Stations and Lines: An Introduction to Nodes and Edges
why a subway map is built from just two kinds of elements — stations and the lines connecting them — and how a graph database organizes data around this exact same simple, powerful pair.
Part 1
Part 2
What Makes a Station a Station: Node Properties and Labels
why a subway station is more than just a dot on the map — it has a name, an accessibility rating, a category — and how node properties and labels carry this same descriptive richness in a graph.
Part 3
Not Every Line Runs Both Ways: Directed vs. Undirected Edges
why some subway lines genuinely run in only one direction while others carry traffic both ways, and how directed and undirected edges capture this same real, meaningful asymmetry in a graph.
Part 4
The Line Itself Has a Story: Edge Properties and Weighted Relationships
why a subway line carries its own attributes — travel time, capacity, fare zone — separate from the stations it connects, and how edge properties give a relationship its own genuine richness.
Part 5
Riding the Rails: Traversal and Why Graphs Answer Different Questions
why a rider planning a multi-stop journey thinks in terms of hopping from station to station along connected lines, and how graph traversal answers questions relational joins genuinely struggle with.
Part 6
Six Stops to Anywhere: Shortest Path and Degrees of Separation
why a rider wants the fastest route between two stations, not just any valid route, and how shortest-path algorithms find genuinely optimal connections through a graph's real structure.
Part 7
The Interchange Station: Modelling Many-to-Many Relationships Natively
why a major interchange station connecting a dozen different lines is the most natural thing in the world to draw on a subway map, and why graphs handle many-to-many relationships this same natural way.
Part 8
When Every Rider Looks Alike: Node Labels and Schema in a Graph
why a transit system still needs a genuine, if lighter-touch, way to categorize riders and stations even without rigid ticket-gate rules for every possible type, and how schema works in a graph database.
Part 9
The System Map vs. One Rider's Journey: Global vs. Local Graph Queries
why some questions genuinely require seeing the entire transit network at once while others only ever need to trace one rider's specific path, and how global and local graph queries answer these differently.
Part 10
Express Lines and Local Stops: Indexing and Traversal Performance
why a transit system runs both express services that skip most stations and local trains that stop everywhere, and how indexing shapes which parts of a graph traversal can move fast versus slow.
Part 11
A Line That Loops Back: Cycles and Recursive Relationships
why a circular subway line that eventually returns to its own starting station needs a genuinely different kind of route-planning logic, and how cycles and recursive relationships require the same deliberate care.
Part 12
The Property Graph vs. The Signal Map: Property Graphs vs. RDF Triple Stores
why a modern transit map with rich station and line detail differs from an older, minimalist signal diagram built from a strict, uniform pattern, and how property graphs and RDF triple stores reflect this same real difference.
Part 13
When the Whole Network Needs a Redraw: Graph Schema Evolution
why a transit authority adding a genuinely new kind of transportation to its network needs a deliberate plan for updating the map without breaking every rider's existing understanding of it, and how graphs evolve safely.
Part 14
Counting Passengers at Every Station: Centrality and Graph Analytics
why a transit planner needs a genuine, systematic way to identify which stations actually matter most to the network's overall function, and how centrality algorithms answer this same question for any graph.
Part 15
Spotting the Unofficial Shortcut: Community Detection and Clustering
why a transit analyst notices that certain groups of stations are used together far more often than the official line map alone would suggest, and how community detection algorithms surface these same hidden clusters.
Part 16
Not Every Trip Needs the Whole Map: When a Graph Database Is Overkill
why a rider making a simple, one-stop trip doesn't need the full transit authority's route-planning department involved, and why not every dataset with relationships genuinely warrants a graph database.
Part 17
From Track Plans to Ticket Sales: Migrating Relational Data Into a Graph
why converting an old, table-based ticketing system into a genuine relationship-aware network map requires more than just redrawing the same information differently, and how a real relational-to-graph migration works.
Part 18
Surveying the Line Before It's Built: AI-Assisted Knowledge Graph Construction
why a survey team can scan an unmapped region and propose where the stations and lines should genuinely go, directly from the terrain, and how AI-assisted knowledge graph construction works from raw, unstructured data.
Part 19
When the Passenger Just Asks Where to Go: AI Agents and Graph-Based Reasoning
why a modern transit assistant increasingly just asks a rider's destination in plain language and works out the route itself, and what this means for how graphs need to be built and organized.
Part 20
One Network, Every Journey
the stations and the lines, the interchange and the shortcut, every article's lesson reassembled into one coherent network capable of carrying every kind of journey the moment genuinely calls for.