Probably this was proposed before, but I wanted to share an implementation that I found in trillium:
To have this in Obsidian would be helpful to organize thoughts using small graphs ( < 25 nodes).
I think the nodes can (should) be the notes themselves that we drag into the canvas. We can refactor them visually; for example, if we rename a note in Obsidian, this should also be renamed in the diagram. Perhaps this can be extended to draw groups or connect MOCs instead of just notes
Ideally, data could be stored in the markdown itself, to keep it forever simple 
You could check out the Juggl plugin for now if you want
Oh yes! I love juggl, and I use it to make sense of the notes.
However I was thinking about something super lightweight to author few nodes (<25)
I’ve been hunting for this exact tool. Juggl is really close and actually could do it if its easier to pin items and lock them in place, resize… i may also just not be proficient enough yet to make these sort of changes.
I am also looking for a feature like that. It would be amazing if one could just link another note and add the type of relationship like: [ [Test Note1] ] (is_part_of).
Check out ExcaliBrain by @zsviczian. It is a beta release, available via the BRAT plugin. I know this sounds hyperbolic, but it is a game changer for sense-making graphs. The interface will feel familair to those who’ve ever use TheBrain. The link to this video is only a few days old and Zsolt has already added many new features. Follow him on Twitter to keep up with ExcaliBrain’s development. Obsidian ExcaliBrain - Beta Release v0.0.1 - YouTube
Loving the look of the ExcaliBrain plugin. I’ve used theBrain off and on since the old Personal Brain days. I’ve never been able to let go of that program completely, no matter how many friction points I encounter, because I’ve found the Plex so visually appealing and easy to navigate. Nothing’s ever really compared… until now. Can’t wait to see how it develops further, but even in this early beta stage it does enough of what I need!
Check out the latest video and update from @zsviczian. I know this is cliché, but the possibilities are endless with so many ways to view data: links, virtual notes, inferred nodes which include siblings and backlinks, folders, tags, visual customization of all the above, and visual customization of the kind of Parent, Child, and Friend. I was demonstrating how a Parent node can be something other than a Parent to someone yesterday and found that grasping alternate Parent, Child, and Friend categories was easier when I talked about the gates rather than using Parent, Child, and Friend as nomenclature.
There are top-gate relationships, side-gate relationships, and bottom-gate relationships.
For instance, a Parent thought connects at the top gate, but relationships called based on, author, and source can all be top-gate relationships, too. Similarly, you can have Children with connections from the bottom-gate, but you could also have consists of, supports, and leads to coming from the bottom-gate.
It blasts the door wide open to different contextual paradigms for different types of data. It also greatly increases the usefulness of both the daily note, especially when they get lengthy. YMMV, of course, but I think about what happens after writing a lot of unlinked atomic notes sometimes as dropping them into a well. They’re there but in a pile at the bottom of a well. The visual feedback from Excalibrain at minimum shows from which daily note they started, assuming you use daily notes this way. It is like a webbing that catches atomic notes on their way down the well. I know this is a weird metaphor, but it was always hard to explain to people why TheBrain software isn’t really a mindmap as much as it is an alternate way of navigating, connecting, and developing thoughts. Excalibrain has now surpassed, IMHO, the usefulness of the TheBrain, but it comes at the price of choice-complexity that some might find overwhelming.
