Which statement best captures why a graph fits connected data better than a table?

Graphs let new relationship types appear without redesigning the schema.

Graphs use less memory than tables.

Graphs let new relationship types appear without redesigning the schema.

Graphs are always faster to query.

In a graph, where does the *meaning* of a fact live?

On the label of the edge connecting two nodes.

Meaning is implicit and never written down.

You add a brand-new kind of relationship between two existing entities. What changes in a graph database vs. a relational table?

The graph just gets one more edge with a new label; the table needs a new column or join table.

Both require a schema migration.

The graph just gets one more edge with a new label; the table needs a new column or join table.

Only the graph needs a migration.

Nodes, Edges, and Meaning — Semantic Web Academy

Theory

A graph is two things:

Nodes — the things you talk about (people, books, cities, ideas).
Edges — the typed relationships between them.

A table forces you to know every column up front: add a new kind of relationship and you have to alter the schema, migrate the data and redeploy the app. A graph has no fixed columns — every new fact is just one more edge with a label. The shape of your knowledge can grow freely as you learn more.

Key idea: in a graph, new relationship types can appear at any time without redesigning the schema.

Here is the contrast at a glance:

Aspect	Table (relational)	Graph (RDF)
New relationship	Add a column or join table	Add one more edge with a new label
Schema up-front?	Yes — every column must exist	No — facts can show up any time
Meaning lives in	Column headers + foreign keys	The label on the edge itself
Cross-source merge	Painful (schema reconciliation)	Natural (shared IRIs glue graphs)

Analogy

A family tree is a graph: people are nodes, parent-of and married-to are edges. Try fitting all of human kinship into a spreadsheet — you can't, because new relationship types keep appearing. Graphs welcome new edges; tables resist new columns.

This is the Semantic Web superpower: because edges are data (not hard-coded columns), two strangers' graphs can be merged by simply taking the union of their triples — no schema negotiation required.

Visualization

LayoutLabelsClick a node to focus its neighbourhood · drag to pan · scroll to zoom

A 4-node social graph. Notice how the meaning lives on the edges.

Reflect

Pick a domain you know well (your job, a hobby, a TV show).

Doing this exercise out loud is how graph-thinking becomes muscle memory: you start to notice which facts are really edges hiding inside columns. The fifth bullet is the most revealing one — it usually surfaces the relationships your current data model is silently dropping.

›List 5 entities you'd model as nodes.
›List 5 relationship types you'd model as edges.
›Which relationships would be *painful* to model in a spreadsheet?

WatchOptional

Reading in progress · 0 of 3 activities done