The Meta-Layer: Year in Review

A Meta-Layer Year in Review

As this year closes, I want to take a moment to say where I think we actually are. Not aspirationally or optimistically. Structurally.

When something goes wrong, I’ve learned to ask myself a question:

How could this be the best thing ever?

When the Meta-Layer Infrastructure SIG failed to make it through the ISOC selection process, I didn’t have an answer right away. It was frustrating. We had done the work, built a serious global team, modeled governance on IETF practice, and showed up in good faith. Losing that vote felt like a setback.

The answer didn’t come immediately. A couple weeks later, it came in a leadership meeting, almost casually, when Twahir Hussein Kassim suggested that we start a Meta-Layer chapter instead.

I felt it the moment he said it. The room shifted. I remember telling Twahir it felt like a genuine hack of the system. Until that moment, chapters had been strictly geographic. The assumption was so baked in that I wasn’t even sure a topical chapter was allowed. We paused everything else and made checking that assumption a priority. When we confirmed that topical chapters were, in fact, permitted; it felt like a door quietly unlocking. What had looked like a dead end suddenly revealed another dimension.

With a little distance, it became clear that a Meta-Layer chapter inside ISOC may actually be a stronger move than a SIG.

Chapters have access to significantly more funding. They carry more visibility and influence inside the organization. And most importantly, a chapter is not a side working group. It is a durable civic structure.

In other words, the loss of the SIG was the best thing that could have happened.

So today, we submitted an application for the Meta-Layer chapter to ISOC.

One additional thing I want to be explicit about: I will not be taking a board seat in the Meta-Layer chapter. The reason being, we need to develop and spread the leadership and I already have a lot of responsibility being Director of the Meta-Layer Initiative and facilitator of the Meta-Layer Infrastructure SIG (which exists despite ISOC's rejection). I am happy to be an active member and special advisor to the board to the extent they want that support.

We will run an open process to determine chapter leadership: chair, vice chair, secretary, and treasurer plus additional board seats. Any founding member will be eligible for a board role, and I believe at least 50% of the board seats should be reserved for youth.

That structure matters as much as the work itself.

What We Actually Did This Year

With that context, here is what we actually did this year. Not positioning. Not aspiration. Work.

• Conducted deep research on AI agents that led us to an entirely meta-layer approach to AI safety, including an unpublished paper on the Safe AI Trifecta and a vision for a meta-layer coordination protocol grounded in MCP

• Engaged deeply with people-centered Internet and AI work, shaping the Meta-Layer as civic infrastructure rather than a technical add-on

• Launched themetalayer.org, making the work public and legible

• Ran an open Call for Input, receiving roughly 50 substantive submissions from builders, researchers, and civic actors

• Built the Desirable Properties application (https://app.themetalayer.org), enabling structured review and synthesis of community input across DP submissions

• Formed the Meta-Layer Infrastructure SIG, recruited a global team, and designed governance modeled on IETF practices. This exists as a program of the Meta-Layer Initiative that is unaffiliated with ISOC.

• Submitted the SIG to ISOC, received a rejection driven by a deeply flawed SIG selection process, and provided direct feedback to ISOC leadership on those failures

• Recruited members and partners for a Meta-Layer chapter and formally submitted the chapter application to ISOC

• Drafted the first chapter of the Desirable Properties of a Meta-Layer essay, responding to the September 16, 2024 challenge issued by Vint Cerf, including the full chapter for "DP1: Federated Storing Authentication and Accountability"

• Authored Digital Vellum Note #3 on digital provenance. (Digital Vellum is a public-interest research and publishing initiative founded by Vint Cerf, focused on long-horizon thinking about digital provenance, trust, and Internet governance.)

• Forked the IETF RFC management codebase to support Meta-Layer governance workflows

• Built the foundation of Canopi, an application for on-page presence and shared attention, which is scheduled to enter beta in Q1

• Raised $1,100 to enable two young women engineers (including an intern) to continue building on the Meta-Layer via GoFundMe.

None of this was hypothetical. None of it was easy. And it was done with zero funding and zero compensation - which is quite obviously unsustainable.

None of this was hypothetical. None of it was easy. And it was done with zero funding. Nada. Zilch.

The Meta-Layer is no longer just an idea.

It has artifacts. It has contributors. It has institutional surface area.

And it has already encountered resistance.

The Threshold Moment

The Threshold Moment

We are entering a period where the constraints of the existing web are starting to matter in ways they never did before.

Artificial intelligence is advancing faster than our institutions, norms, and shared sensemaking can adapt, and it is being fused into an Internet that was never designed to concentrate or exercise this much power.

The effects are already visible. Reality is fragmenting. Trust is thinning. Work is being destabilized, while wealth concentrates at levels we have not seen in generations.

And yet, there is remarkably little scrutiny of the underlying structure that is shaping these outcomes.

The Quiet Choice We Already Made

Without much public debate, the future of intelligence has been locked into an extractive, one‑layer web.

A web optimized for capture rather than coordination. For engagement rather than understanding. For attention rather than truth. For profit rather than collective problem‑solving.

Even institutions that once provided steady moral and architectural grounding for the decentralized web are beginning to buckle as AI-driven systems introduce speed, agency, and power concentrations that exceed what norm-setting, advocacy, and voluntary coordination can realistically absorb.

In 2020, Mozilla spoke about fixing the Internet. In 2023, it shifted to reclaiming the Internet. Now we hear resignation in the language of a post‑naive Internet.

Here is the quiet truth.

We ceded our power. Increment by increment. Interface by interface.

Not maliciously. Gradually. Almost without noticing.

From Maps to Screenshots

The Web does not represent reality so much as compress it. Context falls away, collective presence disappears, and time, provenance, intent, and relationship are flattened into fragments that travel well but mean less. We are not lost because reality has become too complex. We are lost because the tools we rely on to navigate it are dimensionally wrong.

The old parable of the blind men and the elephant isn’t about foolishness. It’s about partial perspective. Each person is faithfully reporting what they can touch, but no one has access to the whole. That is not simply an epistemological failure. It is an interface failure.

Depth Perception, Not Better Maps

This is not about fixing the map. It’s about recovering depth.

A two‑dimensional interface can surface isolated statements, linear feeds, and atomic actions. But lived reality is layered. It involves overlapping perspectives, shared attention, emergent meaning, and the slow accumulation of history, power, and relationship.

When conversations collapse online, it is rarely because people disagree about reality itself. More often, each of us is operating from a different flattened slice of it, with no shared way to reassemble the whole.

The Evolutionary Leap

We often talk about layers as if they were frosting on a cake, something decorative that can be added later. That metaphor breaks down at scale. This shift is closer to biology than design.

Single‑cell organisms don’t fail because they lack intelligence. They fail because they cannot coordinate internally. Multicellular life did not emerge because cells became smarter, but because they learned to signal, specialize, and share boundaries.

The Meta‑Layer is not a feature added to the web. It is the missing nervous system that allows coordination to emerge.

The Coordination Gap

The Web today is alive, but it is not coordinated. There is no shared nervous system, no immune response, and no durable collective sensemaking. As a result, we are accessing perhaps one percent of what this system could become.

Artificial intelligence will not correct this mismatch. It will scale it. AI optimizes whatever interface it is given, and if the interface remains flat, AI will harden that flatness. Once interfaces ossify, reclaiming depth becomes far harder.

From Flat Systems to Living Ones

Moving from one layer to many layers is not an incremental upgrade. It is evolutionary. It is the difference between drifting cells and an organism capable of coordinated action.

This is not about belief. It is about capacity. We are not trying to build a better map. We are trying to grow the organs required for shared reality. And evolution does not pause for comfort.

Peering into 2026

As this year closes, one thing is becoming difficult to ignore. 2026 is not just another milestone on a roadmap. It is a threshold year.

It is the year overlays stop feeling optional and a layered web shifts from theory to necessity. Not because it is fashionable, but because a one‑layer web can no longer carry the load we are placing on it.

Artificial intelligence, labor disruption, civic fracture, and concentration of power are converging. Convergence requires coordination.

The next evolution of the Internet will not come from more content, faster models, or smarter feeds. It will come from restoring a missing dimension: shared context, visible presence, and collective sensemaking above the page.

This possibility was anticipated decades ago. The Internet was never meant to be just a document system. It was meant to support collective intelligence.

Whether we rise to that possibility is still unresolved. But pretending a flat web is enough will stop working very soon.

________________________________________________________________________

I’ll share a separate post this weekend focused on what comes next.