Hierarchies Considered Harmful

[wagle]

I've spent half my life failing to shoehorn my mind into hierarchies. It. Does. Not. Work.

I often have "ten" different reasons to do anything. There is no preferred "parent". Making the wrong adhoc shoehorning decision makes the child inaccessible to the other "parents", and it is lost.

In real life, I have two parents, each having two more parents, ad infinitum. I have one sibling, and we might each have 1 spouse, having their own ancestry, 2 children, again ad infinitum.

I was attracted to obsidian after a very long search because it represented my DAG's, which are huge and so hard to reason about globally. Many people appear to have no problem shoehorning their minds into hierarchy/outlines, and so have no problem insisting that I do the same. It can't be done.

I'm looking into juggl, which seems to do DAG's, but it uses breadcrumbs, so I'm trying to do multiple parents in it.

I don't see the glory in reducing a DAG to a TREE. Please explain?

[wagle]

For example, in your README.md at https://github.com/SkepticMystic/breadcrumbs, you say:

Frontmatter field. This is a field in the yaml format at the very top of your note. For example the me.md file could have the following frontmatter:

---
parent: [[dad]]
parent: [[mom]]
sibling: [[sister Lara]]
child: [[child Andre]]
---

Note the multiple parents?

[wagle]

I do seem to be trying to impose a dag (instead of a tree) on the cyclic obsidian graph.

[neyer]

Maybe I'm confused here, how are DAGS not hierarchies?

I agree that trees aren't great, since they're too inflexible. But i'd consider a dag to be hierarchical in that if you 'keep going up' you'll end up finding some root or set of roots.

[wagle]

Hierarchies are trees. Each non-root node has a unique parent. If a node is a task, then there can be only one reason to do it. In a DAG, you can have the same task serve multiple purposes.