Yesterday, I finished the software framework of the website enough that I could put up the text content of each chapter.
I left out the images for now, and will probably need to fix up the references, but it’s basically all there.
As I was copying over the information, I left out one or two chapters because they might not be good enough even for a basic read-through. I left out chapters on nutrition, and on how a person should live if they want to live forever without worrying about money, etc.
I want to talk a little more about the software end of the website, because it’s going to be helping out in the next few years, telling me what pages are the worst in the book.
The basic idea I have is that the measurement of how readable a page is, is how many people clicked from it into the next page.
I’ve broken the book down into chapters and sections. Each section is a stand-alone article discussing a specific topic. For example, the “mathematical universe” in the “quantum immortality” chapter is a defined topic, so can be separated into a whole section on its own.
Each section will have multiple “variations”, with very slight wording changes, image changes, etc. The idea is that visitors to the website will be given a random variation of the section that they want to read, and I will then be able to measure how interesting and readable that variation was, by how many people finished reading it and clicked on “Next” at the bottom to go to the next section (link not yet added at this moment – will add that this evening).
If statistics are kept for hundreds or thousands of reads, then a pretty accurate statement can be said about which section variation is “better” than any other one, because there is a definite goal: the reader should want to go to the next page.
It’s important that the statistics be done over as broad a number of reads as possible, because search engines will read the whole book anyway, as will scrapers, so a larger sample set is needed to help dampen down those effects.
Another issue is that sometimes someone will come to the book through a search engine for a specific topic, and leave after reading that section. In those cases, how do we know how well written the section was?
The solution I’ll use here is that sections themselves will be broken down into artificial “pages”. If someone clicks through each page of a section, then it’s good, and if a high percentage stop reading at a certain point, then that indicates a page that needs work.
Reading through the pure text of the book at the moment, I know it’s not very good, but I expect that over the next few months, it will improve dramatically.