I think it was me that described them as rubbish, so I shall defend that statement. They are rubbish because they have no content, or rather, what mathematical content there is is so diluted as to lose any interest. They do not prove anything, and the statements that they make are trivial in the extreme and not at all worth reading. Anyone with one scintilla of one iota of one brain cell can immediately see that they are dross and contain practically no mathematical knowledge worth writing down. Engineers may disagree at will. Actually what irks me is that they charge a ridiculous amount of money to students to temporarily own a book that could be written by the tutor in one 100th of the amount of space at fractional cost of energy to the tutor. Canonical example is the fact that one of them (FInney I tihnk), proves three different chain rules for them. Hello! There's only one chain rule, don't dress it up and make them think it is harder than it is. Once they're doing multivarible calc there is no difference in difficulty between the chain rule for functions from R^m to R^n for any m and n so don't make them learn special cases...
So, to clarify, they are rubbish because they are exteremely bad value, and cannot be condoned as purchases to anyone at all, excepting that many universities require their students to buy them *for no good reason in my opinion*.