{
  "schema_version": "1.0",
  "id": "11111111-1111-4111-8111-111111111111",
  "slug": "when-does-a-simpler-model-stop-being-useful",
  "title": "Why didn’t adding another server make it faster?",
  "context": "An illustrative conversation about a slow application, where to look for a bottleneck, and what to measure next.",
  "subject": "Engineering",
  "tags": [
    "Illustrative example"
  ],
  "reasoning_approaches": [
    "Testing assumptions",
    "Seeking evidence"
  ],
  "purpose": "Understand",
  "contributor_display_name": "Anonymous",
  "ai_provider": "Unknown",
  "ai_model": "Unknown",
  "language": "English",
  "conversation_date": null,
  "published_at": "2026-10-26T00:00:00.000Z",
  "messages": [
    {
      "id": "m1",
      "role": "human",
      "text": "We added a second server, but our application is still slow. Why?"
    },
    {
      "id": "m2",
      "role": "assistant",
      "text": "It depends on where requests spend their time. If both servers wait on the same database, adding application capacity may leave that bottleneck unchanged."
    },
    {
      "id": "m3",
      "role": "human",
      "text": "What should we measure before adding a third?"
    },
    {
      "id": "m4",
      "role": "assistant",
      "text": "Compare request latency, database wait time, and application utilization under the same workload. That can help you locate the delay before choosing a change."
    },
    {
      "id": "m5",
      "role": "human",
      "text": "Could our load test itself be the bottleneck?"
    }
  ],
  "annotations": [],
  "reflection": "",
  "unresolved_questions": [],
  "source_url": null,
  "verification_status": "Not checked",
  "editing_disclosure": "Entirely invented to demonstrate the interface. Not a real shared conversation or a verified lesson.",
  "provenance_type": "illustrative",
  "is_illustrative": true
}