12.11 — The SaaS industry, ARR fraud, the con

12.11 — The SaaS industry, ARR fraud, the con#

Ocarina rejects SaaS. The SaaS industry is a circular ecosystem where every tool sells to every other tool, where declared revenue is inflated by convention, and where every layer of the capital chain (LP, VC, founder) deliberately maintains a plausible deniability so as to sue the others when the time comes.

1. Contemporary SaaS#

LayerProviders (sample)Model
Hosting & CDNVercel, Netlify, Cloudflare, Fly.io, RenderSubscription per project / per bandwidth
DBSupabase, Neon, PlanetScale, Upstash, MongoDB AtlasSubscription per compute / per GB
AuthClerk, Auth0, WorkOS, StytchSubscription per MAU
MailResend, SendGrid, Postmark, MailgunSubscription per sent email
AnalyticsPostHog, Mixpanel, Amplitude, Plausible, Vercel AnalyticsSubscription per event
Error trackingSentry, BugSnag, Rollbar, HighlightSubscription per event
ObservabilityDatadog, New Relic, Honeycomb, Grafana Cloud, AxiomSubscription per metric volume
Feature flagsLaunchDarkly, GrowthBook, Statsig, ConfigCatSubscription per MAU
E2E testBrowserStack, Sauce Labs, LambdaTest, Cypress DashboardSubscription per run / per parallel
CIGitHub Actions, CircleCI, Buildkite, Vercel CISubscription per minute
Code reviewGraphite, Codeball, CodeRabbit, SourcerySubscription per PR / per dev
Internal commsSlack, Linear, Notion, Figma, Loom, PitchSubscription per seat
BillingStripe, Lago, Orb, Metronome% of processed revenue
CRMHubSpot, Salesforce, Attio, PipedriveSubscription per seat
SupportIntercom, Zendesk, Plain, Help ScoutSubscription per seat + per message
AI toolingOpenAI, Anthropic, Cursor, Vercel AI SDK, ReplicateSubscription per token + per seat

2. Cross-subscriptions#

   ┌──────────┐ pays     ┌──────────┐
   │ Vercel   │─────────>│ Stripe   │
   │          │<─────────│          │
   └──────────┘     pays └──────────┘
        │                     │
   pays │                     │ pays
        v                     v
   ┌──────────┐          ┌──────────┐
   │ Linear   │<────────>│ Notion   │
   └──────────┘   pay    └──────────┘
        │                     │
        v                     v
   ┌───────────────────────────────┐
   │ Slack, Figma, Loom, Sentry,   │
   │ Datadog, Clerk, Cursor, …     │
   │ all subscribed to all         │
   └───────────────────────────────┘
  • Vercel pays a subscription to Stripe (billing), Datadog (obs), Linear (planning), Figma (design), Loom (internal presentations), Slack (comms), Notion (docs).
  • Stripe pays a subscription to Vercel (hosting for its dashboards), Linear, Figma, Slack
  • Linear pays a subscription to Vercel, Stripe, Notion, Figma, Slack.
  • Notion pays a subscription to Linear, Stripe, Slack, Figma.

And so on. Everyone pays everyone.

12.12 — DHH, Paul Graham, Haters, Mean People Fail

12.12 — DHH, Paul Graham, Haters, Mean People Fail#

The Holy Book quotes two big names: David Heinemeier Hansson (DHH) and Paul Graham (pg). These two authors aren’t cited at random; each embodies a precise rhetorical posture. This file explains who they are, what they wrote, and why Paul Graham’s Haters essay (and its younger sibling Mean People Fail) is the direct intellectual precedent of the “First feedbacks” chapter.

12.13 — Lambda calculus, ROP, Haskell / F# / OCaml, monads

12.13 — Lambda calculus, ROP, Haskell / F# / OCaml, monads#

Ocarina invents nothing on the theoretical front. It applies: from Alonzo Church’s lambda calculus (1936) to Eugenio Moggi’s (1989) and Philip Wadler’s (1992-95) monads, formalized as Railway Oriented Programming by Scott Wlaschin (2014).

1. Lambda calculus (Church, 1936)#

Definition#

Lambda calculus (λ-calculus) is a formal system invented by Alonzo Church (1903-1995) to study computability.

PrimitiveNotationSemantics
Variablex, y, zA name
Abstractionλx.MDefinition of a function with parameter x, body M
ApplicationM NCall of the function M on the argument N
RuleEffect
α-conversionRenaming a bound variable (λx.xλy.y)
β-reductionApplying a function: (λx.M) N → M[x := N]
η-conversionλx.(f x) ≡ f if x not free in f

Why it’s foundational#

  1. 1936: Church proves λ-calculus is Turing-complete (Turing publishes his machine in 1937). Church-Turing thesis: anything mechanically computable can be expressed in λ-calculus.
  2. Alan Turing was Church’s doctoral student at Princeton (PhD, 1938).
  3. λ-calculus is simpler than the Turing machine: only 3 constructions. Everything else — integers, booleans, data structures — is encoded with those 3 bricks (Church encoding).

Typed lambda calculus (Church, 1940)#

Church added types in 1940 (simply typed lambda calculus, STLC) to forbid self-application paradoxes (λx.x x). Each variable has a type, and function application is only allowed if types match.

12.14 — AI, and how Ocarina applies it for real

12.14 — AI, and how Ocarina applies it for real#

AI-powered” became a totally idiotic marketing slogan in 2023-2026: 99% of products claiming it bolt a chat.completions endpoint onto an existing product. Ocarina proposes, on the contrary, a software infrastructure designed so an AI can contribute like a senior developer. ocarina-with-ai-example is the proof by example.

1. What AI is today#

LayerExampleAppearance date
Statistical learningRegression, k-means, decision trees1960s-1990s
Deep learningCNN (vision), RNN/LSTM (text)2012 (AlexNet — mainstream adoption)
Transformer-based foundation modelsGPT, Claude, Gemini, Llama2017 (paper), 2018 (BERT), 2022 (ChatGPT)
Agentic workflowsAutoGPT, LangChain, Claude Code, Cursor agents2023+

Note: CNNs were invented by Yann LeCun as early as 1989 (MNIST). AlexNet (Krizhevsky, Sutskever, Hinton, Toronto, 2012) was the industrial breakthrough. LeCun, Hinton, and Bengio share the 2018 Turing Award for their foundational contributions to deep learning.

12.15 — Typing, ISTQB, Reinforcement Learning, probes: why AI works in Ocarina

12.15 — Typing, ISTQB, Reinforcement Learning, probes: why AI works in Ocarina#

Why so much obsession over strict typing, formalizing every interface, alignment on ISTQB, refusing to execute an assertion without having observed? The mechanism is borrowed from Reinforcement Learning: an agent only progresses under constraint + reward signal.

1. A naked LLM is just a bad collaborator#

An LLM without a structured environment is:

  • Optimistic — it assumes its code works.
  • Persuasive — its output is convincing even when wrong.
  • Memoryless — it forgets conventions between sessions.
  • Feedbackless — it has no way to know it was wrong.
  • Loves what it deems “most likely” — it drifts toward popular patterns, not correct ones.

Without an environment, an LLM is an over-confident junior dev producing code that looks right, breaks in prod, and learns nothing from the failure.

12.16 — The underground internet of 1995-2010: forums, tools, banalization of cruelty

12.16 — The underground internet of 1995-2010: forums, tools, banalization of cruelty#

To understand the rage running through Ocarina’s Holy Book, you have to understand what the internet really was before Facebook (2006-2008), before Cloudflare became the gatekeeper (2010+), before moderation was taken seriously. It was a Wild West where state-level capabilities developed in teenagers’ bedrooms, where an adversary’s privacy could be demolished in 48 hours, and where classical legal recourse was inoperative. This page describes the mechanics, without complacency, without heroization.

12.17 — The survivor's psyche: "enemies", "narcissism", "hell", "survival", and how it bubbles up into the code

12.17 — The survivor’s psyche: “enemies”, “narcissism”, “hell”, “survival”, and how it bubbles up into the code#

Ocarina’s Holy Book author uses strong words: enemies, narcissism, hell on earth, survival. These words aren’t stylistic flourishes. They’re the precise descriptors of an experience lived by a whole cohort that went through the 2000s underground internet described in 16-underground-internet-2000s.md. This page explains why these words are accurate, what this lived experience produces on the psyche, and how these psychological adaptations imprint deeply into the way one programs, architects, always stays up to date.

12.18 — "Villains but not Monsters", DEF CON, "live as a white hat or die as a black hat", and Ocarina's ideology

12.18 — "Villains but not Monsters", DEF CON, “live as a white hat or die as a black hat”, and Ocarina’s ideology#

Ocarina is neither an aesthetic whim nor an out-of-the-blue UFO. It’s the technical expression of a coherent ideology inherited from the hacker movement: the Villain, but not a monster ethic, the tribal culture of the conferences (DEF CON foremost), and YTCracker’s stage motto: “live as a white hat, or die as a black hat”.