Using Logseq to research and plan for a video
Based on Tools on Tech's video on YouTube. If you like this content, support the original creators by watching, liking and subscribing to their content.
Use Logseq topic pages as long-lived single sources of truth for research, and keep per-video content pages as short-lived drafting spaces built from templates.
Briefing
Logseq is used as a two-layer system that turns scattered research into repeatable YouTube output: a long-lived “topic” page for single-source-of-truth knowledge, and a short-lived “content” page that acts as a working template for each specific video. The key idea is to keep research stable and reusable while letting the per-video page stay flexible—so notes don’t get duplicated across multiple places and later videos can start from existing references instead of rebuilding context.
In practice, each topic page stores enduring material such as PDFs (like a keyboard manual), customization details, URLs, tags, and—crucially—where that topic appears in past or future videos. The workflow runs in two modes. “Focusing on this topic” means spending a few hours working directly inside the topic page, since that’s where the information will ultimately live. A second mode captures “bright ideas” midstream: new observations or articles get logged into a journal, then linked back to the relevant topic page as TODO items. When it’s time to work, a simple query surfaces open tasks tied to that topic, making it easy to see what still needs processing.
For long-form writing, the system leans on Logseq’s outliner approach rather than forcing a traditional table of contents. Folding lets the writer collapse details into a readable structure and expand subpages only when needed. To avoid the common “too many panes” problem, the workflow favors a sidebar-based cross-check: open related notes in a side window while verifying, then close them once done. Keyboard navigation (Alt + arrow keys) supports fast movement through the outline.
The transcript also highlights a practical limitation—side-by-side documents can feel clunky—so the workflow instead moves supporting context into the sidebar and keeps the main work moving top to bottom. For email handling, another query-based approach organizes messages linked to the topic/content pages. Emails get broken into blocks: either a single quote block for quick reference or bullet points for granular linking to specific talking points.
Research happens inside the topic pages, but processing is intentionally deferred. Notes are captured and linked to related “namespaces” (for example, keyboard sets), yet the writer doesn’t try to clean or fully structure everything immediately. Instead, references are left clickable so future work can quickly bootstrap understanding. A concrete example shows keyboard research: notes about one keyboard are stored under the keyboard topic page, while the per-video content page pulls only what’s needed for the current outline.
The per-video “content” page is built from a template and includes properties like type (video), state (e.g., doing), and an episode label used to track which drafts are active or simmering. Queries list all pages marked as videos in a given state, enabling quick access to titles, related topics, and links to supporting notes. The outline itself is treated as a living draft: it can drag in multiple topic pages (keyboard, split keyboard, Moonlander), and linked sections like pro/con lists update automatically from the underlying sources. The result is a workflow designed for a chaotic brain: capture fast, link aggressively, process only when output is near, and rely on queries and templates to move from research (A) to production (B) without duplicating effort.
Cornell Notes
The workflow uses Logseq to separate durable knowledge from per-video drafting. Each topic gets a long-lived “topic page” that stores research (manual PDFs, URLs, customization notes, and where it’s used later). Midstream ideas are logged as TODOs and journal entries, then surfaced via queries so unfinished notes don’t get lost. For each video, a separate “content page” template tracks status (type, state, episode) and holds an outline that pulls in multiple topic pages. This structure reduces duplicated work, keeps references clickable, and avoids messy multi-pane editing by using sidebar cross-checking.
Why split work into “topic pages” and “content pages” instead of keeping everything in one place?
How does the system capture ideas that show up while researching or writing?
What role do Logseq folds and outliner navigation play in long-form writing?
How does the workflow handle the need to cross-check notes without cluttering the workspace?
What’s the approach to processing research notes—capture everything now or clean later?
How do queries and properties help manage multiple video drafts?
Review Questions
- How does linking TODOs from journal entries back to topic pages change what happens during later drafting sessions?
- What specific problems does the workflow try to avoid by using folds and sidebar cross-checking instead of many panes or a traditional TOC?
- Why does the system delay note “processing,” and how do clickable links compensate for imperfect organization in the short term?
Key Points
- 1
Use Logseq topic pages as long-lived single sources of truth for research, and keep per-video content pages as short-lived drafting spaces built from templates.
- 2
Capture midstream ideas as journal entries linked back to the correct topic page using TODOs, then rely on queries to surface unfinished work.
- 3
Treat Logseq as an outliner: use folds for structure and Alt + arrow keys for fast navigation through nested sections.
- 4
Avoid clutter from multiple side-by-side documents by using sidebar cross-checking and closing side panes once verification is done.
- 5
Organize emails with queries and break messages into blocks (quote blocks for quick reference or bullet points for granular linking).
- 6
Defer heavy note cleanup until output is near; clickable links and references let imperfect notes remain useful later.
- 7
Use content-page properties (type, state, episode) and queries to manage multiple drafts, and drag in multiple topic pages to build each video outline.