Clarke is famous for using a fictional premise as a vehicle for discussing scientific ideas, and rightly so. His ideas are so interesting, and his explanations are so clear that no one minds if the story sometimes takes a back seat to the science. In _Earthlight_, a secret agent visits the lunar colony to find the spy leaking classified material to Earth's other colonies. Clarke uses this premise to explore two main ideas: what life in an established lunar colony would be like, and how a conflict between Earth and her colonies might arise.
Clarke's description of a 200-year-old colony is fascinating. His discussions of the problems created by the hostile lunar environment, and his solutions to them, are clear and interesting. What was surprising to me is that these descriptions, written almost 50 years ago, haven't really dated. There are some anachronisms; all communications are auditory, for example - television wasn't forseen, much less the Internet. But the majority of the ideas are as plausible and compelling as when they were written.
The development of the conflict between the Earth and the Federation of colonies is somewhat less successful. The main reason for the conflict - competition for limited resources - rings true, and Clarke captures the feeling of life during a time of political tension leading to conflict. However, his descriptions of events and motivations feel a little too simple for what we'd expect to be a complex political situation. The concluding battle, employing highly speculative science, is a jarring contrast to the earlier, reasoned discussions of the science behind the lunar colony.
These are nitpicks, however. _Earthlight_ succeeds in the ways most important to Clarke and to his readers. If you're familiar with Clarke's other work, I think you'll enjoy this novel. If you haven't read Clarke before, this novel is a fine introduction, although you would do equally well with his better-known works.