This is a great introduction to the philosophy of science, and it remains quite contemporary even though it's already 16 years old. James Ladyman goes through the main discussions in philosophy of science in a very clear, accessible and engaging way. The book covers the famous problem of induction and of Baconish naive inductivism and the attempted responses to it, as well as Popper's account of the logic of science and its problems, closing the first part, which is about the scientific method, with Kuhn's historical challenge to the rationality of the dynamics of theory change in science.
The second part of the book is dedicated to introduce the still very alive debate between scientific realism and antirealism, and here Ladyman is very wise to break down the different requirements (these are the semantic, metaphysical and epistemological requirements) involved in the scientific realist position, since by making them explicit it's much easier to understand how there are different ways to be a scientific antirealist, not all of which equally plausible. A great feature of the book, in my opinion, is the attention given to a specific kind of antirealism, which is Bas van Fraassen's constructive empiricism. Unlike certain popular types of antirealism, constructive empiricism concedes that science is an epistemically privileged enterprise and that it constitutes a paradigm of rational inquiry, but goes on to deny that we ought to go beyond the empirical adequacy of scientific theories that postulate unobservable entities (e.g., electrons), thus promoting agnosticism about these entities and thereby denying the epistemological requirement of scientific realism. To my mind, van Fraassen is successful in putting a good deal of pressure on realism about unobservables, showing in the process how antirealism can be a perfectly intellectually respectable position to hold after paying close attention to actual scientific practice. One of the most interesting things about the discussion around constructive empiricism in the book is van Fraassen's criticism of abductive reasoning (also known as "inference to the best explanation"), which is a valuable tool for scientific realists. The other problem that was given a more extented attention was the so called "pessimistic meta-induction" against scientific realism, which is one of the most important arguments in the realism x antirealism debate together with the "no-miracles argument", as well as the underdetermination problem. The second part, and along with it the book itself, closes with a very brief overview of structural realism, a pretty recent and increasingly important contender in the debate, and it felt like Ladyman should have extended this section a little more.
As any good introduction, it leaves many problems open and avoids taking sides as much as possible. At the same time, though, Ladyman doesn't shy away from offering his own critical evaluations of some arguments here and there, which isn't a bad thing at all since he manages to keep as much impartiality as possible even then. At the end of every chapter we find a very useful list of further reading suggestions, as well as some somewhat amusing dialogues between two fictional characters, Thomas (who plays the role of a skeptic about scientific rationality and scientific realism; I wouldn't be surprised if the name choice was a clever reference to Kuhn) and Alice (who is committed to the rationality of science and to scientific realism), which have a triple function: the dialogues sum up what was discussed in the chapter and, secondly, give the reader a hint about the problems to be discussed in the following chapter; the third function is more subtle and I'm not sure whether it's easily recognizable: Alice claims certain things that are not explicitly presented in the book's chapters and some of them hint at what Ladyman himself might think about what is the best direction to take relative to a given issue (this was most obvious to me when Alice tells Thomas that maybe the demarcation problem has more to do with the institutional context where scientists trade information and debate ideas than with a set of distinctive procedures and attitudes held by individual scientists). To be sure, I might be completely off here, but by being familiar with some of Ladyman's work I was left with this impression.
The only thing that really bothered about the book is that it ended quite abruptly, like another reviewer has pointed out. A nice summary/conclusion would be in order for a book like that. But anyway, again, the book is an amazing introduction today as it was when originally published. I highly recommend it to anyone interested in getting to know what philosophy of science is about, and it's likely that I will re-read it (if not all, a good portion of it) in the future!