Ok, as of now, I'm not a huge fan of algebra. I feel that this is a result of using Artin's algebra text (which has a very strong flavor that you may or may not like) and the fact that algebra feels like a collection of topics as opposed to a coherent theory. However, this book makes me tolerate algebra, and I must certainly applaud it for doing so.
Herstein is one of the best mathematical writers I have read. I feel that he tops Spivak, Stillwell, and possibly even Rudin. I'd probably rank him with Axler. He writes with clarity and enthusiasm, and his obvious love for the subject is dripping off every page. Hertein is somehow able to take a very typical algebra book, and make it into something enjoyable. One important thing to note is that this book is not quite as flavored as Artin's is, and this is the result of Herstein treating conventional topics in a rather conventional way. People tend to either love or absolutely hate Artin, but no one could truly hate this book's presentation.
As commented on by many reviewers, this book is especially strong in group theory. This book takes time to build up more theory than most other books, and it does so in an exciting way. However, after Herstein's discussion of groups, this book becomes quite shallow in many areas. He dedicates only one subsection to modules, and many reviewers have commented on the skimpiness of his field and Galois theory sections. This is a book that will be easily outgrown by anyone who uses it, and this is the reason I have given it four stars. I cannot really comment on what good references for undergrads would be on modules, but Stillwell's Elements of Algebra: Geometry, Numbers, Equations (Undergraduate Texts in Mathematics) is a good introduction to field and Galois thorey. The sheer fact that a student would need to supplement an already expensive book with others is quite annoying. Artin, on the other hand, spends a chapter on rings, and chapter on modules, a chapter of fields, and then finally a chapter on Galois thoery. The fact that Artin gives decent discussions of each of these topics has caused me to begrudgingly return to his book and start looking for buyers of this one. I feel that this is the reason that so many more classes will opt to use Artin as opposed to this book.
However, I don't think that this book is useless or will become obsolete. I feel that anyone who works through this book will easily be able to begin using a book like Lang's Algebra. So I guess the choice of whether to use this book or Artin's will come down to the professor's (or buyer's) preferences in what should be covered and where to put the emphasis.
Like I said before, I'm not a huge fan of algebra, but I did enjoy this book. So I'm guessing that if anyone who actually likes algebra picks this book up, then they would probably view this book as the greatest thing ever. One word of caution, this is a more beefed up version of his Abstract Algebra, and so only stronger undergraduates should consider using this. This book is, after all, conventionally used in honors sequences.