YAML in Obsidian - A Prerequisite For Dataview

Obsidian’s Dataview plugin becomes far more powerful once notes are annotated with custom metadata—and YAML is the key mechanism that makes those annotations both readable and queryable. Dataview automatically generates many metadata fields (like file creation date, file size, title, and path), but relying only on built-in metadata limits what can be searched and filtered. YAML lets users attach their own structured metadata to individual notes, so Dataview can query using parameters the user chooses rather than only what Obsidian provides by default.

The transcript frames metadata as “data about data,” explaining that digital files already carry system-generated metadata (such as resolution, file type, and geolocation when available). In Obsidian, Dataview similarly creates automatic metadata for each note, but users can also add their own—like a star rating, genre, or tags—directly inside the note. That custom layer is what unlocks Dataview’s flexibility. YAML is presented as a human-friendly way to embed that metadata, comparable in purpose to JSON or XML but easier to write and interpret.

A practical walkthrough follows, starting with an Obsidian setting: enabling “show front matter” in the editor settings. YAML metadata in Obsidian is added as a “front matter” block at the very top of the note, delimited by three dashes, then another three dashes. Inside that block, key-value pairs define metadata fields. Examples include tags: evergreen notes, alias: our yaml presentation, and type: youtube. When the YAML header is valid, Obsidian recognizes fields like aliases—so searching for “our yaml presentation” surfaces the note—and preview mode confirms that the structure is being interpreted correctly.

The transcript emphasizes YAML’s strictness, which is also its benefit. YAML must be the first content in the note; even a single character typed above the YAML block breaks recognition. Spacing matters: removing the space after a colon (for example, writing “tags:evergreen notes” instead of “tags: evergreen notes”) triggers an “invalid” YAML warning in preview mode. For list-like fields such as multiple tags or multiple aliases, items must be separated with commas. The speaker also notes an alternative formatting approach: using indentation to represent a list, where each item is preceded by a space to maintain valid YAML structure.

Finally, the transcript clarifies why some fields appear under Obsidian’s “metadata” area while others don’t. Obsidian natively surfaces certain metadata types—specifically tags, aliases, CSS class, and publish—so those show up in the metadata panel, while other YAML keys (like type) remain in the YAML header display. The payoff is threefold: YAML flags mistakes immediately via “invalid YAML,” keeps notes cleaner than scattered annotations, and enables Dataview queries based on user-defined parameters. The next step—Dataview usage—is positioned as the natural follow-up once YAML annotations are in place.