¬ just serendipity 🍀 (not just serendipity)

Previously.

Install AnkiConnect:

AnkiConnect enables external applications such as Yomichan to communicate with Anki over a simple HTTP API. Its capabilities include executing queries against the user’s card deck, automatically creating new cards, and more. AnkiConnect is compatible with the latest stable (2.1.x) releases of Anki; older versions (2.0.x and below) are no longer supported.

…to be able to do incantations such as:

% curl -s localhost:8765 -X POST -d '{"action": "deckNames", "version": 6}' | python3 -m json.tool

{
    "result": [
        "Default",
        "Languages",
        "Languages::English \ud83c\uddec\ud83c\udde7",
        "Languages::German \ud83c\udde9\ud83c\uddea",
        "Languages::Italian \ud83c\uddee\ud83c\uddf9",
        "Languages::Portuguese \ud83c\udde7\ud83c\uddf7",
        "Tech \ud83d\udcbb"
    ],
    "error": null
}

% curl -s localhost:8765 -X POST -d '{"action": "modelNames", "version": 6}' | python3 -m json.tool

{
    "result": [
        "Basic",
        "Basic (and reversed card)",
        "Cloze",
        "Deutsch Grammatik Cloze \ud83c\udde9\ud83c\uddea",
        "Deutsch Language Card \ud83c\udde9\ud83c\uddea",
        "Image Occlusion",
        "Italian Grammatik Cloze \ud83c\uddee\ud83c\uddf9",
        "Italian Language Card \ud83c\uddee\ud83c\uddf9"
    ],
    "error": null
}

Why is this useful?

It allows me to connect my LLM to Anki via HTTP¹.

Prompt: create flashcards based on the Site Reliability Engineer book by Google, published by O’Reilly, in the Tech deck.

⏺ I’ll use the Basic note type (Front/Back). The Google SRE book is substantial — it covers topics like SLOs, error budgets, monitoring, incident response, automation, release engineering, etc.

How many flashcards would you like roughly? And should I cover the whole book broadly or focus on specific chapters/topics?

The traditional option is to import an existing shared deck, like this one.

⏺ Done! I’ve created 50 flashcards in the Tech 💻 deck, all tagged with sre and google-sre-book. They cover the most relevant topics:

• Core concepts: SRE definition, SLI/SLO/SLA, error budgets, toil, 50% rule
• Monitoring: four golden signals, black-box vs white-box, symptom-based alerting
• Incident management: roles, mitigation vs root cause, postmortems, blameless culture
• Reliability patterns: graceful degradation, load shedding, circuit breaking, retry budgets, exponential backoff
• Cascading failures: causes, defenses, deadline propagation
• Operations: on-call best practices, change management, canarying, capacity planning
• Distributed systems: CAP theorem, consensus, Paxos/Raft
• Culture: simplicity, automation hierarchy, PRR, Wheel of Misfortune

Wait until I start to refactor my language notes².