🌿 Alternef Digital Garden

Information Economy Meta-Language (IEML)

5 min read

Edit on GitHub

IEML (Information Economy Meta-Language) is an open-source artificial metalanguage designed to represent the semantic content of linguistic signs in a computer-readable way. It was developed by Pierre LĂ©vy as part of his work on collective intelligence, aiming to encode meaning in a way that is both computable and aligned with the expressive power of natural languages.

About Pierre LĂ©vy

Pierre Lévy is a philosopher and pioneer in the study of the impact of the Internet on human knowledge and culture. His seminal work Collective Intelligence: Mankind’s Emerging World in Cyberspace (French 1994, English 1999) described a kind of collective intelligence that extends everywhere and is constantly evaluated and coordinated in real time—a collective human intelligence augmented by new information technologies.

  • Canada Research Chair in Collective Intelligence at the University of Ottawa (2002-2016), where IEML was developed with over $1 million in federal funding
  • Adjunct Professor at the Literature and World Languages Department, UniversitĂ© de MontrĂ©al
  • Co-founder and CEO of INTLEKT Metadata Inc.
  • Fellow of the Royal Society of Canada

Lévy’s research focuses on reconciling artificial intelligence with collective intelligence, and exploring the intersections between philosophy, science, religion, and technology.

Philosophical Context

IEML can be understood as a modern realization of historical ambitions in symbolic logic and language:

  • Ramon Llull’s Ars Magna: The medieval combinatorial system that inspired computational approaches to knowledge
  • Leibniz’s Characteristica Universalis: The dream of a universal symbolic language that could express all human thought and enable logical reasoning through calculation

While IEML is not a complete calculus ratiocinator (it does not possess a true “alphabet of thought”) or a philosophical language per se, it builds on these foundations using modern understanding of computer semantics and philosophy of language.

Core Concept

The goal of IEML is to make real-world data machine-readable by proposing a standard representation that enables the mapping of semantic representations with data in a computer-friendly manner.

IEML’s design starts with a small set of primary concepts (semantic primitives) arranged in a matrix, which are composed together to create more complex concepts. This process repeats recursively, allowing increasingly complex concepts to emerge through a fractal, matrix-based design. This architecture makes representations:

  • Easy to manipulate
  • Quick to calculate semantic distances between concepts
  • Simple to encode

Unique Properties

As of 2021, IEML is the only metalanguage that combines all of the following properties:

  1. Fully Regular Grammar: A completely regular and recursive grammar with approximately 3,000 dictionary words organized in paradigms (systems of substitution)
  2. Self-Defining: With each concept automatically determined by composition and substitution relations, IEML serves as its own metalanguage
  3. Computable Semantics: Not only syntactically computable but also semantically computable—semantic relations are computable functions of syntactic relations
  4. Natural Language Expressivity: Has the expressive power of a natural language with an algebraic structure
  5. Web Standards Compliance: Complies with Web standards and can be exported in RDF

USLs: Uniform Semantic Locators

IEML expressions are called USLs (Uniform Semantic Locators). They can be read and translated into any natural language. The dictionary has been translated into French and English and could translate any natural language.

The Semantic Sphere

The semantic sphere brings together all possible texts in the IEML language, translated into natural languages, including the semantic relations between all texts. On this conceptual playing field:

  • Dialogue and intersubjectivity arise
  • Pragmatic complexity emerges
  • Open games allow free regulation of categorization and evaluation of data
  • Ecosystems of ideas representing collective cognitive processes can be cultivated in an interoperable environment

Applications

IEML is intended for use across multiple disciplines:

  • Artificial Intelligence: Foundations for a new generation of AI with computable meaning
  • Business Intelligence: Unified semantic layer across disparate data sources
  • Data Science: Semantic interoperability between databases and disciplines
  • Digital Humanities: Heritage conservation and cultural knowledge representation
  • Digital Communications: Enhanced metadata and content discovery
  • Knowledge Management: Reflexive collective intelligence systems

Relationship to Semantic AI

IEML is closely related to Semantic AI in that it serves as a foundation for semantic computation and semantic interoperability between systems. It is part of a neuro-symbolic architecture (specifically a neuro-semantic architecture) that emphasizes solving the problem of semantic computation.

IEML addresses the challenge of semantic interoperability among databases, languages, and disciplines by providing:

  • Unambiguous and computable semantics
  • A concept coding system that reconciles maximum originality and complexity with interoperability
  • A coordinate system for a common knowledge base feeding both automatic reasoning and statistical calculations

This fulfills the promise of the Semantic Web through computable meaning and interoperable ontologies.

A Project for a New Humanism

Lévy positions IEML as more than a technical tool—it is “a project for a new humanism.” By enabling reflexive collective intelligence, IEML aims to augment human collective thought in the algorithmic medium, fitting new institutions to the new technological environment.

External Resources

Backlinks

Graph View