Your text started "We should be able to evaluate a book" and ended "The opinion I express is more of a generic attitude on life.", which made me believe it was a general statement on all books, not on a limited subset. I think I can be excused for assuming it was a more encompassing attitude.
As for "designed to be complete", I don't know what that means in this context. Perhaps you mean stand-alone, as it does not depend on having access to other resources? Otherwise, the only definition I can come up with is tautological, as it covers everything the author meant it to cover. A technical book on "advanced differential equations" is not complete in that it assumes the reader already has a basic understanding of differential equations.
I have read many technical papers. They are still human documents. They can and often do contain meaning between the lines. I am reviewing some of the literature in my field from the 1960s, and it's easy to infer the social context. For a simple example, in most of the papers refer to a generic chemist as 'he' and clerical staff as 'she'. More subtly, there can be veiled attacks, like when one paper says "it's surprising that X and Y are connected" and then a paper a few years later says "as many of the people who founded X came out of Y, it's not surprising that there's a close connection between the two". It's not hard to read between the lines and infer that the second author is chiding the first author for not knowing the history.
More classically, in 'Dialogue Concerning the Two Chief World Systems' Galileo put the words of Urban VIII into the mouth of the character Simplico. The pope and many others interpreted that as a hidden attack on the pope, despite being a book on scientific ideas.
Have you never come across mathematical proofs written by kooks? The language, and frequent references to numerology and other occult philosophical traditions make it effectively impenetrable. Unfortunately, I can't find one easily. Visit your local math department and ask around - it's not uncommon. Sometimes much more sane versions of these get published, like http://www.math.utep.edu/Faculty/sewell/AML_3497.pdf .
In any case, math proofs are supposed to be interesting. Otherwise we would just hook up a theorem prover to a journal publisher. If you look at the early papers in statistics, they use games of chance as examples. Sure, the proofs stand alone, but from the text you can also determine that gambling was not strictly taboo during that era.
BTW, I agree with the statement that none of Zed Shaw's internet persona is evident in this book. By all accounts, it is a good resource for learning how to program. In my cursory scan, I found no hidden or double meanings. However, that's a specific property of this book, I disagree that that's true of every arbitrarily chosen technical book.