Getting started with ExcaliBrain - starting from an empty Obsidian Vault (Part 2 of 3)

Excalibrain’s relationship graph becomes truly usable once link styling and ontology visibility are aligned—so the “remote work policy” node doesn’t just connect to other nodes, it clearly shows what those connections mean. The session starts by revisiting Excalibrain’s ontology and link styling settings, then turns on label and arrow display so the graph communicates direction and relationship type at a glance. A key fix is changing the base link style so ontology information (like the “position” type added earlier) appears on the links, and adjusting arrowheads to reduce confusion about which side is the start versus the end.

Arrow direction is handled through two approaches: Excalibrain can use an inverse arrow direction by default, or users can keep that logic and instead switch which arrowhead is set to an arrow. In practice, the “remote work policy” node ends up with arrows pointing toward the two connected nodes that represent positions, with the direction tuned to match the user’s internal reading order. The session also highlights a practical navigation feature: embedded mode. When embedded mode is enabled, the connector between the leaf note view and the Excalibrain view disconnects; turning embedded mode off reconnects it. A separate “synchronize navigation” toggle in Excalibrain behavior settings controls whether the two navigation controls act together or independently.

The workflow then shifts to managing embedded nodes and theme consistency. The “remote work policy” node is selected, embedded mode is turned on, and the user observes how the connector behavior changes. Next comes a styling issue: embedded markdown nodes follow the Excalidraw theme by default. In light mode, embedded nodes look correct; switching to dark mode without matching settings can produce ugly colors. The fix is found in Excalidraw settings—specifically the option “markdown embeds to match excalidraw theme.” After toggling that setting, the user may need to close and reopen Excalibrain for the change to apply.

With the graph readable and the interface stable, the session builds the argument structure. The supporting argument relationship is added to the ontology using Excalibrain’s “add data view field to ontology” command (via the command palette). After the ontology updates, “supporting argument as an ontology” appears on the relevant link, and a new markdown document is created as a supporting argument node. That document is formatted as an argument with an “argument” tag and a question: “how can remote work improve employee work life balance?” The question itself is then added to the ontology as a “next” friend type so it appears on the right side of the graph.

Finally, the session polishes the visual semantics. Supporting argument links are reformatted to be green and thicker (set to 3px), while objecting arguments are intended to be red (set up as the desired color scheme). The result is a graph where “remote work policy” connects to a clearly labeled supporting argument, which contains a question node about work-life balance. The session closes by previewing the next step: adding evidence, including web links, and embedding those links into the graph with further formatting and settings in Part 3.