fileoriented

File-oriented refers to an approach in computing that treats data primarily as files stored on a filesystem, rather than as entries within a database or as objects managed by an application. In file-oriented systems, data is accessed by path, filename, and file metadata; manipulation occurs through standard file I/O operations, typically reading, writing, and appending to files. This model emphasizes the organization of data into discrete, self-contained units.

Historically, many software tools and data interchange formats are file-oriented, such as plain text, CSV, JSON, and XML. Version control systems, log collectors, and archival storage often follow file-oriented workflows. By contrast, database-oriented approaches rely on structured schemas and query languages to access data, while a file-oriented approach may constrain querying and integrity to the logic of the application and the format of the files themselves.

Characteristics include reliance on filesystem semantics, directory hierarchy, and file metadata; data is typically serialized in human- or machine-readable formats. While simple and portable, file-oriented designs can lead to scattered data, duplication, and weak cross-file consistency.

Advantages include simplicity, portability, ease of backup through file copies, and human readability. Disadvantages include limited transactional guarantees, lack of built-in querying, potential redundancy, and challenges in enforcing consistency across many files.

Use cases include configuration files, logs, data exchange payloads, and archival datasets. See also file system, flat-file database, document store, and relational database.