Research-To-Writing Source Log

What is a Research-To-Writing Source Log?

A Source Log is a pre-writing checklist that sits between your research and your published article. It answers five questions before you write a single sentence of prose:

  1. Where did my sources come from? — Kindle highlights, wiki notes, financial data, legacy archives, or external references.
  2. Am I allowed to use each source publicly? — Privacy screen: no personal health data, no family details, no proprietary signals, no employer confidential information.
  3. What is my one-sentence thesis? — If you can't say it in one sentence, you're not ready to write.
  4. Have I validated the article against quality gates? — Source registry, privacy check, article review, internal links, publish-log entry.
  5. How will I distribute and measure it? — WordPress, WeChat, X, newsletter — and what does success look like in 7 days?

Why use it?

Most writing failures happen before the writing starts:

  • You cite a source you shouldn't have cited (privacy leak).
  • You publish without checking whether the facts are traceable (credibility risk).
  • You write 3,000 words and realize you don't have a clear thesis (wasted time).
  • You publish but forget to plan distribution (invisible article).

The Source Log prevents all four. It takes 10-15 minutes to fill before writing — and saves hours of rework after.

Download

The template is a Markdown file. Open it in any text editor, Obsidian, Notion, or your preferred note-taking tool.

Example: How I used it

Here's how I filled the Source Log for my article 精神谱系：从斯多葛到贵格会，从磨镜人到摘茶人 — an 8,000-word essay drawing from Kindle highlights, wiki notes, and legacy archives:

  • 5 source families mapped (CLAUDE_REFERENCE, Authenticity.md, Wiki, Kindle, public bios)
  • Privacy screen: all 7 risk categories checked → 0 present
  • Thesis: "These seemingly unrelated traditions converge on one formula: doing what duty requires, in places no one sees."
  • Validation: 6/6 checks passed before publishing

The full filled example is available in the GitHub repository.

Related