “Bad people use buildings, too,” but it’s necessary to have these shelters and protections because “we all do better if we have some walls and doors.”

Private spaces strengthen communities

These private spaces create safety, allow for individualism, creativity, freedom of expression, and allow for agency in how people interact with those around them, both in their chosen communities and outside of them. Private spaces also allow us to section each other off, which is better than it sounds. By being able to have closed communities, we create stronger bonds through shared experience.

Ultimately, online communities should be designed more like natural habitats, with varying levels of privacy, intimacy, and protection, because people need sheltered spaces as much online as they do in the physical world. Kissane stressed that we must make sure that our online communities become better because they are real. “[Our online interactions] will come out of the computer and get us. So we have to take it seriously.”

Resources