Without going into a long-winded diatribe, if you don't want me to see your code, don't expect it to be run on my computers. Firstly, his sort of thing is dramatically counter to part of the open-standards issue of HTML; it's not a medium in which secrets last long, so development proceeds at an accelerated pace. Secondly, never trust code you haven't looked at yourself, and then don't trust it unless you understand what it's doing (encryption software is the traditional example; most of it is incomprehensible to those lacking advanced training or experience in several esoteric areas of mathematics).
PageParser's author writes:
Incidentally, I fully believe in the ability to keep your code secret, and to maintain privacy in your own data communications. But I don't believe, generally speaking, in keeping code secret when its function is public. If something is intended for my consumption, I wish to be able and permits repairs or optimizations on my own time, without waiting for new releases &c.
That said, here's the meaningful stuff:
Note from the future: A long time later, I came across a program called TagsLock from a company named AtomPark, which did much the same thing only more abusively. So I wrote another countermeasure.