Internet

This is the most quickly changing section. There are too many Internet books and they come out and become out-dated too soon for me to ever feel like I'm up-to-date. So in this list, I'm just trying to describe the best books I've seen on Internet related topics. I'm adding books to this list for CS193i.

• Spinning the Web, by Yuval Fisher. 1996. Springer. 536 pages. $28 softcover. Currently the best overall book on server-side programming. Includes reasonably detailed technical discussions on the whole range of HTTP server issues: MIME, HTTP, HTML, CGI programming. Much more informative than the numerous and forgetable "HTML for Dummies" type books out there.
• How to Set Up and Maintain a World Wide Web Site, by Lincoln D. Stein, Addison Wesley, 1995. 469 pages. $40 softcover. Similar to the Fisher book. Provides good coverage of a variety of server-side issues: configuring and installing server software, HTML, CGI. Well written and complete.
• Learning Perl, by Randal Schwartz. 1993. An O'Reilly Nutshell guide. 246 pages. $25 softcover. The thinner of the two O'Reilly Perl books-- it's the one with the Llama on the cover (the other has a Camel). An excellent introduction to Perl. Covers all the important features with lots of little examples. For most people, this is all the Perl background they will need.
• Programming Perl, by Larry Wall and Randal Schwartz. 1991. An O'Reilly Nutshell guide. 465 pages. The thicker of the two O'Reilly Perl books-- this one has a Camel on the cover (the other has a Llama). Explains everything about Perl. I think the overall treatment in this book is not quite as straightforward as in the Llama book, but this one is more complete. Includes a long section of detailed 1-2 page Perl code examples demonstrating a variety of Perl/Unix applications.

TCP/IP

• The Internet Book, by Douglas E. Comer. Prentice Hall. 1995. 312 pages. Paperback. Also known as "The Thin Red Internet Book". Fills a unique niche in Internet books-- it explains what the Internet is, where it came from, and how it works from a largely non-technical point of view. The book's strength is its clean, well-written explanations of the "plumbing" aspects of the Internet-- how the bytes get back and forth. Explains namespaces, routing, packets, ...the whole deal. The book does not have any source code, detailed routing strategies, or socket interfaces. It's not that sort of book. Instead, it describes in medium detail how all the pieces fit together and why. The second half of the book describes, in slight detail, what the basic idea is with some of the services built on the Internet-- gopher, HTTP, WAIS.... I found that part of the book less appealing. Overall, a very well written, gentle introduction to the technical underpinnings of the Internet. Does not assume a technical background and does not provide technical details. Has a lot of useful and interesting content packed into a relatively short, readable book.
• Internetworking With TCP/IP, Volume 1, Principles, Protocols, and Architectures. Douglas E. Comer. 1995 (3rd Ed). Prentice Hall. 613 pages. Hardcover $52. This is an excellent book. Covers everything about the role and implementation of all the parts of TCP/IP, but just a little about writing client code to use TCP/IP. Discusses the whole range TCP/IP components, many in sufficient detail to begin to understand their implementation. This book is probably overkill if you just want to write some TCP/IP client code. This is an excellent book to understand what's really going on with TCP/IP for technically sophisticated programming or network administration. A bit expensive.
• Internetworking With TCP/IP, Volume 3, Client-Server Programming and Applications. Douglas E. Comer and David L Stevens. 1993. Prentice Hall. 498 pages. Hardcover $63.50. Another excellent book from Mr. Comer. In constrast to volume 1, this book focuses on building things on top of TCP/IP. Covers everything and features lots of code examples. The book comes in two versions-- a Socket interface version and a version based on the TLI interface associated with AT&T System V Unix (if you have no reason to care, get the socket version). More than just code, the book has well written discussions of the design philosophies and tradeoffs associated with various client server strategies. Covers many issues for both client and server side programming: connection vs. connectionless, state oriented vs. stateless, server side concurrency issues, implementation notes for RPC and NFS. Even more expensive than volume 1.
• Unix Network Programming. W. Richard Stevens. Prentice Hall. 1990. 772 pages. All about communications on Unix. Not specific to TCP/IP channels but it is very Unix specific-- covers all varieties of Unix communications. Covers both BSD Sockets and AT&T TLI. Lots and lots of sample code. Includes sections on many common historical Unix protocols: ping, lpr, rshd, rlogin. Does not cover newer, non Unix oriented protocols like HTTP etc.. Stevens also has a series of very thick, very detailed books called TCP/IP illustrated, volumes 1, 2, and 3. I haven't read them, but by all accounts they are extremely complete on both implementation and applications issues.

Human Interface Design

C++ and OOP

• Mastering C++, by Cay Horstmann, John Wiley, 1991. Covers C++ through release 2.0. Concise and readable- the thinnest book on C++ around. Intended for people who already know C or Pascal. Assumes you basically know how to program, and concentrates on the most important language features. Sadly, this is getting out of date.
• C++ Primer, 2nd edition, by Stanley Lippman. The complete reference for release 3.0 of the C++ language. Technically rigorous. The explanations can be a little dense. Makes a good reference. 614 pages.
• The C++ Programming Language, 2nd edition, by Bjarne Stroustrup. Similar to Lippman but written by the original designer of the language. Some prefer this to Lippman. 669 pages.
• C++ How To Program, by Deitel and Deitel. Not quite as technically deep/obscure as Lippman or Stroustrup, but very complete and readable. Has good examples and explanations for all common C++ programming issues. Includes a review of C. No doubt includes much of their earlier C book. 950 pages. Covers 3.0 features. There are many books which basically fill the same niche as this one.
• Thinking in C++, by Bruce Eckel, 1995, 813 pages. A competitor to Deitel and Deitel. Does not review C issues. $31.95.
• Object Oriented Design, by Grady Booch. A 580 page hard cover treatment of the style, techniques, and philosophy of OOP. Not specific to any particular language. Sufficiently abstract that it supposedly can be read even by "executives."
• The Design and Evolution of C++, by Bjarne Stroustrup, Addison Wesley, 1994. Bjarne Stroustrup's other book about C++. Discusses what the objectives and concerns were in the design of many of C++'s features. Helps you to understand many of C++'s more advanced features, and gives you a better appreciation for why things are the way they are in C++. This book helped me to find C++ less irritating.

Advanced Programming

• Expert C Programming, by Peter Van Der Linden. 1994. 351 pages. Discusses advanced and seldom described areas such as the most subtle distinctions between arrays and pointers, the linker, shared libraries, and stories about famous bugs. UNIX C oriented. No discussion of basic C issues, but otherwise full of interesting stuff. $32.
• C Traps and Pitfalls, by Andrew Koenig. 1989. A 147 page book devoted to all the subtle bugs you can get in C. Good reading when you feel like you mostly know your way around the language and you want exposure to some advanced issues.
• High Performance Computing, by Kevin Dowd, O'Reilly & Associates, 1993. The best, most practical book on current performance techniques. Starts with a good introduction to the state of the art in processor and memory architectures. Then goes through compiler optimizations, loop optimizations, memory optimizations, and finally, real obscure optimizations. 371 pages. $26.
• The C Programming Language, 2nd ed. by Brian Kernighan and Dennis Ritchie. 1988. The second edition of the original language reference from its designers. Thin and to the point. Defines behavior without long explanations or examples. Some people really like its style. Stay away from the first edition which describes pre-ANSI "old-style" C. $34.
• C How To Program, by Deitel and Deitel. 1992. 588 pages. Complete discussion with examples of all the major C constructs. Includes some discussion of C++. There are thousands of books pretty much just like this one, so you're best off flipping through a couple at your bookstore and picking the one you like best. Stay away from the really thick, slow moving ones. This one is distinguished by the fact that it's current, complete, and they sent me a free copy.
• Writing Efficient Programs, by Jon Bentley. A 170 page book on how to write C/Ada/Pascal code which runs fast. A little dated.
• Programming Pearls, and More Programming Pearls, both by Jon Bentley. 1986, 1988. Two books of interesting little stories and problems about coding design and efficiency. Interesting, fast reading. Basically just extracts of his column in the Communications of the ACM.
• Principles of Concurrent and Distributed Programming, by M. Ben-Ari. 1990. 225 pages. Good coverage of all the basic principles of concurrency problems, abstractions, and implementation.
• Parallel Programming by R.H. Perrott. 1987. 252 pages. Same basic introduction as above, but more emphasis on FORTRAN type parallelism abstractions.

LISP

• Common Lisp The Language, 2nd ed. by Guy Steele. A snappy 1029 page language reference for Common Lisp from the guy who designed it. You use it as a reference to look up the behavior of the vast number of Common Lisp built in functions. The 2nd edition includes the definitions for the OOP extensions. Does not teach how to program in Lisp at all.
• LISP, 3rd ed. by Patrick Winston and Berthold Horn. A 611 page guide to how to program in Common Lisp. Includes OOP stuff and AI oriented examples.
• Common LISP, by Wade Hennessey. A more dense, to the point LISP book at 395 pages. An excellent book, which I think has sadly gone out of print.
• The Common Lisp Companion, by Timothy Koschmann. Teaches all the major areas of Common Lisp, including OOP, 459 pages.

Symbolic Systems / History of Mathematics

Miscellaneous Computer

• The Mythical Man-Month, by Fred P. Brooks. 1975. 195 pages. The classic essays on why software is hard to do, especially when coordinating a large group of programmers. The source of such gems as "Adding more software engineers to a late project makes it later" and "Plan to throw one away." Dated but entertaining. $24. Supposedly there's a new edition in late 1995, but I don't know if it's been seriously updated.
• Dynamics of Software Development, by Jim McCarthy, 1995, Microsoft Press, 184 pages, $24.95. An excellent book about the development process. Very readable. Lots of stories drawn from the author's experiences at Microsoft. A slight "touch-feely" tone, but don't let it bother you. The book is not oriented just at engineers-- it discusses the entire development process and its players" engineers, managers, QA engineers, marketers, and customers.
• Accidental Empires, by Robert X Cringely. 1993. 336 pages. An outstanding, funny history of all the little stories, which have made silicon valley. Filled with interesting, gossipy stories about famous people, companies, products, bugs...Really quite entertaining. Helps explain why things in the computer industry work out the way they do. $11.
• The Macintosh Way, by Guy Kawasaki. There are Macintosh people, and there are PC people. Guy is a Macintosh person, but if that doesn't bother you too much, this is an entertaining exposition on the development of the Mac. In hindsight, it's clear that the stealing the GUI ideas from Xerox and developing the Mac was an outstanding and influential idea. The main theme of the book is avoiding mediocrity. Witty and entertaining. Always cheers me up.
• The Soul of a New Machine, by Tracy Kidder, Avon, 1981. Winner of the Pulitzer Prize. Semi-fictional story about the development of a new computer. Captures the flavor of how hard engineers can push themselves when they are immersed in a product. A good read.

Witty/Cynical/Trendy

Science Fiction / Fantasy

Ok, there's a lot of forgettable science fiction. God knows, I've read and forgotten tons. But some of it's pretty good.

• The Foundation Trilogy, by Isaac Asimov. Doubleday, 1951. One of the greatest works of science fiction of all time. Creative, engaging, and relevant. 20 million nerds can't be wrong.
• Ringworld, by Larry Niven. Ballantine, 1970. Probably my personal favorite work of science fiction. Very inventive and entertaining. Unfortunately, re-reading it now that I'm not 11, I see that Niven is a bit of a sexist. But the outstanding quality and creativeness of the plot help to make up for it.
• Snowcrash, by Neal Stephenson. The first few chapters are hilarious and inventive. Presents an interesting view of how the "information society" of the future might turn out. Then the plot gradually goes downhill as the novel progresses. Overall, the book is ok, but the beginning is outstanding.
• Ender's Game, by Orson Scott Card. Awesome. A real page turner.
• The Hobbit and The Lord of the Rings Trilogy, by J.R.R. Tolkien. Awesome, epic story. The greatest, basic Good vs. Evil story every told. In a recent gathering, we ended up trying to name books that everyone had read and liked. The Tolkien works were the immediate and unanimous winners.

Economics / Finance

• The Work of Nations, by Robert Reich, Vintage, 1991. The best book on the development and future of "work" and the economy. In particular, the book deals with the role and future of the information economy. Should be mandatory reading for anyone who wishes to be an informed inhabitant of our planet. Reich is currently our secretary of labor, where, as far as I can tell, he has accomplished surprisingly little for a man with such powerful and insightful ideas. On the whole, this serves as a depressing indictment of the ability of our political system to crush all that is innovative in its trustees. There is no doubt a related lesson in Al Gore's accomplishments as vice-president after writing Earth in the Balance. Remember the BTU tax? Grazing fee reform? It is at least comforting that Clinton hires bright people, even if they can't do anything.
• A Random Walk Down Wall Street, by Bernard Malkiel. The best book on the operation of the modern financial world. Half entertainment, and half ruthless explanation and criticism of the facts and foibles of the financial "industry". As an "efficient market" person, Malkiel argues that the collective analysis, rumors, and speculation of all the financial analysts and salespeople doesn't really do much useful except generate a lot of commissions. The book advocates that you put your money in a low-overhead buy-and-hold type mutual fund such as the Vanguard Index 500 or the Schwab 1000. Having gotten that one basic investment decision right, you should then ignore investing and concentrate on having a nice life (unless you want investing to be your hobby in which case you should not kid yourself about what you ard doing). Strangely, salespeople, financial institutions, and advertisements seldom advocate such a strategy.
• The Only Investment Guide You'll Ever Need, by Andrew Tobias. Bantam Books. Originally published in 1978 with many revisions since then. The title pretty much says it all. Thin, non-nonsense guide to your financial life.

Miscellaneous

• Adrift: Seventy-Six Days Lost At Sea, by Steven Callahan, Ballantine, 1987. This is an amazing book. It's the narrative of the guy who set the world record for surviving the longest in the open ocean after his sailboat sank. I read it in one marathon sitting; I could not put it down. What's great about it is how he needs to deal with ordinary problems but with a life-and-death urgency. The water-distiller things he has don't work quite right, so he needs to fix them using whatever is at hand. He needs to spear fish to eat, but there's the constant problem of poking the raft itself. I think it just happens to hit the combination of adventuring and tinkering which I find most appealing. It's a little reminscent of debugging. A sort of Apollo 13 of the sea now that I think of it. Would make, a good chapter-a-night type bedtime story. Anyway, I liked it a lot, but not all my friends liked it that much. The book is not especially well known, and some bookstores do not carry it.
• QED, by Richard Feynman. Princeton University Press, 1985. "The Strange Theory of Light and Matter" An amazingly conversational and approachable discussion of the forbidingly named field of Quantum Electrodynamics. Transcribed from a series of lectures Feynman gave. Save it for one of your more quantitative moods.
• How Much is Enough?, by Alan Durning, Norton, 1992. A very reasoned criticism of consumer society. Explains the pitfalls of a consumer attitude which confuses physical possessions with happiness. Thought-provoking, inspiring, and potentially depressing. From the folks at the World Watch institute.