Many-to-Many in an ERD Shouldn't Mean Hand-Building the Junction Table

Short summary

Schemity, a desktop ERD tool, auto-generates junction tables for many-to-many relationships instead of requiring the six manual steps typical in tools like SSMS. The post argues that junction tables are mechanical consequences of N:N relationships and should be tool-generated output, not human-assembled input. It also recommends using a composite primary key within the same bounded context but switching to a surrogate id with a unique constraint when junctions bridge contexts or grow lifecycles.

• •Schemity auto-generates junction tables with foreign keys, composite PK, and naming conventions from a single drag-and-drop gesture
• •Manual junction table construction in tools like SSMS involves six error-prone steps per relationship
• •Use composite PK within same bounded context; switch to surrogate id plus unique constraint when bridging contexts or when the junction grows a lifecycle

Generated with AI, which can make mistakes.