uid

UID stands for unique identifier, a label assigned to an object to distinguish it from all other objects within a defined scope. A UID is intended to be unique within that scope, allowing systems to reference resources, records, users, or devices unambiguously. Scopes can be local, such as a single database or operating system, or global, such as across Internet-scale systems.

Common forms include integers, strings, or structured alphanumeric sequences. Some UIDs are sequential; others are generated using randomness or timestamps to minimize collisions. A widely used category is UUIDs (universally unique identifiers) or GUIDs (globally unique identifiers), which are 128-bit values defined by standards such as RFC 4122. UUIDs come in several versions that use different generation methods, including time-based, name-based, and random, and are designed to be unique across space and time. They are widely used to identify resources in distributed systems, databases, and APIs.

In Unix-like systems, a UID is a numeric value assigned to a user account. The kernel uses real UID, effective UID, and saved set-user-ID to enforce permissions. The root user typically has UID 0. User IDs are stored in system files such as /etc/passwd or in directory services, with mappings to human-readable usernames.

In databases and applications, UIDs are often surrogate keys—opaque identifiers used as primary keys rather than meaningful data. Security and privacy considerations include avoiding exposure of predictable or sequential UIDs in public interfaces, which can enable enumeration or inference attacks.

Choosing a UID scheme involves trade-offs among simplicity, scalability, privacy, and collision resistance, depending on the domain and deployment.