Why Trees?

If trees are so great, then why aren’t they being used now?

Because in order to make a tree-based search possible, somebody has to create a tree of everything. This is impossible.

Even if we create a big tree with millions of entries, such a tree is impossible to work with.

True, some big company can host a big tree & parcel out pieces of it to client websites using web services like REST, but the amount of traffic would be onerous.

Building trees from tags

The Blind Guy figured it out. Build many Trees from a site’s tags. This way, trees are auto-generated and only show what things are present in that website.

Since tags are names and are partially or completely universal, tags from many websites can be processed to create multiple trees.

The Blind Guy wrote a software machine that crunches content and counts tags at the site level, branch & each category of the original WordPress trees. The machine runs multiple times, generating ranked tags first, then, sets of top tags called Essences and using Essences, creating flavored trees, each based on a single important concept.

The result is a set of trees, each based on an important concept, that an automated librarian can offer people as they wander the many streams of content in a website, and later on, automatically select them.

Learning Trees

Trees based on an important concept, called Flavored trees, are paired with posts that have that concept as a tag. This is flavoring.

The Athena treeadds a rule that each link in a branch only have posts that contain the flavor & the previous tag link, each tree, the links, and each link’s posts reflect a facet of the flavor.
each post shares the context of the branch. This means that although a link may appear many times in that tree, each may have a different number of posts. This is called faceting, which transforms a branch into a Neural-like Pathway.

This brain-like pattern makes the tree brain-friendly, which enhances learning. This is why we call these automated trees Learning Trees.

Branch-centric

Traditionally, trees are link-centric. WordPress stores these links, relationships & posts in 3 database tables and use recursive algorithms called Walkers to build them. This works nicely for a single website, but is cannot translate between websites or the web itself. But branches built from tags can. Branches can be dynamically assembled with other branches to form other trees. They are also simple to render.

Dynamic Search

Generating many trees is not enough. There needs to be a way for an automated librarian to decide which tree to give you. By giving each tree an important concept, it will be possible for an automated librarian to decide which tree to open when your discovery goes down a different path. As of August 2024, we have not build this feature yet, but it will become critical for the MemeWeb.