I made a post on reddit and got a few responses on it. The question I posted was if someone wanted to help me going forward with the AI NLP path of the evolution. The idea is to take the character set and see if we can create any kind of coherent language. So I started talking to someone that is pretty well versed in NLP. (Thank you Sean!) I gave him the character set and actually the whole book in digital version. We won’t rush it, but I’m very excited to see where it is going. If all goes well, maybe there will be a new version of the book out pretty soon. But first, let’s talk a little about why, what and how.
Why?
The concept always was to have this work evolve. That is (partly) why I chose the original source text to be “The origin of species”. Historically, evolution has been environment-driven.
There is an advantage in the world for taller people
==> so the population tends to favour reproducing with taller people
==> shorter people don't get to procreate as much
==> the genealogy of taller people remain, while other come to a stop.But in the last centuries, the world has changed so much that man is able to change the environment around them. So some of the advantages that some people might have had in an more agrarian culture, don’t have as much an effect on the population. Technology is taking over a lot of the factors that otherwise might have triggered an evolutionary response in a certain direction.
So it would only be fitting that this work evolves in function of a new technological process: NLP.
What?
NLP or Natural Language Processing in short is how computers learn to read and understand text. Traditionally, you feed a lot of language in an algorithm and the application learns from it. It can be used, e.g., to help call centre operators respond to customers: the customer asks a question, the application interprets the language and suggests a response to the operator. Or maybe a little closer to home: if you have a gmail account and want to reply to a mail, it might give you a suggestion already.
But (!) in this case we turn it around. In stead of giving the application a lot of text and tell it to learn; we give it characters and tell it to create text. It’s something completely different of course. (very exciting) I’ll leave it to my collaboration partner to maybe explain in the future how it worked exactly.
How?
So this is a complete mystery to me. I know the power of AI and NLP. But I don’t know how far it might go. I’m truly excited to see what might come of it. So keep watching the site and see what happens. Let’s all keep our fingers crossed and cheer on Sean!
Oh and if you haven’t already, you can always by the first version. I’ll keep all versions available for purchase all the time.