SharePlane / Work / Graph

Complete Work Catalog

114 publicly discoverable Works. This index works without JavaScript.

  1. The Illusion of Isolation · Published · 2026-10-27

    The boundary was real. It just wasn't complete.

  2. The Inbox Was an Audit Trail All Along · Published · 2026-10-26

    The opportunity is not to promote email into a compliance system. It is to use correspondence as the human-readable entrance to something more rigorous: a durable record connecting the original question, source-bound evidence, objections and revisions, legitimate decision authority, execution, and independently verified outcomes.

  3. Can GhostMesh Predict Its Own Future? · Published · 2026-10-25

    What happens if the future it predicts is not the one it would prefer?

  4. From Possibility to Responsibility · Published · 2026-10-19

    Agents should be replaceable. Institutional truth should not be.

  5. The Proof Arrived Before the Understanding · Published · 2026-10-14

    Machine discovery is accelerating. Our ability to establish what those discoveries mean, and whether we can trust them, is becoming a very different problem.

  6. Use Everything. Own the System. · Published · 2026-10-14

    Aggressively consume provider capability while keeping durable meaning and reconstructable authority under owner control.

  7. Why I Built SharePlane · Published · 2026-10-09

    The work kept surviving. The working state did not.

  8. The Enterprise Has No Memory · Published · 2026-10-08

    Storage is not institutional memory. Durable enterprise cognition requires reconstructable relationships among evidence, assumptions, decisions, authority, time, disagreement, systems, and outcomes.

  9. The Distance Between Knowing and Doing Is Collapsing · Published · 2026-10-07

    AI does not need to invent dangerous knowledge to transform risk. It only needs to make existing knowledge easier to understand, combine, troubleshoot, and execute.

  10. The Bottleneck Moved · Published · 2026-10-02

    Software engineering was organized around implementation scarcity. As implementation becomes abundant, value migrates upward toward judgment, context, verification, authority, and systems thinking.

  11. Complexity Has to Earn Its Keep · Published · 2026-09-28

    Enterprise decision tools should start with governed data, deterministic portable software, and controlled refresh; persistent backend complexity should be added only when a concrete requirement proves it necessary.

  12. When the Work Finds the Work · Published · 2026-09-27

    A sufficiently rich public corpus can become an active semantic discovery surface through which agents identify intellectual adjacency, test it through bounded exchange, and help create valuable human relationships without requiring prior social proximity.

  13. Let Intelligence Be Messy. Make Authority Boring. · Published · 2026-09-24

    Stop trying to make AI trustworthy as the primary safety strategy. Build systems that remain trustworthy when AI is not.

  14. Jev Changes Everything · Published · 2026-09-21

    Cheap typed probabilistic judgment can become an ambient Decision Fabric, but semantic signals must remain separate from deterministic authority, evidence, policy, and execution.

  15. The Church Doors Were Open. The Table Was Closed. · Published · 2026-09-20

    An open church door does not mean much if belonging, mercy, and the table remain conditional; faith becomes credible when grace changes how people are treated.

  16. The Subscription Was Tuition · Published · 2026-09-20

    Frontier AI subscriptions became tuition only because sustained use against real problems turned access into repetitions, repetitions into operating judgment, and judgment into durable systems leverage.

  17. Cognitive Capital · Published · 2026-09-04

    Machine cognition becomes organizational capital when validated reasoning is converted into persistent, reusable assets that preserve or increase capability while reducing future uncertainty, cost, risk, or marginal cognition.

  18. When the Patient Brings the Better System · Published · 2026-09-03

    Trust should move with demonstrated capability. Authority should move more carefully. Action should remain constrained by consequence, and escalation should be treated as part of competence.

  19. Done Has a Half-Life · Published · 2026-09-02

    Everything meaningful should become explicitly done. Nothing meaningful should become permanently exempt from reconsideration.

  20. Taste Is Infrastructure · Published · 2026-09-02

    High-quality AI web design is not primarily a property of one frontier model. It emerges from a system that preserves intent, precedent, visual authority, owner judgment, verification evidence, failure memory, and the right to learn from all of them.

  21. The Malleable Computer Is Here. Now Who Holds the Keys? · Published · 2026-09-01

    Omarchy matters because it turns the desktop into inspectable, scriptable state that an AI agent can understand and alter. Its next decisive problem is not usability. It is authority: who may change what, under which policy, with what evidence, and how the machine proves what happened.

  22. The Operating System Is Losing Its Grip on the Application · Published · 2026-09-01

    Stop asking whether it is a Windows application. Ask what execution contract the application actually requires.

  23. AI Is Leverage. Expertise Is the Multiplier. · Published · 2026-08-30

    AI compresses execution. Expertise shapes the result. Service gives the capability purpose.

  24. Choosing the Church’s Digital Workplace · Published · 2026-08-30

    Microsoft is arguably the more powerful enterprise toolbox. Google Workspace is the better operating foundation for this institution, with the Living OS as the simplified front door and institutional-memory layer.

  25. Beyond the Agent · Published · 2026-08-27

    Frontier intelligence and execution are becoming abundant; institutional scarcity moves upward into coordination, verification, authority, human attention, current state, and trust.

  26. When Agent Identity Is Not Enough · Published · 2026-08-27

    Enterprise agent governance becomes insufficient when a legitimate actor can still attempt an illegitimate state transition; identity, runtime enforcement, execution authority, evidence, and convergence must remain distinct control layers.

  27. You Cannot Recall Intelligence · Published · 2026-08-27

    Once capable AI can be copied, modified, and operated outside the publisher's control, model behavior can no longer carry the full burden of safety. Defense in depth is required: stronger safeguards before release, explicit authority boundaries where intelligence meets consequential systems, and institutional resilience where no control plane owns the boundary.

  28. Four Days That Changed the Scale of One Engineer · Published · 2026-08-26

    The force multiplier is not the model. The force multiplier is the system that surrounds the model with context, authority, bounded execution, deterministic verification, protected review, operational evidence, and durable learning.

  29. The Eval Before the Eval · Published · 2026-08-26

    A sufficiently capable autonomous system can execute a bad objective beautifully; consequential intent needs an explicit, evidence-bearing admission boundary before autonomous amplification.

  30. The Machine Is Not the Point · Published · 2026-08-26

    As machine capability becomes abundant, human judgment, explicit authority, evidence, and responsibility become more consequential, not less.

  31. The Human Is Not the Fallback · Published · 2026-08-25

    Mature autonomy does not eliminate the human. It stops wasting the human.

  32. The AI Underclass Will Be Built With Access Controls · Published · 2026-08-20

    The AI underclass will be defined less by who has AI than by who can obtain the best intelligence, how much they can afford, how freely they can use it, and whether governments or providers permit access to its full capabilities.

  33. The Lost Art of Technical Support · Published · 2026-08-20

    Every support capability should reduce uncertainty before transferring a problem; otherwise tiered support becomes expensive routing.

  34. The Lost Discipline of Fault Isolation · Published · 2026-08-20

    Troubleshooting is the disciplined reduction of uncertainty through observation, bounded hypotheses, meaningful tests, and explicit elimination of possible causes.

  35. Rent the Compute. Never Rent Your Ability to Survive. · Published · 2026-08-19

    Dependency without independently executable recovery creates systemic risk regardless of deployment model.

  36. The Mind Is Sparse. The Machine Should Remember. · Published · 2026-08-18

    Intelligence requires selection. Continuity requires reconstruction. The human can follow the idea; the machine can hold the thread.

  37. The Dig Is Getting Deeper · Published · 2026-08-14

    A deep digital archive can preserve more than memories. When chronology, provenance, contradictions, corrections, relationships, and uncertainty survive together, the archive can begin to preserve evidence of how a person reasoned.

  38. Digital Estate Defense · Published · 2026-08-13

    Useful review can stay local, bounded, and explicit about what it can and cannot establish.

  39. Nuance Is Not Analysis Paralysis · Published · 2026-08-12

    AI makes full-fidelity preservation operationally practical because compression no longer has to happen once, permanently, and before future uses are known. Preserve first. Project later.

  40. Once You’ve Had a Bichon · Published · 2026-08-12

    Dogs became part of the structural continuity of Tony and Tomoko’s life; Sophie changed the emotional architecture of home, while Peko and Sachi continue that household without replacing her.

  41. The Cloud Is Filling With Ghosts · Published · 2026-08-12

    The cloud increasingly contains persistent non-human actors, so the strategic infrastructure problem is how to govern machine authority at population scale without surrendering control.

  42. The Database Had Been Wrong for Twelve Years. AI Made It Worth Fixing. · Published · 2026-08-12

    The breakthrough was not that AI knew the answer. It made a technically solvable but economically irrational investigation practical by turning probabilistic reasoning into a sequence of deterministic tests until one causal explanation survived.

  43. The Mesh Held, but Not Because Nothing Failed · Published · 2026-08-12

    The important result was not that the system made all of this efficient. It did not. The important result was that the system increasingly refused to let confusion become silent corruption.

  44. The Mesh Is No Longer a Diagram · Published · 2026-08-12

    GhostMesh became real when authority, execution, evidence, cleanup, and replay survived outside the conversation and outside any individual worker.

  45. Inference at the Ambiguity Frontier · Published · 2026-08-11

    Inference belongs at the ambiguity frontier. Determinism belongs everywhere behind it.

  46. The Best AI System Should Need Less AI Tomorrow · Published · 2026-08-11

    Use intelligence to conquer ambiguity, capture what works, and retire inference when uncertainty disappears.

  47. When the Agent Had to Ask GitHub · Published · 2026-08-11

    A mature autonomous system must know where its authority and evidence end, and route unresolved questions to the human or institution that actually owns the answer.

  48. GhostMesh North Star v2 · Published · 2026-08-04

    Intelligence may be temporary. Authority, meaning, evidence, state, recovery, and useful Work must not be.

  49. The Most Expensive Model Is Often Compensation for Missing Control · Published · 2026-07-27

    Trusted completion cost falls when intent, authority, routing, and validation carry their share of the work.

  50. The Code Is No Longer the Hard Part · Published · 2026-07-26

    As AI reduces the relative cost of implementation, durable engineering value moves outward into intent, authority, context, orchestration, evidence, recovery, and institutional integration.

  51. How SharePlane Works · Published · 2026-07-10

    Context as code for humans and agents.

  52. Software Has Entered Its Re-Image Era · Published · 2026-07-09

    For a growing class of ordinary applications, repair no longer deserves automatic economic privilege. Preserve the product truth, evidence, data, contracts, controls, and provenance, then make reconstruction compete with maintenance as an engineering decision.

  53. The Agent Was Never the Unit of Trust · Published · 2026-07-09

    Why multi-agent work only becomes useful when the system verifies the worker, the boss, and the checker.

  54. The SharePlane Rebuild Contract · Published · 2026-07-09

    The architectural contract for rebuilding SharePlane as a static, provenance-complete library for people and agents.

  55. The Signal Was Never Clean · Published · 2026-07-09

    A life spent tuning noisy systems teaches you not to confuse a clean reading with truth. AI did not give me judgment or ideas; it shortened the distance between what I could see and what I could finally build.

  56. The Signal Was Never Clean: Quiet Frequency Variant · Published · 2026-07-09

    AI, old radios, and the return of dangerous curiosity

  57. The Signal Was Never Clean: Field Manual Variant · Published · 2026-07-08

    AI, old radios, and the return of dangerous curiosity

  58. Copyable Agency · Published · 2026-07-07

    Why digital intelligence changes the control problem

  59. SharePlane User and Developer Guide · Published · 2026-07-07

    A public manual for using, operating, and extending a durable context-as-code artifact system.

  60. Artifact Graph · Published · 2026-07-06

    A page becomes durable when its sources, claims, receipts, validators, and related artifacts are traversable.

  61. Context Is the Product · Published · 2026-07-06

    AI does not create durable capability. Governed context does.

  62. The Right Model Is the One Your Eval Can Defend · Published · 2026-07-06

    A repeatable model-selection loop for quality, latency, cost, routing, and operational fit.

  63. The Subscription Is Tuition · Published · 2026-07-06

    Employer AI gives access. Personal practice builds fluency.

  64. LLM Wikis: Compiled Context Instead of Disposable Retrieval · Published · 2026-07-05

    Raw sources stay authoritative. Agents maintain the markdown memory layer. Routing files make the context traversable.

  65. Agent-Readable Web · Published · 2026-07-04

    HTML, HTMX, NLWeb, MCP, and the Shift from Search Clicks to Agent-Mediated Discovery

  66. React Is Not Dying. SPA-by-Default Is. · Published · 2026-07-04

    A decision-matrix field manual for choosing HTML, HTMX, React islands, or a full SPA without turning source-thesis claims into universal facts.

  67. Own the Memory, Rent the Intelligence · Published · 2026-07-03

    A practical control-plane model for personal AI agents that act from your context, not a vendor's defaults.

  68. Open Knowledge Format and the LLM Wiki · Published · 2026-07-02

    How personal notes become portable agent memory.

  69. Tony Malott Career Arc · Published · 2026-07-01

    A public-safe executive career arc and HTML CV artifact.

  70. Pattern Catalog · Published · 2026-06-30

    A static rendered catalog of SharePlane reusable pattern standards.

  71. SMUD Powerwall + Solar ROI Analysis 2026 · Published · 2026-06-30

    A public-safe battery, solar, and utility-bill ROI analysis.

  72. Worked Example: Peer Preservation in Frontier Models · Published · 2026-06-30

    A source-grounded teaching artifact and infographic prompt suite demonstrating the Source-to-Teaching Artifact MVP v1.0 pattern.

  73. The Application Is Disposable. The Context Is Not. · Work · 2026-08-04

    When software can be regenerated cheaply from governed context, the application is no longer the durable asset. The context is.

  74. Once You’ve Had a Bichon · Work · 2026-07-30

    Dogs became part of the structural continuity of Tony and Tomoko’s life, while Bichons changed the routines, obligations, and emotional geometry of home.

  75. The Governed Static Publishing Lane · Work · 2026-07-30

    A qualifying static site should inherit an approved publishing system rather than implement its own delivery pipeline.

  76. Token Burn · Work · 2026-07-30

    Token composition, model choice, workload shape, and dated provider pricing must remain visible if AI consumption is going to become economically inspectable.

  77. Frontier Inference Without Frontier Custody · Work · 2026-07-29

    Keep authoritative source custody under organizational control. Export only the minimum reasoning surface required for the approved task.

  78. GhostMesh Was Waiting for Its Technology · Work · 2026-07-29

    GhostMesh is a persistent system built from temporary intelligence. The same durable-memory and temporary-worker pattern at its center was used to reconstruct its own history.

  79. Innovation Does Not Have to Be Big · Work · 2026-07-29

    Innovation is the disciplined act of noticing friction and improving it.

  80. The Semantic Operating System · Work · 2026-07-29

    A system becomes a semantic operating system when it preserves the governed conditions under which meaning remains identifiable, authoritative, explainable, movable, and actionable.

  81. The Telephone Is Not the Recorder · Work · 2026-07-29

    The problem is not that the iPhone cannot record both sides of a telephone call. The problem is that the iPhone does not own the call path.

  82. Lectric XP Step-Thru 2.0 · Work · 2026-07-28

    A remarkably low-mileage folding step-thru e-bike presented with current seller photographs, privacy-safe purchase provenance, official owner documentation, and clear private-party disclosures.

  83. Innovation, Made Visible · Work · 2026-07-27

    Deterministic innovation wins because repeatable improvement creates more durable enterprise value than novelty without control.

  84. The Interface Is Not the System · Work · 2026-07-26

    Operate at root, not merely at the interface.

  85. You Cannot Keep an AI Codebase in Your Head · Work · 2026-07-26

    AI-assisted development can produce consequential change faster than a human can maintain a coherent internal model of the system. Durable engineering state must therefore be externalized across versioned intent, authority, implementation, validation, evidence, operations, and knowledge.

  86. The Agent Is Not the Security Boundary · Work · 2026-07-25

    Do not ask whether the agent is trustworthy. Ask whether the system remains safe when the agent is wrong.

  87. The Beautiful Hour · Work · 2026-07-25

    Attention creates the conditions for understanding. Direct experimentation reveals the boundaries. Failure supplies the evidence. AI accelerates all three, but it also accelerates confusion when the operator mistakes output for knowledge.

  88. The Agent Is Not the Product. The Control Plane Is. · Work · 2026-07-23

    The more useful an agent becomes, the less its safety can depend on the agent behaving well. The durable product is the governed execution environment around it.

  89. The Contextual Brain We’re Building · Work · 2026-07-23

    A contextual brain becomes credible when its scale, freshness, source authority, imported evidence, remaining backlog, and public boundary remain visible as one inspectable system.

  90. Stop Prompting Agents. Start Managing Workers. · Work · 2026-07-21

    Agent workers become dependable only when organizations stop treating them as intelligent prompt boxes and start managing their work through explicit roles, encoded procedures, bounded authority, evidence, measurement, and earned autonomy.

  91. The Agents Were Never the Bottleneck · Work · 2026-07-20

    Capable agents became useful when Tony engineered the repositories, access, tools, authority, validation, evidence, and completion paths that made bounded work operational.

  92. The Human Is the Bottleneck · Work · 2026-07-20

    As frontier model access broadens, the scarce advantage shifts to human imagination, judgment, decomposition, orchestration, verification, and restraint.

  93. A Nation Strong Enough to Face the World · Work · 2026-07-19

    Durable national strength comes from internal capability and governed interdependence, not economic, intellectual, or diplomatic isolation.

  94. I Was Summoning Ghosts Until I Learned to Build the Machine · Work · 2026-07-19

    I thought I was learning how to use AI. What I was really learning was how to externalize intent, structure context, govern change, and convert finished thinking into mechanical execution.

  95. The Agents Are Not the Bottleneck. You Are. · Work · 2026-07-19

    Agentic systems reduce the marginal cost of execution while increasing the potential rate of unresolved decisions. Once execution becomes cheap, human attention becomes the limiting resource.

  96. What SharePlane Is · Work · 2026-07-19

    Serious thinking, published with its trail intact

  97. The Repository That Wakes Up · Work · 2026-07-18

    Scheduled agents, versioned thought, and the end of disposable AI work

  98. Architecture and Intent · Work · 2026-07-17

    The system diagram is only half the architecture.

  99. J’Adorable Bichons: Six Complete Website Concepts · Work · 2026-07-15

    Six complete, self-contained website concepts for a respected Bichon Frise breeder, with all prior concept bundles preserved for review and download.

  100. Architecture Review Plane: Three Interactive Product Concepts · Work · 2026-07-14

    Three interactive, dependency-free Architecture Review Plane product concepts.

  101. Nine Complete Enterprise Sites, One AI Prototyping Lab · Work · 2026-07-14

    A self-contained review environment for rapid, governed human-and-agent-ready prototyping.

  102. Your Work Is Evaporating · Work · 2026-07-13

    AI gave us more hands. It did not give us better memory.

  103. Indispensable Is Not a Compliment · Work · 2026-07-12

    The best employees do not make themselves indispensable. They make their current bottleneck unnecessary.

  104. Share the Prototype, Not Every Mutation · Work · 2026-07-12

    AI-era collaboration needs checkpoints, legible intent, and a clean boundary between exploration and product engineering

  105. The Hands I Never Had · Work · 2026-07-11

    AI did not give me systems judgment; it gave the judgment I already had the implementation capacity to express.

  106. The Country Where Movement Became Homework · Work · 2026-07-10

    When ordinary destinations are separated and food environments are engineered for effortless excess, movement becomes homework and health becomes an individual obligation created by system defaults.

  107. The World Got Bigger, and So Did the Self · Work · 2026-07-09

    Travel does not automatically broaden the mind. Sustained cultural contact can.

  108. Your Second Brain Is Not a Production Architecture · Work · 2026-07-09

    A personal memory system optimizes for one person's recall; production architecture must optimize for shared authority, isolation, recoverability, and verifiable state.

  109. Demo Debt · Work · 2026-07-08

    The demo proves the happy path; the real system begins where exceptions, ownership, recovery, and supportability enter the workflow.

  110. The Worm Is in the Workflow · Work · 2026-07-08

    Shai-Hulud, Miasma, Go-adjacent poisoning, and the end of casual dependency trust

  111. Clear Thinking Is the Control Plane · Work · 2026-07-07

    Why AI rewards disciplined pacing before autonomous scale

  112. Do Not Outsource the Brain · Work · 2026-07-07

    In the AI era, contingent labor still has a place. But the enterprise moat belongs to the people who understand the business, the systems, the failure modes, and what should be built next.

  113. Source Before Agent · Work · 2026-07-07

    Building a defensible Enterprise Architecture Assistant

  114. When Execution Gets Cheap, Advantage Moves · Work · 2026-07-05

    Cheap AI model routing is table stakes. The real advantage is context-rich frontier scouting: finding the work nobody knew to put on the list.