Introduction: Is Social Media Just Digital Grooming?

How do you use social media in your daily life?

Liking your friends' posts, sharing funny videos, tweeting about your day. These behaviors are sometimes compared to "grooming" in primates—when monkeys pick bugs off each other's fur. It feels good for both parties and reinforces social bonds. Much of social media functions as this kind of closed, intimate space.

Or perhaps it's a place where you passively receive broadcasts from celebrities and influencers. Information flows from top to bottom, like an extension of television.

But couldn't social media be something more?

In this article, I'll introduce an attempt to understand "ATmosphere"—the space created by the decentralized social network Bluesky and its underlying AT Protocol—from a new academic perspective.

What Makes Bluesky Different?

Traditional social networks like Twitter (now X), Instagram, and TikTok are all managed by a single company's servers. This is called "centralized" architecture. Your account and all your data belong to that company. If they change the rules, you have to follow them. If your account gets suspended, you lose everything.

Bluesky is different. Using a technology called the "AT Protocol," it achieves a "distributed" architecture.

Three key points:

  • Your identity belongs to you: Through DID (Decentralized Identifier) technology, your account doesn't depend on any single company

  • You can take your data with you: Even if you move to a different server, your followers and data come with you

  • You can customize what you see: You choose which posts to see (Feeds) and how content is labeled (Labelers)

This entire ecosystem—not just the Bluesky app, but all services and communities using the AT Protocol—is called "ATmosphere." Literally, the "atmosphere" of AT (Protocol).

ATmosphere as a Public Sphere

In 18th-century Europe, there existed a "public sphere" where citizens freely exchanged ideas in cafes and salons. A place where people could engage in rational dialogue about society regardless of their status or position. German philosopher Jürgen Habermas saw this as the foundation of democracy.

ATmosphere has the potential to become something similar:

  • Not controlled by any single company

  • Users retain ownership of their speech and data

  • People with different values can selectively connect

Of course, reality doesn't always match the ideal. But in its technical design aimed at creating "a space for equal dialogue," it's fundamentally different from traditional social media.

What Academic Field Does This Belong To?

Here's a question: If we want to study ATmosphere academically, what discipline does it fall under?

  • Engineering? → Protocol design and implementation is engineering

  • Information Science? → Deals with information flow and processing

  • Sociology? → Analyzes communities and human relationships

  • Linguistics? → Studies communication patterns

The answer is "all of the above, and none of the above."

ATmosphere consists of multiple layers:

Layer 4: Social phenomena (communities, discourse, norms)
Layer 3: User behavior (posts, follows, moderation)
Layer 2: Applications (Bluesky, various clients, Feeds)
Layer 1: Protocol (AT Protocol, DID)
Layer 0: Infrastructure (servers, networks)

Each layer falls under different academic disciplines. Therefore, ATmosphere isn't a single "X Studies" but rather a research object where multiple disciplines intersect.

Just as "Internet Studies" never became a single discipline called "Internetology," ATmosphere is best understood as an interdisciplinary field: "ATmosphere Studies."

The Essence: Distributed Networks

So what's the most important characteristic for studying ATmosphere?

It's being a distributed network.

Mastodon is often called "decentralized," but it's actually "federated." Your account is tied to a specific server, and if that server shuts down, your account disappears.

In ATmosphere, your identity through DID doesn't depend on any specific server. Even if you migrate servers, you remain you. This is true "distribution."

The Gap Between "Perceived World" and "Actual World"

Now we get to the heart of the matter.

The human brain contains approximately 86 billion neurons forming networks. The activity of these networks generates our mind—our thoughts, emotions, and memories.

But we can't directly see our brain's network structure. The subjective experience of "how I think" and the objective fact of "what's happening in my brain" are different things.

The same applies to social media.

  • Actual network: Who follows whom, who replies to whom (measurable as data)

  • Perceived network: Who you feel connected to, who you think is influential (subjective perception)

These two often don't match.

In psychology and organizational studies, this has been studied under the name "Network Perception." And the findings are surprising:

  • People cannot accurately gauge their own popularity or centrality

  • In fact, those in central positions tend to misperceive their position more severely

  • People can sense the direction of change (whether their popularity is rising or falling) but cannot see the overall structure

Why Study Network Perception in ATmosphere?

So why is it valuable to study this in ATmosphere specifically?

With traditional social networks (like Twitter/X), obtaining network data has become increasingly difficult. API restrictions, data access limitations, platform black-boxing. It's getting harder for researchers to access data every year.

ATmosphere is different. Because it's an open protocol, anyone can in principle observe the network structure.

Furthermore, ATmosphere's unique structure generates new research questions:

  • How does Feed selection change perception?

    • When you only see your self-selected Feeds (information sources), does your perception of the overall network become distorted?

  • How does Labeler selection change how you see others?

    • When you can choose who is "spam" and who is "trustworthy," how does network perception change?

  • How does server migration affect perception?

    • In a distributed system where migration is possible, does the perception of "my community" become more flexible or more confused?

  • Does distributed identity change self-perception?

    • Does the awareness that "my ID belongs to me" affect how I perceive my position in the network?

These are questions unique to ATmosphere that cannot be asked about centralized platforms.

"Network Perception in ATmosphere" as a Research Field

Based on all this discussion, we propose a new research field:

Network Perception in ATmosphere (NPA)

This research field requires contributions from:

  • Network Science: Quantitative analysis of graph structures

  • Cognitive Psychology: Mechanisms of perception and cognition

  • Sociology: Community formation and social relationships

  • Information Science: Design and impact of distributed systems

  • Linguistics: Communication and meaning-making

No single discipline can answer these questions alone. That's precisely why interdisciplinary collaboration is necessary.

A Space Where Lone Wolves and Researchers Can Both Participate

One more important point.

When you hear "academic research," you might imagine university professors writing papers and presenting at conferences. But we believe ATmosphere research can take a more open form.

Engineers can contribute by writing code and building tools. Visualization tools for graphs, bots for data collection, libraries for analysis. "Building with your hands" becomes a direct contribution.

Researchers can use that data to construct theories. Why do these gaps occur? Under what conditions does perception become accurate? How does this fit into historical and theoretical contexts?

And lone wolf types—people who don't belong to organizations and want to work at their own pace—can also participate. Simply sharing your observations is like releasing something into the atmosphere, contributing to the whole. No forced direct collaboration. Yet just by existing, you become part of the "sphere."

Perhaps that meaning is also embedded in the name "ATmosphere."

Conclusion: Challenges Ahead

"Network Perception in ATmosphere" is still at the stage of just having a name. What's needed going forward:

  • Pilot studies: First measure the gap between actual graph structure and perception on a small scale

  • Methodology development: How do we measure "perceived networks"?

  • Knowledge accumulation: Record findings, share them, critically examine them

  • Community building: Create spaces where interested people can gather and discuss

This blog post is the first step.

In this new space called ATmosphere, what do we see and what do we miss? How different is the "connection" we think we have from the actual "connection"?

The journey to answer these questions begins now.