At SVA, the critique room worked like this. You hang your work at the front of the room. Your peers and your teachers sit around you. Everyone evaluates in good faith. The feedback is specific, grounded, and coming from multiple perspectives at once. Where they agree, you have a strong signal. Where they disagree, you have a decision to make. The accumulated decisions are yours. The work that comes out belongs to you, not to any single voice in the room.

The Formwork Protocol rebuilds that room as a system. Distinct layers of concern, each staffed with evaluative personas extracted from real practitioners, run independently and read together. The layers are the plates. The personas are the trades. The maker is the general contractor.

The word “plates” comes from screen printing. Each color in a print is a separate plate. You build them independently, each one handling its own ink and registration. The final print only exists when the plates combine. If one plate is off, the image misregisters. But you diagnose the problem at the plate level, not at the whole-image level.

Evaluation works the same way. A website has at least two plates: structural and narrative. The structural plate asks: is this well-built? Grid, typography, spacing, hierarchy, economy. The narrative plate asks: does this feel like the thing it’s supposed to be? Identity, world-feel, voice.

A design can score perfectly on the structural plate and fail the narrative plate (well-built but generic). A design can nail the narrative plate and fail the structural plate (the vibe is right but the grid is broken). Both plates must register for the work to hold.

Each plate is staffed with multiple personas. The structural plate might have Vignelli (restraint and systematic limitation), Rams (economy, everything earns its place), and Muller-Brockmann (grid as compositional discipline). They don’t agree on everything. Vignelli’s restraint sometimes pulls against expressiveness that a different structural thinker would allow. That tension is the valuable signal.

The narrative plate might have Victore (bravery, the maker visible in the work), Millman (vulnerability, real human moments), and a peer benchmark. Again, tensions. Victore pushes confrontation. Millman pushes directness. Those aren’t the same thing.

Where personas agree within a plate: strong signal. Act on it. Where they disagree: decision point. The maker chooses. The choices, accumulated across dozens of evaluations, produce work that belongs to the maker. No single persona dominates. The synthesis across all of them produces something that couldn’t have come from following any one influence.

Convergence happens when the plates are read together. The structural plate says the grid is clean. The narrative plate says the identity is buried. Both verdicts are correct. They measure different things. The convergence maps where the two plates agree and where they diverge, and names what’s at stake in the divergence. The maker resolves it.

On a recent pass of my bio page, the vulnerability lens scored strong (the page opens with a real moment from the household) and the restraint lens scored weak (too much personal detail for the visual economy). The tension: raw personal directness vs. visual discipline. That’s a real creative decision. The framework surfaced it. I made the call.

That’s the system. Distinct layers. Independent evaluation. Convergence where the layers meet. The maker at every decision point. The same architecture as the critique room at SVA, running as a repeatable protocol instead of requiring a room full of people.