disease. :) Nowadays, I think there are better ways than COM or CORBA for writing reusable components. (And, incidentally, those better ways don't involve making all languages a thin veneer over one runtime.)
No doubt about it, but with COM, it is often a matter of using EXISTING components. Which is where Visual Basic used to have so much of a foothold, and still has, as far as I can tell.
I doubt this will ever change regardless of what books exist. I don't believe [skipped] The same kind of environment typically doesn't exist in computer science.
But there's another difference too -- practicing industrial programmers often stand to benefit from improving their skills and sometimes are even pressured to do so :-) Good books are the #1 aid in this quest. I don't think this situation is anywhere close to what you have for math or physics (sorry, that's what I think about when I say "science" -- must be my own background showing through) -- unless you are in research, you're not very likely to need [to be current in] them so much in your work.
Relevance to what, exactly?
To what is being done (and used) in the world at large. Much as I am frustrated by the legacy of bad design decisions that the industry is doomed to carry (and slowly replace with other bad design decisions, or so it seems) I remember all too well the fate of Algol 68, Prolog, VAX/VMS -- to name just a few. ...Max...