‘Futuristic’ Unison functional language debuts

Unison, a statically typed functional language with type inference, an effect system, and advanced tooling, has reached its 1.0 release status.

Announced November 25, Unison 1.0 marks a point where the language, distributed runtime, and developer workflow have stabilized, according to Unison Computing. Billed as “a friendly programming language from the future,” Unison is purported to bring benefits in compilation and distributed system development. With Unison, a definition is identified by its actual contents, i.e. a hash of its syntax tree, not just by the human-friendly name that also referred to older versions of the definition, according to Unison Computing. As a result, each Unison definition has a unique and deterministic address. All named arguments are replaced by positionally-numbered variable references, and all dependencies are replaced by their hashes. Thus, the hash of each definition uniquely identifies its exact implementation and pins down all its dependencies, according to the company.

The Unison ecosystem leverages this core idea from the ground up. Benefits include never compiling the same code twice and limiting versioning conflicts. Further, Unison promises to simplify distributed programming. Because definitions in Unison are identified by a content hash, arbitrary computations can be moved from one location to another, with missing dependencies deployed on the fly, according to Unison Computing. Unison can be viewed as a descendant of Haskell, with similarities including type inference and pattern matching, but is smaller and simpler than Haskell, according to a Unison FAQ.

Download and installation instructions can be found for Homebrew, Windows, Linux, and MacOS at the Unison website. Unison can be used like any other general purpose language, or used in conjunction with the Unison Cloud for building distributed systems. Unison code is stored as its abstract syntax tree in a database, i.e. the “codebase,” rather than in text files. Unison has “perfect” incremental compilation, with a shared compilation cache that is part of the codebase format. Despite the strong static typing, users are almost never waiting for code to compile, Unison Computing said. Unison’s hash-based, database-backed representation also changes how code is identified, versioned, and shared. The workflow, toolchain, and deployment model emerge naturally from the language’s design, enabling better tools for working with code, according to Unison Computing.