Obsidian Bases: Obsidian's Biggest Upgrade (Complete Guide)

Obsidian Bases’ biggest upgrade isn’t just faster organization—it’s the way it turns note links into live, contextual “dynamic links” that update as soon as you click into a note. That shift changes how people navigate an “idea verse”: instead of browsing a static database or scrolling a backlink list with no context, Bases can surface the right subset of related notes (movies, quotes, people, projects, calendar entries, and more) in a sidebar, sorted and filtered to match the note you’re currently viewing.

Speed is the first practical win. Creating a new base via Command-P (Mac) or Control-P (Windows) is described as effectively instantaneous, with built-in optimizations for scrolling large tables. The workflow is tested against vaults with tens of thousands of files, and the experience remains smooth while moving through large result sets.

The second major feature is that a single base can hold multiple views—table, cards, and more—each showing the same underlying dataset through different lenses. In the “sources” example, one base contains pre-built views for books, shows, and other categories. Clicking between views immediately changes layout and filtering without rebuilding anything.

The third feature is live editing with autocomplete, letting users rapidly update properties (like ratings) and see those changes propagate into the base’s sorted results right away.

Under the hood, Bases rely on note properties. Properties live inside individual notes (for example, a “book” note can store metadata such as year, rating, and other fields). Bases then use those properties to create columns, sort order, and filters. The transcript emphasizes that properties are the control panel: once they’re present, Bases can filter and present notes that match specific parameters.

Views are where the system becomes usable. A “main view” can apply a global filter (such as “folder contains sources”), and then each duplicated view adds its own narrower filter (such as “folder contains …/books”). This structure keeps shared constraints consistent while letting each view focus on a different slice of the vault.

The most transformative capability is embedding views via Obsidian syntax and “dynamic links.” When a note includes an embedded reference to a specific Bases view, the embedded results auto-update based on the note you’re currently in. The transcript demonstrates this with directors like Peter Jackson, Hayao Miyazaki, and James Cameron: clicking between director notes updates the sidebar to show only the movies in the database that match that director, sorted by personal rating.

The walkthrough then scales up into a full personal system (“Ideaverse Pro 2.5” and related bases). Bases are organized around an “ACE folder framework” (Atlas, Calendar, Efforts, X/Extra, plus an Add folder). High-level bases (like “Atlas,” “Dots,” “Maps,” “Sources,” and “Projects”) are intentionally broad, while views inside them zoom in by note type—people, questions, quotes, statements, and “dots” (timeless knowledge units). The result is a navigational map where clicking a note reveals connected context across time, projects, and collections.

Practical automation also appears: Claude Code is used to populate missing book cover images by finding notes with an “image” YAML property and then searching online (via a Google Books API approach) for covers, ultimately updating dozens of book notes.

Finally, the transcript argues that Bases make links more meaningful rather than replacing writing. Dynamic links and contextual sidebars are framed as “echolocation” for memory—reinforcing relationships over time—while guidance urges users to start small (10–20 notes) and expand gradually, rather than trying to retrofit properties across thousands of notes at once.