I learned photography from books, and from people who knew what they were doing. I learned book-binding from the Internet. The differences are, well, they're illuminating.
Although this is less and less the case, when you're writing a book about how to do a task, you tend to organize the ideas together. You test procedures you're recommending, you research ideas you're proposing. You try to pull together a coherent set of more-or-less true statements that present a more-or-less complete picture of how to do the task.
When you're writing a web page on how to do a task, you pull together random bits and pieces of bullshit from a quick google search, stitch it together with a few things you've actually tried out superficially, and you fling the whole mess up there. The result is incomplete, riddled with nonsense, and disorganized.
In order to learn to sew a perfectly ordinary Smyth-sewn book binding, I had to pull together information from at least three different web pages. Were the authors of the web pages simply leaving bits and pieces out through carelessness? Do these authors actually just sew bindings incorrectly? Or do the authors in fact not actually know anything, but are filled with an inexplicable desire to teach others how to do this thing they do not know how to do?
I don't know and it doesn't matter. The bottom line is all the web sites on book binding that I have found have errors and/or serious omissions.
Web sites which purport to teach photography are just as bad. Incomplete, incorrect, and assembled by people who have no idea what they're talking about.
In a very real sense, the vast Internet which allows uniform access to knowledge to all has returned us to a tribal state in which "knowledge" is a mixture of truth, superstition, and cargo cult logic.
This witch-doctor's brew is copied and passed on not by word of mouth but through the mechanic of web searches. It is inherent in copying that further errors are introduced. Without mechanisms of testing, editing, correcting, the errors remain to be themselves copied, so the each generation is always, unambiguously, worse than the previous.
Thus we get endless bullshit about "the medium format look" and "rule of thirds" and "exposure triangle". We also get endless overly technical descriptions, because these things feel weighty and "true" but often fail to actually describe anything. We're given depth-of-field calculators to play with, which may or may not be correctly implemented, but which most assuredly tell us not one damn thing about what our pictures will look like. We're given detailed descriptions of Airy discs and the optical details of diffraction, which may or may not be riddled with errors, which tell us not one damn thing about what our pictures will look like.
In this modern era, we increasingly see this same approach taken for books, formerly (albeit only to a degree) a kind of bastion. Book authors copy rubbish from one another, and from wikipedia, and the Internet, with increasing abandon, to churn out the how-to title du jour.
This is not exactly progress.
If you want to learn how to do something, get a book. An old book.