Formatto

Formatto is a hypothetical data interchange format designed for efficient serialization of structured data in both human-readable text and compact binary representations. It is intended as a learning tool and reference for comparing serialization strategies, rather than a widely deployed consumer technology.

The core data model includes primitive types such as booleans, numbers, strings, and null, along with composite types like arrays and maps. Formatto supports an optional nominal type system and the use of explicit type annotations to improve interoperability across programming languages.

There are two encoding forms. Formatto-Text resembles a simplified, typed subset of JSON with optional tags and comments, while Formatto-Bin uses a compact, binary encoding based on a length-prefixed scheme to enable streaming and incremental parsing. The binary form emphasizes space efficiency and fast deserialization, suitable for low-latency applications.

Schemas provide validation and versioning capabilities. A Formatto schema defines object structures, field names, and allowed value types, with rules for optional and default fields and strategies for forward and backward compatibility as schemas evolve.

Implementations and usage. As a hypothetical format, several educational libraries exist in languages such as Python, Java, and Rust to illustrate parsing, schema validation, and streaming. Formatto is not widely adopted in production systems, but it is discussed in academic contexts and in comparative studies of serialization formats.

Related formats include JSON, Protocol Buffers, MessagePack, and YAML. See also discussions of data serialization, schema evolution, and cross-language data exchange.