You are viewing a single comment's thread from:
RE: Demo coming soon? Tau Meta Language in C++ updated on Github.
I would hope Ohad at least provides a demo which people can relate to before the TML in TML. People need to see visually the progress not just see code. I know how to read the code but I'm not 99% of the audience. For the 99% seeing Tau in action is the game changer.
From what I see on Github doesn't Ohad still have to complete the Earley parser component or is that completed somewhere else? I would think he would need the parsing mechanism before he can code TML in TML? But it does look like he is nearly complete with regard to PFP and BDD which are the two major components that make TML special. The Early parser if I understand right is the component which will allow discussions to translate into the logic component.
EDIT: The demo comes before TML is written under TML. I made a mistake initially. Demo is the next main step
I had to ask Ohad
Yes, the Earley Parser is a very important component to the TML, but it will be applied on the next layer (TML written in TML).
There are a lot of reasons why TML has to be rewritten in TML (rather than just in C++). Firstly, the entire idea of Tau revolves heavily around self definition. Tau is about defining its own rules using its own rules, which is how we'll get to a protocol that can amend itself in subsequent blocks according to the real time consensus of its users in the current block.
Additionally, implementing the TML using BDDs gives it a unique feature: it enables the TML defined under itself to carry significant performance increases over the original. This means it no longer makes any sense to implement the parser over the original C++ TML as one would rather implement it over the faster version and the Parser itself is not necessary for self definition.
Similarly, the demo will come once TML is defined under itself. <- wrong, it's coming before that
BDDs are one of the best things to have happened to the project since the discovery of Partial Fixed Points (only known decidable, self defining logic that supports negation). It is truly very well suited to our needs and offers extraordinary advantages over alternatives.
Finally, Ohad doesn't anticipate this next step (TML over TML) to be as difficult as the previous one and it shouldn't take anywhere near as long.
Interesting. I have no idea how he will implement TML over TML without some kind of parser but I guess I will find out. I mean I suppose he could do it but it seems like it would be quite difficult. What I can see from the current Github code direction is you can definitely implement the rules, and you can for sure handle the logic, and I can sort of see how you can do stuff like deductive reasoning and model checking etc.
I admit the self definition part seems to be conceptually the more difficult part. From what I understand the Futamura projection component is what ultimately will allow TML to be implemented in TML. From what you say it seems he will have to reverse engineer the C++ code base and do it again in the TML language which I'm hoping will be simple else we will not be able to even read it.
I guess I can look forward to the Futamura projection which is the next key component I expect Ohad to implement as it's the logical next step. For instance PyPy seems to be an example of what Ohad might be trying to do next where Python is using the Python language to rewrite a better version of itself. The benefit seems to be "automatic compiler generation and optimization". This might explain how TML could be generated from TML itself to be far superior and better optimized than done in C++. Strategically speaking this makes a lot of sense but it also means we might not see or know the benefits until later.
Reference
sorry, I made a mistake about the demo
Ohad plans to demonstrate TML with the initial version, it doesn't have to wait until TML over TML
So it shouldn't be very far off