“I want to know what a node actually represents in rule space”. If we analogize to networks in other applications (e.g., computers connected to Internet, cells in a body, or persons in a social network) then nodes can be born, die, and connect or disconnect to other nodes. In a fundamental physics context, I would expect nodes (under whatever specified conditions) can do the same. Links between nodes can represent communication lines that allow for passage of information from one node to another node. Along the lines of what happens in cellular automata, I would also suggest that nodes process a limited amount of information and can change states as a result.

“Why do [nodes and links] seem to be freely created and destroyed without cost?” Inflationary theories of cosmology suggest that a whole universe can be generated from nothing (or almost nothing). Similarly, Wolfram suggests we can start from very slim initial conditions and create a whole universe. Hence, there would be no material cost to copying existing nodes to create new nodes, but there is a time cost. Similarly, there would be time costs to processing information within a node, transferring information between nodes, or creating new links between existing nodes. I would conjecture: (a) The time cost of processing information within a node, is less than (b) the time cost of transferring information between nodes, is less than (c) the time cost of creating new links between existing nodes, is less than (d) the time cost of creating new nodes. Wolfram’s 2020 book ignores (a) and (b) and implicitly assumes the time cost of (c) and (d) are equal. Also, the 2020 book ignores the possibility that nodes might be in different states, or exhibit different behavior, depending on their own states or the states of their neighbors.

“When connections pop up between existing nodes that had no previous connection, how does a node ‘know’ how to form that new edge?” For a node (A) to form a link with non-connected node (B), I would suggest three requirements: There must already exist an indirect link via other node(s) (C or Cs). There must be a movable link that proceeds from node A to node B via nodes C or Cs. All the in-between nodes C or Cs must cooperate in passing the link onto B and B must accept the proffered link. Naturally, the time cost for linking up to a distant node will be larger than for a nearby node.

And what name do we use to refer to the situation before the first node is drawn? (or after the last one is removed).

I won't be surprised if the new Wolfram Model has some of the secrets.

But to be fair, we already know the answer is 42. But we have to figure out the question.

Just remember to bring your towel with you.

I'll get right on it. Should only take me a decade or two. No biggie.

Wouldn't it be funny if the fundamental rule was a computational expression of "Whatever makes a good story." Narrative becomes the fundamental concept at the heart of our universe? ;)