Publications

Alexander Karp and Nicholas Zamiska have written a bestseller that poses an important question and gives the wrong answer. The Technological Republic asks how democratic societies should organize AI infrastructure. Cube Commons proposes the other answer: commons-governed, polycentric, and built to outlast any single vendor. This paper is a sketch of what that other answer is, and why it is the more durable democratic architecture.

The Other Answer moves fast through ideas with long histories. This addendum slows down and walks through each concept for readers who haven't encountered the sources before. Organized in three tiers — general reader, civic technologist, and policy reader — it covers the commons, Ostrom's polycentric governance, subsidiarity, Illich's convivial tools, free software licensing, local-first software, Estonia's X-Road, and the relicensing wave of 2018–2025.

A companion expanded with three new reader tiers for Global South civic institutions, European publics, and non-aligned readers. Seventeen concepts — from multipolarity and tianxia to the Digital Silk Road and the Karp/Schmidt frame — each explored through up to six lenses, making explicit how the same argument reads differently from Lagos, Berlin, Jakarta, or anywhere that rejects the bloc-choosing frame entirely.

Alexander Karp and Nicholas Zamiska have written a serious book, and it deserves a serious response. This seven-part essay engages The Technological Republic on its own terms: what the diagnosis gets right, where the argument slides from "Silicon Valley should reconnect with society" to "concentrated state-aligned firms should be that connection," what the historical record actually shows about computing's foundational vision, what the live empirical record shows about the architectural model the book defends, the China question reframed, the moral question taken seriously, and what a reconstruction would look like.

Formal submission to NIST Docket NIST-2025-0035 (Federal Register Vol. 91, No. 5, January 8, 2026), submitted March 9, 2026. Proposes Intrinsic Access Control (InAC) and the Enforcement Location Principle (ELP) as candidate vocabulary for federal AI agent security guidance. Includes formal definitions, analytical findings, and actionable recommendations. A full Technical Companion accompanies this submission.

The extended version of the NIST/CAISI submission. Contains complete formal proofs, extended independence arguments, detailed attack payload analysis, governance maturity assessment rubrics, a federal agency deployment checklist, a worked deployment example, and a full glossary. Appendices A through H. References of the form "(Technical Companion, Section X)" in the main submission point to this document.

Applies the rendering pipeline — projection, rasterization, compositing — as a mathematical framework for understanding multi-agent coordination. Demonstrates that coordination can be understood as the problem of rendering a consistent shared state from distributed, partial observations. Cube Commons Technical Report CC-TR-2026-003.

Formal specification of Intrinsic Access Control as a sixth access control model, distinct from MAC, DAC, RBAC, ABAC, and PBAC. InAC proposes that the agent subject is simultaneously the enforcement mechanism — collapsing the traditional enforcement location problem into the agent's identity. Submitted as part of the NIST/CAISI RFI process and developed independently as a standalone specification.