You'll see this acronym crop up a lot in our documentation - DAG stands for 'directed acyclic graph'. Yes, this is a bit of computer-science jargon, but it's also a crucial concept that's becoming more relevant in data analytics, and is already important in data engineering. Let's break it down:
Directed - the only reason a pair of cells can be related is that one cell is influencing the results of another, and this influence must be in one direction only.
Acyclic - it is impossible to construct a cycle of dependencies in a DAG (and therefore Count). For example, if you write two cells which look like
a: select * from bandb: select * from a, both cells will display an error.Graph - this is just the generic name for a bunch of boxes with arrows between them (apologies to any reader with a formal background in graph theory).
The DAG is represented in Count by the colored arrows which appear between cells. You don't have to think about it, Count will construct and manage the DAG automatically, but it's very helpful to be aware of the relationships between cells as those determine how cells execute in response to changes.