惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

Google DeepMind News
Google DeepMind News
B
Blog RSS Feed
量子位
aimingoo的专栏
aimingoo的专栏
V
Visual Studio Blog
Y
Y Combinator Blog
Vercel News
Vercel News
云风的 BLOG
云风的 BLOG
宝玉的分享
宝玉的分享
Engineering at Meta
Engineering at Meta
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
GbyAI
GbyAI
人人都是产品经理
人人都是产品经理
博客园 - 叶小钗
Stack Overflow Blog
Stack Overflow Blog
大猫的无限游戏
大猫的无限游戏
Microsoft Security Blog
Microsoft Security Blog
B
Blog
Last Week in AI
Last Week in AI
有赞技术团队
有赞技术团队
博客园 - 聂微东
腾讯CDC
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
J
Java Code Geeks

DEV Community

Authentication Security Deep Dive: From Brute Force to Salted Hashing (With Java Examples) Why AI Systems Don’t Fail — They Drift Spilling beans for how i learn for exam😁"Reinforcement Learning Cheat Sheet" I Replaced Chrome with Safari for AI Browser Automation. Here's What Broke (and What Finally Worked) How Python Borrows Other People's Work The $40 Architecture: Processing 1 Billion API Requests with 99.99% Uptime Vibe Coding: A Workflow Guide (From Zero to SaaS) Most webhook security guides protect the wrong side. The scary part is delivery. Headless CMS for TanStack Start: Build a Blog with Cosmic EU Age Verification App "Hacked in 2 Minutes" — What Actually Happened Comfy Cloud’s delete function does not actually remove files Running AI Models on GPU Cloud Servers: A Beginner Guide Event-driven media intelligence with AWS Step Functions and Bedrock I scored 500 AI prompts across 8 quality dimensions — here's what broke How to Call Google Gemini API from Next.js (Free Tier, No Backend Needed) The Portal Protocol: Reclaiming Human Connection in the Age of AI How to Fix Your Team's Scattered Knowledge Problem With a Self-Hosted Forum Intro to tc Cloud Functors: A Graph-First Mental Model for the Modern Cloud Designing Multi-Tenant Backends With Both Ownership and Team Access I Built a Neumorphic CSS Library with 77+ Components — Here's What I Learned PostgreSQL Performance Optimization: Why Connection Pooling Is Critical at Scale Cómo construí un SaaS multi-rubro para gestionar expensas en Argentina con FastAPI + Vue 3 🚀 I Built an Ethical Hacking Scanner Tool – Open Source Project I Replaced /usage and /context in Claude Code With a Single Statusline A Pythonic Way to Handle Emails (IMAP/SMTP) with Auto-Discovery and AI-Ready Design I Collected 8.9 Million Polymarket Price Points — Here's What I Found About How Markets Really Move EcoTrack AI — Carbon Footprint Tracker & Dashboard Everyone's Using AI. No One Agrees How. 5 self-hosted ebook managers worth trying in 2026 Building Your First AI Agent with LangChain: From Chatbot to Autonomous Assistant
From Eclipses to P95 Latency: What the Joseon Dynasty Can...
Lenard Francis · 2026-06-01 · via DEV Community

The Joseon Dynasty ruled Korea for more than five centuries, from 1392 to 1897.

That is longer than the United States has existed. Longer than the printing press has been in widespread use. Five hundred years of one government, one bureaucracy, one record-keeping system.

And they documented everything.

The 朝鮮王朝實錄 (Veritable Records of the Joseon Dynasty) is one of the most extensive continuous historical records ever produced. Royal decrees, diplomatic correspondence, criminal cases, military campaigns, natural disasters, celestial observations, agricultural conditions, and administrative decisions were meticulously recorded.

Every eclipse. Every comet. Every drought. Every flood. Every tiger that wandered into a village.

At first glance, it reads like a civilisation obsessed with omens. Look closer and it begins to resemble something else: an accountability system operating at a national scale.

The Mandate of Heaven as an Accountability Mechanism

Joseon inherited the concept of the Mandate of Heaven from Chinese political philosophy.

The basic premise was simple: Heaven's approval of a ruler could be inferred from events in the natural world. Stable harvests, favorable weather, and orderly skies suggested good governance. Floods, eclipses, unusual celestial events, and other disruptions demanded attention.

Whether or not one accepted the underlying cosmology, the system functioned as a powerful accountability mechanism.

When an eclipse occurred, someone had to observe it. Someone had to record it. Officials had to discuss its significance. The court had to determine whether action was required. The response had to be documented. And the entire process became part of a permanent historical record.

A king could not plausibly claim ignorance of a reported eclipse.
An official could not quietly invent a justification for a policy years later. The record existed. The deliberation existed. The decision existed.

Accountability was structural.

https://ajin.im/is/building/omen.ops/

A Dynasty Rendered as Telemetry

Recently, Ajin built omen.ops, a project that renders the Veritable Records as a modern observability dashboard.

It is one of the most serious pieces of digital scholarship I have encountered. Every entry is sourced, annotated, and presented with the gravity the original record-keepers intended. Rather than merely digitising the records, the project reinterprets them through the lens of modern operations, observability, and incident management.

Suddenly, centuries-old historical events look remarkably familiar. Eclipses appear as system alerts. Comets register as anomaly spikes. Droughts become degradation events. The Mandate of Heaven itself is represented as a system health score.

The effect is both humorous and strangely illuminating.
A guest star observed over thirteen consecutive night-watches in 1592 appears as a P1 incident. The court astronomers of the Gwansanggam—the royal bureau responsible for celestial observation—tracked its position relative to known stars, recorded its persistence, and noted the absence of any established remediation procedure.

In modern operations language, the alert was acknowledged, classified, documented, and ultimately deemed unactionable.

Every engineer who has ever been paged at 3 AM has encountered the same category of problem.The dashboard also presents a derived metric called the Mandate Volatility Index. It compresses centuries of recorded anomalies into a single score relative to a reign's baseline conditions.

The historical court and the modern SRE team face the same challenge: overwhelming amounts of signal with limited human attention. Different centuries. Different tools. Same problem.
They needed summaries. We use dashboards. They had volatility indexes. We have P95 latency graphs.

The Tiger Incident

The most memorable entry may not involve the heavens at all.

In 1571, a white-browed tiger reportedly killed hundreds of people and livestock near the capital. The response escalated rapidly. The court mobilised a specialised tiger-catching commander and launched a coordinated hunt.

Then a secondary problem emerged. The soldiers sent to eliminate the tiger began looting civilians. The court was suddenly managing two incidents instead of one. The tiger threat was eventually mitigated. Multiple animals were killed. The military operation was scaled back. Reports of civilian misconduct were documented.

Incident opened. Response initiated. Unexpected side effects detected. Mitigation adjusted. Incident closed. The terminology changes. The workflow does not.

The Governance Problem Hasn't Changed

The tools have changed completely. The governance problem has not. What fascinates me about the Joseon system is not the astronomy. It is the process.

Observe - Record - Deliberate - Authorize - Act - Document

That sequence appears repeatedly throughout the Veritable Records. It is also the sequence behind every mature operational system. Modern monitoring platforms are excellent at collecting signals. They can detect latency spikes, memory pressure, queue backlogs, failed deployments, and infrastructure degradation in seconds.

What many systems still struggle with is everything that happens after detection. Who saw the alert? Who approved the response?
What evidence was considered? Why was a particular action taken?
Can someone reconstruct the decision six months later?
Detection is only the beginning. Accountability begins when decisions become traceable.

Building for Active Control

This idea sits at the centre of what I am building with FastAPI AlertEngine.

FastAPI AlertEngine is incident intelligence for FastAPI services. A free SDK adds health monitoring to your application with a single line of code. When degradation is detected, a managed orchestrator investigates the likely cause and sends you a WhatsApp or Telegram notification containing a single-use approval link. Nothing executes without your authorisation.

The goal is not simply to collect more alerts. Most organisations already have more alerts than they can reasonably process. The goal is to preserve the decision chain.

Observe the signal. Capture the evidence. Present the context.
Recommend an action. Require authorisation. Execute the response.
Record everything.

The Joseon court performed this process with a brush, ink, astronomers, and royal historians. We perform it with telemetry pipelines, machine reasoning, and APIs. The difference is speed. The court might take days to process an eclipse.
We can detect a P95 latency spike, identify likely causes, generate remediation options, and request approval in seconds.

But the underlying governance problem remains unchanged.
How do you ensure that the people responsible for a system are confronted with evidence, required to make a decision, and unable to rewrite history afterward?

Five hundred years ago, the Joseon Dynasty answered that question with brush and ink. We are still figuring it out with Redis and JWT tokens.

Explore the historical dashboard: https://ajin.im/is/building/omen.ops/

If you run FastAPI in production and want incident response that asks before it acts, the free SDK is available through FastAPI AlertEngine.