Relationships

Content is captured in the form of pages.

Each page is a node in a graph.

Each page has a set of relationships to other pages.

Each relationship has a name.

Each relationship has a direction.

The most common relationship is "tag". Which is pretty vague but does contain direction.

Possible Relationships

• "vanilla extract" is FOUND ON the "shelf" // is this helpful for discovery/orientation?
• "sauna roof underlayment" is a post ABOUT THE "sauna"
• "sauna" is a TYPE OF "project"
• "sauna" is a MEMBER OF "building"

perhaps I can make relationships dynamically with x-attributes?

xFoundOn: 'shelf'

anything that starts with an x is a custom relationship. this way we can continue to build these up dynamically?

what about moving these into the body of the post? --> currently the "references" section captures this kind of body-linked relationship.

• is a
• is a kind of
• is a member of
• is a part of

Types

What is the goal here? Illuminate the taxonomy of the content captured.

examples:

• alcohol -> liqour -> rum -> light-rum
• person
• room

question:

• should the types map include members of the type? seems like yes!

Tags

Whats the goal here? we want to capture relationships between pages/posts/items.

ok. why are we capturing these relationships? what are we going to do with them?

I think the goal is discovery and orientation.

1. discovery: at best we will want to make new connections that surprise even the authors of the content

2. orientation: at best, we want to solidify the connections about the content that we are grasping at for deeper understanding and recall

Tags vs Named Connections

what if nothing was allowed something as vague as "tagged"? what if all connections/relationships had to be named? made more explicit?

Sketching Examples...

• "vanilla extract" is FOUND ON the "shelf" -> found on
• "sauna roof underlayment" is a post ABOUT the "sauna" -> about