I believe that having the entire text of a book online will not reduce sales of the bound work for several reasons.
First, the online reading experience, for a book of this length and density, is barely tolerable for short extracts, and not at all satisfactory for extended reading. True, the online text has the advantage that the cross-references are live and clickable. But I'd be surprised if anyone has the patience to read one of the longer chapters, even from the good screen of a desktop system. In fact, I'd be equally surprised if anyone can spare that much time for reading, out of the limited time they spend at their desk keyboard.
Put either the HTML or the PDF on the screen of a laptop so you can take it to bed or on an airplane, and the smaller, dimmer screen will soon have you blinking.
True, if someone were determined to read the book but not to pay for a bound copy, he could download each of the 13 files and print them. Printed, the HTML lacks both typographical beauty and page numbers. The PDF files are paginated, but are set to be nonprintable. True again, it is not hard to find hacker code that will break Adobe's print lock, if you don't mind violating federal law (the infamous Digital Millenium copyright law).
However the online files are printed, the result is not a book but an unbound stack of 250-odd sheets, still lacking the indexes and table of contents, whose printing will have cost the stubborn person, in paper, toner, and wasted time, several times the price of the book.
There is a very slight chance that someone might copy the files and put them up on a website somewhere, without attribution. Science-fiction writers complain of this being done with their short stories from time to time. This is a clear violation of copyright, easily detected with a Google search and easily proven for purposes of legal action. It can be argued whether or not it actually has any effect on the author's income; but in any case it is not hard to prosecute.