14 comments

[ 2.7 ms ] story [ 52.1 ms ] thread
What is the target knowledge representation that you expect to produce here? Are you suggesting the neural nets will output formal knowledge represented according to an ontology? This is very vague.
I think the answer is "it doesn't matter" because it will never happen. This guy has set himself an absurdly ambitious, poorly defined task.
just wanted to say that like button doesn't work
Trying to understand how you “compile a book to JavaScript”. Just write code based on the examples or what? Oh well, you got a subscribe so I can finds out...
I honestly have no idea what this means
I don't know what this project means at all. But if you're interested in projects that formalize the knowledge encoded in stories, check out http://groups.csail.mit.edu/genesis/index.html

Professor Winston isn't alive to drive the project, fyi, and I don't know what other similar projects there are.

1 book a week is kinda crazy. So something should be coming in a few days?
this is the intro plot to a bad 90's movie about how an evil programmer amassed enough wealth to take over the world
While the original post makes no sense there are some interesting articles in the comments which I find mildly amusing.
Strange that the author starts with a book that began as an online, runnable, interactive book. (Taking a lot more than a week to write.)

Presumably he would end up with a library of functions that somehow capture some of the mathematical models found in some books. Without any of the understanding of what they mean or how to apply them, this is meaningless.

If you really want to do something meaningful and ambitious, spend a year trying to write a program that can understand a single sentence.