What is Tauchain (Tau)?


Tauchain (Tau) is an open source, decentralised network with a governance mechanism that uses user discussions to create a scalable, self defining and dynamic system. Tau’s large-scale discussion platform can detect consensus points or disagreements between all users and execute or self-implement the network instructions in real time . Agoras token will leverage Tau’s technology to have an economy governed by its users. Agoras tokens aim to enable markets for knowledge, computational resources, and derivatives

What Problems is Tau Solving?

Tau solves societal challenges with decentralised technology:


  • The problem of scaling discussions and decision making (governance)
  • Collaborative discussion & decision making about a desired computer program, to then be translated into code.
  • Agoras solves the inability to directly monetize knowledge. It is nowadays impossible to directly monetise a piece of knowledge. We monetise knowledge only indirectly by being recognised as experts in some fields.


  • A program in which its requirements and code are determined by its many users.
  • The theoretical nonexistence of one universal language to be suitable for all types of expression.
  • The Halting Problem or more generally Rice’s theorem, which demonstrate that reasoning over computer programs is mathematically impossible in most cases

To improve the structure of online discourse, history has shown that great tech advances include two things. The efforts of people, being able to build on each other’s work and tools to make that possible.

So for example, today if you are going to the discussion forum to find a particular piece of information or you wanted to fact check something that someone said. Chances are it’s probably cumbersome / difficult to do and once you are done no one can easily build on your work.

Tau, by using a self defining and decidable logical framework, empowers people to structure and tailor information and discussion environments to what they wish and allows users to build on each others work. Giving the ability to scale collaborative endeavors between thousands of mathematicians, doctors, computer scientists, engineers and other experts to greatly accelerate the production of knowledge.

Anyone can interact with the tauchain network and participate in this digital economy, without the need for third parties and without the risk of censorship.

Tau uses a decidable logical framework which allows it to synthesize and execute code based on user specifications. Users can create their own teams under Tau to discuss and collaborate on a project with its own private knowledge base. After taking advantage of Tau's ability to scale discussions, teams can quickly reach a consensus on the specifications and Tau can automatically generate and execute code, according to those specifications.

What is Tauchain?


  • Ohad Asor
  • Zennet.sc
  • Old Tau: Neglected the social side and was also mistakenly relying on an undecidable logic, with decidability being a main requirement.
  • New Tau


  • Knowledge market and economy, Agoras, knowledge 2 cash transactions (the ability to) maufacture, ship and price knowledge)
  • Monetisation of formalised knowledge
  • Computational resources market
  • Advanced currency
  • logical solver
  • internet of languages
  • TML : A fully declarative logic programming language that satisfies 3 requirements for the language: decidability, non monotonic logic, self define-ability https://youtu.be/8QZDYIO1EGA?t=1085

Technical and societal aspects

Decentralised resources

  • Internet of languages.
  • Discussion platform featuring machine-comprehensible languages.
  • Further those languages are decidable, so in particular detection of agreement and disagreement, is solvable.
  • Converting a discussion about a program into actual program code.
  • Risk-free interest without inflation, by featuring a non-leveraged derivative market, and then implementing zero-delta portfolios.
  • the ability to fairly rent (and rent-out) computational resources, under acceptable risk in the user's terms (as a function of cost)
  • featuring a decentralised search engine that, to some probabilistic extent, ensures that the index contains the correct knowledge (isn't modified, replaced, or omitted).
  • using a pricing formula to eliminate the risk of mispricing, a risk that can be significantly exploited.

Ohad's Pricing Formula

  • probabilistically verify unverifiable computations by calculating the same thing more than once (by randomly choosing more providers), so increasing the cost linearly, decreases the risk exponentially (e.g. x10 more cost yields¹⁰ less risk)

Tau Meta-language (TML) Overview

This is an overview. If you'd like to learn more
Read about TML, in detail, within the whitepaper


  • There can not be a single universal language as no one language is optimal or adequate for all needs.
  • Nobody knows how machines can use human languages, therefore tau can't use natural languages but machine comprehensible languages, until someone in the future will develop such an ability, if any.
  • Languages can be evolved and become more and more human-convenient over time, since Tau leaves the languages end open to development and change. It is not only a self-defining program to evolve according to its users, but even the languages it supports are self-defining and self-evolving by its users.
  • There are many formalisms of natural language that are close to natural language and comfortable for humans to work with (what we refer to "simple enough English that machines can understand")


  • TML is intended to be a compiler-compiler, to allow translation from one language to another. To define a new language would mean to define how it translates from an existing language. by that the choice of language of expression (among the languages to be available over time in the internet of languages), becomes irrelevant.
  • Machines need to understand the meaning of what we say, but computers expect operational information, while humans use declarative language. In Tau people focus on the "know-what" and machines figure out the "know-how". Tau uses formal logic as it is natural to humans and something machines can work with (cf. "Knowledge Representation and Classical Logic" by Lifschitz et al)
  • A meta-language that can define new languages and is able to redefine itself and change, a self-amending language
  • Users define new languages by specifying logical formulas to describe what it means for two documents in different languages to have the same meaning. To define a new language, one needs to define how it translates into an existing language while preserving semantics.
  • Tau uses Partial Evaluation to prevent considering the logic of the language(s) again and again with every compilation of documents written in it. This adds desirable features for a compiler-compiler, in the form of Futamura projections.
  • Binary decision diagrams used for performance of logical solver.

Read more about the concepts behind
TML and Logic Programming

Use TML.bot : Get first hand TML experience with an easy-to-use guide to interact with the TML bot

The Internet of Languages


  • users define languages over the internet of languages. Users can express knowledge and opinions in these languages in order to communicate.
  • With Tau users express themselves in decidable machine-comprehensible languages. This facilitates Human-Machine-Human communication where Tau organizes what we say and is able to do so since we encode our information in a way accessible to it.


  • IoL is referenced as the infrastructure
  • <insert here the arguments for the language of law, and the argument for the meta language>



  • Gather knowledge and agree or disagree over it, and perform automated actions arising from the discussion
  • a decentralized program that everyone writes, a program that is waiting for its users to tell it what to do. That program can be the rules of the blockchain itself, effectively democratizing the rules of governance at the most fundamental level allowing all users equal rights


  • Tau doesn't guess the people's opinion. that's one reason we use logic, not ML for example
  • Why the logic was chosen with regard to partial fixed point: Tau Chain allows users to change the choice mechanism of itself by having clear rules of changing the rules, to change Tau's code with time recursion in order to deal with rules of changing the rules. This is an important aspect involved in the choice of fixed-point logic for TML, and λY calculus on the Beta (apropos, Bauer showed in "On Self-Interpreters For System-T and Other Typed λ-Calculi" that a language can self-interpret only if it has fixed point, which rules out total programming languages)
  • Tau uses PFP logic known from Finite Model Theory with PSPACE-complete expressiveness, decidability and self-defining properties (Imhof, 1999 "Logics that define their own semantics")



  • governance model, decided by the users
  • choices collaboratively be made over the system, are about the system itself. Tau, is a discussion about Tau.
  • users decide who can change the rules (in a decentralized setting)?


  • a collaboratively self-amending program that can be anything or many programs at once
  • Over Tau many drafts that propose Tau's next full code can be submitted by users. Using the logical formalism of these documents Tau can calculate the precise core that everyone agrees on, and list the points to be resolved. Users discuss the future Tau and the opinions map arises from the conversation, determining the next Tau.


This is an overview. If you'd like to learn more
Read about Agoras, in detail, within the whitepaper

  • Knowledge economy, named after the Ancient Agora of Athens)

Knowledge market


  • consists of users that execute searches and simultaneously maintain the data
  • home and institutional users use and support the infrastructure by running a client. costs usually even-out, though heavy users pay more, heavy maintainers earn more


  • utilising the decentralised search engine and data storage

Computational resources market

  • maintainers index the knowledge on the web, charge a fee for answers on queries that depends on the amount of usage and the cost of maintenance

Derivatives market (newly designed economy offering features like risk free interest without printing new money)

  • advanced monetary system
  • the monetary features that Agoras will support
  • ethical value systems in the broad sense, defining good and bad, better and worse as an economy without some subjective valuation or "utility function" can't exist

Business/use cases

Tau can be anything that its users want it to be, as they ultimately collaboratively determine the very code of the network. Examples:

automated scheduling

  • from app demo on idni.org

scaleable decision making

  • we cannot ever vote on whether something should be voted on in the first place
  • theoretical inability to comprehend the consequences of proposals which could compromise the security of the system
  • social choice mechanism where small or very large group of people repeatedly reach and follow agreements. A chronological and conceptually bottom-up order about the creation of a knowledge society.

    • language transmitting one idea between two people - formalized in a machine comprehensible language - allows for easy explaining, easy understanding, and produces formalized knowledge.
    • knowledge limits that affect flow and processing of information collaborative theory formation Ideas that are formalized in a machine comprehensible language allows users to translate it to other knowledge representation languages, to reorganize it or to compare it to other formalized ideas. From this Tau gains the ability to answer questions about an idea in a decidable manner, without the need to refer the question to the original idea's author
    • discussion collaborative decision making methods of discussions so far suffer from very poor scaling voters go over equally-weighted one million proposals daily
    • collaboration
    • Scaling limits
    • discuss, share and organize knowledge, detect consensus and disagreements, and coordinate actions in forms of programs
    • automatically detect repeated argument by same person
    • collecting what each person said during the discussion
    • map all the agreement and disagreement points
    • list all opinions and then who agrees with them rather (speakers per opinion rather opinions per speaker)
    • organize the information put on the discussion in more organized and readable forms like a wiki
    • comment automatically and automatically express your opinion, based on the information you provided in the past, and relative to the post you're autocommenting to.
    • calculate the set of statements agreed by everyone, network wide, per team, per profiles connected to the users profile, per discussion, and so on.
    • change (choice about choice)
    • opinions flow through certain pipes and reach the voting stage almost empty from the vast information gathered in the process
    • everyone has an equal right to propose what to vote over
    • If rules can change themselves, they inevitably contradict themselves as they try to say something else. This is formalized in a paradox-free manner using recursion.
    • knowledge market
    • based on agoras'
    • hardware rent market using Agoras' decentralised search engine
    • a knowledge economy, on top of the knowledge society
    • using decentralised search engines, data storage and computation

Structured knowledge

  • Similar to wolfram alpha?

knowledge economy

  • based on agoras' knowledge market
  • lives on knowledge society-framework Tau
  • an economic system that supports the production, pricing, shipment, and arena (agora), of some knowledge (the knowledge of the text on the web)
  • deeper and more meaningful knowledge than existing static searched based on thesaurus and ML that has no real understanding of the subjects

Edit this page on GitHub