Input product
The parser turns the user's product of Steenrod squares into monomials that the rewrite engine can inspect.
Like a sentence whose words are in the wrong grammar order.
The workbench rewrites products of Steenrod squares into the Serre-Cartan admissible basis and exposes every local Adem move, parity decision, cancellation, and compressed graph structure.
Each stage is clickable. The same current trace is interpreted as grammar repair, route splitting, binary filtering, graph compression, and final basis filing.
The parser turns the user's product of Steenrod squares into monomials that the rewrite engine can inspect.
Like a sentence whose words are in the wrong grammar order.
AdemScope should not pretend to replace serious mathematical software. It should become the visual proof and translation layer between Steenrod algebra calculations, homological algebra, computed spaces, spectral sequence workflows, and formal proof certificates.
Cross-check normal forms, compare Serre-Cartan and Milnor bases, export change-of-basis evidence.
Click a trace result, ask Sage for the same basis coordinates, then display agreement or convention mismatch as a visual certificate.
Steenrod algebra oracle
Bright cells are not promises of completed integration. They are the most plausible interfaces: traces to Sage, bases to spectral sequence tools, homology to CHomP/Kenzo/HAP-style systems, and certificates toward formal proof environments.
The strongest direction is a verified rewrite graph workbench with Sage comparison, proof certificates, basis conversion, corpus generation, and eventually odd-prime and cochain-level modules.
Typed algebra core, parser grammar, Sq^0 / zero semantics, deterministic ordering, trace recording, pretty + LaTeX output.
Which reduction strategy minimizes intermediate support size?