Building a Smart Constitution on To: Answers to Your Questions

Table of Contents

1. Introduction
2. Building a Smart Constitution on To
3. Town Language Demo Release
4. Hyperdimensional Computing and To
5. Delivering Tangible Results
6. Comparison of AI Agents in Town and Aolis
7. Disruption of Software Engineering by To
8. Social Disruption of To
9. Controlled Natural Language Interface
10. Test Net and Main Net

Introduction

Welcome back to the January update from the Team! In this update, we will address some of the most frequently asked questions (Q&A) from our community members. So let's dive right into the questions and provide you with the answers you've been waiting for!

Building a Smart Constitution on To

One of the community members, Danna, asks if it would be possible to build a smart constitution on To based on what AAD (Autonomous Agreement Discovery) says about the laws of making laws. The short answer is, yes! There is no reason why we cannot build a smart constitution on To. However, there is a tradeoff between expressiveness and decidability. The more complex and expressive we make the language, the less decidable certain things become. But within these limitations, building a smart constitution on To is definitely feasible.

Town Language Demo Release

Andrew is eager to know when the Town Language demo will be released. He has been working on implementing satisifiability and cleaning up ambiguities in the grammar. The good news is that we are making great progress. Once all the patents have been submitted, which will be in the forthcoming weeks, we will share a candidate release of the Town Language. So stay tuned for exciting updates in the near future!

Hyperdimensional Computing and To

Danna brings up an interesting question about hyperdimensional computing, referencing a paper from March 2003 that discusses a vector symbolic architecture for solving Raven Progressive matrices. He wonders if Ahad and the team have any thoughts or opinions on this approach, considering that some AI researchers are advocating for vector-Based ai as a means to achieve logical understanding and meaning. Ahad explains that while Atml, the precursor to To, was based on Binary Decision Diagrams (BDDs), the case is different for the To language. To is more complex and does not rely solely on BDDs. In fact, Ahad spent significant effort exploring logic using hyperdimensional vectors. However, he concluded that while these vectors are frequently used in machine learning, they are not suitable for the complex task of logic. So in short, To does not limit itself to Boolean operations like BDDs and can implement vector-based approaches, but the complexity of logic goes beyond what these tools can handle.

Delivering Tangible Results

Matata is curious if the To team plans on delivering anything tangible soon. After eight years of development, the community is excited to see a test net or even experiment with To itself. The answer is a resounding yes! We are very close to delivering a first demo of the To language itself. Once that is achieved, it won't be long before we have a fully functional test net. So the wait will soon be over, and you'll have the opportunity to experience To firsthand.

Comparison of AI Agents in Town and Aolis

Row wants to know how the AI agents in Town differ from those in Aolis. The AI agents in Town are unique and different from any others out there. In To, you have a specific language where whatever you say is what will happen. Although it may sound like a programming language, it is not. In traditional programming languages, you cannot ask questions about the consequences of your actions. But in To, you can. The language is capable of constraining itself, allowing you to give constraints on all future blocks in a Blockchain. This level of language power is what sets the AI agents in Town apart from others.

Disruption of Software Engineering by To

Donna wonders how software engineering will be disrupted by To. Will the traditional leadership structures of projects change, or will core developers still maintain their leadership positions? The answer is that there will always be room for core developers, individuals who dedicate their lives and work to the product. However, with To, there will be a shift in focus from the "how" to the "what." Instead of explicitly telling the computer what to do at each step, To allows for high-level descriptions, letting the computer figure out the rest. This change in approach will disrupt the way software development is currently conducted.

Social Disruption of To

Sir Andrew raises an intriguing question about the social disruption caused by To. To has the potential to completely disrupt communication and the connection of knowledge in all aspects of society. Current communication platforms are stagnant in the sense that once you leave an opinion, it doesn't do anything with it. With To, opinions can have meaning and significance within the system, allowing for connections and the derivation of new facts and knowledge. This shift in communication and knowledge creation will have a profound impact on society.

Controlled Natural Language Interface

Sir Andrew wonders if a controlled natural language (CNL) interface will be available for users during the test net or at the start of the main net. Unfortunately, for the test net, we do not plan to include a CNL interface just yet. However, for the main net, there are plans to have one. While Fluent knowledge of To and the ability to write To logical formulas will be required to utilize the innovative features of the test net, users will still be able to submit transactions without this knowledge.

Test Net and Main Net

Ken wants to know who the target users for the upcoming test net are. Is it only limited to academic users and programmers, or can members of the To community also understand and test it? The first version of the test net will require fluency in To and the ability to write To formulas. This may limit access to academic users and experienced programmers initially. However, all users will be able to submit transactions, even if they don't have in-depth knowledge of To. As the development progresses, we aim to make To more accessible to a wider range of users within the community.

Note: The remaining headings and subheadings will be covered in the continuation of the article.

Please find the continuation of the article below.

📌 Highlights:

• Building a smart constitution with To
• Town Language demo release in the near future
• The limitations of hyperdimensional computing in AI
• Delivering tangible results with To
• The unique AI agents in Town compared to other platforms
• Disruption of software engineering by To's high-level approach
• Social disruption through improved communication and knowledge connectivity
• The potential for a controlled natural language interface in the future

FAQ

Q: Can To teach chess and explain moves like human players? A: Chess is a complex game with no hard and fast rules for making optimal moves. While AI engines like Alpha Zero can identify these moves, they struggle to explain the reasoning behind them. To, with its focus on high-level descriptions and reasoning, may offer an alternative approach to chess analysis and teaching. However, explaining moves in chess remains a challenge due to its inherent complexity.

Q: Will To be compatible with existing blockchains? A: To uses a chosen blockchain as a layer one infrastructure, but the bulk of the smarts and governance will be implemented through the To language. The aim is to allow To to be compatible with any blockchain by implementing the necessary functionalities as To code. The long-term goal is to have 95% of the blockchain implemented in To.

Q: Will To disrupt traditional leadership structures in software projects? A: While there will always be a place for core developers, To's focus on high-level descriptions and automated decision-making will disrupt the traditional leadership structures to some extent. The emphasis will shift from micromanagement to defining the desired outcomes, allowing computers to figure out the detailed execution.

Q: Will regular users without programming knowledge be able to utilize the test net? A: The initial version of the test net will require fluency in the To language to fully utilize its features. However, regular users will still be able to submit transactions without in-depth knowledge of To.

Q: Can To explain the reasoning behind moves in games like chess? A: While traditional chess engines struggle to explain moves, To's high-level description and reasoning capabilities may offer a different approach. However, the complex nature of chess and the absence of hard rules make it challenging to provide comprehensive explanations for moves.