Probably one of the highest recommended books in the software world, this isn't your typical tutorial/guide/reference book on programming or software engineering. Rather, this book is a series of ideas, more or less independent, on large software projects.
I'll have to admit that you should get some actual software development/management time under your belt before picking up this book, otherwise it is very difficult to make the right connections, which is very important given this is a book full of essays from the 70s software development era. Details might have gone outdated, and many other things won't really work in today's fast paced startups who live by the fail fast mantra. Nevertheless, the human aspect of software is very accurately depicted in the text. Why software projects are difficult to manage? Why does adding man power to an already late project only makes it 'later'? Why is communication important? These are some of the interesting questions that the author addresses in the text with his opinions and experiences.
I won't go into the content of the book, as I would never be able to do justice to it. If you're looking for a short summary, refer the Wikipedia page for the book. Highly recommended if you're a software developer and would like to read one of the rare kind of books in our field, a book on the human aspect of software engineering.