This may answer the question, but it will not be satisfying. There are very few books because the legal department goes after authors who create books without their permission. I am three connections away from an author who was told that screen shots of the software are releasing proprietary information about the user interface and he should stop distributing his book. Perhaps this is why many books show models and have long discussions but don't tell you "click here".
My speculation on the printed material is that it was just to costly to edit, print, and distribute manuals. An e-mail announcing a new release has practically no cost associated; a box full of optical media may cost a few dollars to ship; a box with optical media, and printed manuals may cost more than ten but less than twenty dollars to ship. Considering these costs, and that there may be installations world-wide (increasing shipping costs), and a user base of something like 50,000 people, you are talking about half a million dollars to ship a new release with printed materials. If you create a new release every 18 months, that is a lot of money to spend on a regular basis.
So a solution is to distribute user guides in pdf format. Once a pdf is released, what prevents anyone from e-mailing it to a friend? What prevents anyone from e-mailing it to a competitor? Does this pdf upset approved authors because they are not selling books? I am sure there are many more questions like these that kill distribution of pdf files. The result is an on-line library.
The on-line library requires a username and password to enter. This prevents sharing of resources, or at least slows it down a bit. The on-line resource can also be updated at any time. Corrections can be made daily, if required. The problem is you must pay for access to these resources, and an administrator of the resources has to assign content to you. For example, your college professor may be the adiministrator, but he has to find the content to help you in your specific situation and assign it to you or assign the entire on-line library to you for you to find the answer. The latter is easier for your professor because many times the answer is found in pieces. One piece discusses selecting objects, another discusses creating datums, another discusses creating sketches, another discusses creating sweeps, another discusses creating evaluate features, another discusses global variables, and so on. The PTC folks each describe how to use their tool expecting that you already know how to use other tools in the software. Theoretically, an on-line library allows you to jump around like that better than a book. In actual practice, a book has an index that covers all topics between the front and back cover; the on-line resource has an index covering each section, so you don't know where to look for related topics. Lastly, I can put book marks in a paper book and flip between them in less than a second. Each on-line topic must be setup and try to determine where you were last. That usually takes more than ten seconds and less than 30 seconds. Just long enough to wonder if the Internet is slow or something. As a learner that is very frustrating.
If you are wondering, I have been teaching the basics of Pro/ENGINEER since 1998. I've never been completely satisfied with a paper book, but I have been much more satisfied with paper books than the on-line resource. On one hand the on-line thing is cool because you can watch someone use the software; something that a book can not do. On the other hand, the index of the book is better, access is immediate, and the book doesn't "shut off" after you own it for more than 365 days. I have used Roger Toogood's tutorial and advanced tutorial, books by CADQuest, a printed book from an author, and probably a few I forgot. I currently use the Precision LMS by PTC, but I am considering going back to paper because of student complaints.
