When should I use Pattern versus a plain graph library?

Use Pattern<V> when your data is inherently a decorated sequence — when what you have is a value accompanied by an ordered list of sub-parts, and that structure matters.

Good fits for Pattern<V>:

• Graph elements you are constructing or transforming: nodes, relationships, walks as first-class values that you build up, map over, or serialize.
• Structured notation: Gram documents, formatted text, nested structured data where the shape carries meaning.
• Custom value types: You have a type V and you want to compose values with sub-values while preserving structure through transformations.

A plain graph library is a better fit when:

• Your primary operation is traversal and query over a large connected graph (e.g., shortest path, reachability, community detection).
• You need index structures over millions of nodes for fast lookup.
• The graph is mutable in place with concurrent writers.

StandardGraph bridges both worlds. It stores Pattern<Subject> elements in a queryable structure indexed by identity. Use it when you need to look up elements by ID, query by label, or retrieve relationships from both endpoints. It is not a general-purpose graph database — it is a structured container for patterns that you have already built.

If you are parsing Gram files and building in-memory graphs for processing, StandardGraph is the right tool. If you are running graph algorithms at scale, reach for a graph database or a dedicated graph algorithm library.