The Protocols (TCP/IP Illustrated, Volume 1) by W. Richard Stevens

The Protocols (TCP/IP Illustrated, Volume 1) Reviews

• 
  Michael Finocchiaro

  A classic like all of Richard Steven's books about networking, TCP/IP Protocols Vol 1 is a critical formative text on how the internet was created. As I have said elsewhere, without TCP/IP there would be no Google, no Facebook, and no Drumpf? Oops...
• 
  Casper Gasper

  Still one of the best books to read and study for networking, despite it's age. To give you one idea how old this book is, there is a section towards the end that lists other TCP services and there's a few lines on a new application called the World Wide Web :-)
  Nevertheless, it's surprising how little has changed -- highly recommended for anyone who's involved with networks.
• 
  Pete

  If you really want to understand the TCP/IP protocol buy this book. To go even deeper get volume 2 which discusses the actual code implementation
• 
  Marin

  The range of topics covered and a level of detail that's spot on for each of them made this a dazzling cross-section of TCP/IP networking. Thanks to its numerous detailed but clear descriptions, fragments of computer networking knowledge you might have get woven together into a whole, and to-the-point details that the book is peppered with are the markers that fall into the empty spaces. The dynamics of the protocols (TCP in particular) are so well described (and illustrated!) that you get a real feel for them.
  
  The problems this book has, I think, are attributable to the changes made in the second edition. I've never read the first, but I've skimmed over some of it. The shift in the material covered feels exactly right (the one exception - which made me sad - was ditching the routing protocols), but there were certain points in the book at which I really had to summon my resolve to push through. This happened in the DHCP & ICMP chapters, and almost all the time during the security chapter (EAP, IKE and DNSSEC are vivid examples). Why did this happen? I think the reason is that the writing got too terse and convoluted, and didn't have the same quality that Stevens' writing exhibits. The wording and flow demonstrate that the topic is very easy for Kevin R. Fall, but unless you already know what he's talking about, you'll only too often be struggling to keep up. Then you'll fail and reread whole paragraphs that are unfortunately written with the density of mathematical proofs.
  
  I can't hold that against the book though, and I'd still read the second edition over the first. There are some challenges, but there's so much to like!
• 
  Allisonperkel

  if there is a bible for tcp/ip programming this would be it
• 
  Scott Holstad

  This was both incredibly essential and a life saver for me when this came out. At the time, I had just completed a graduate degree in an entirely different field, but somehow found myself trying to handle two small SCO networks, as well as networking some Power Macs and, funny, some old XTs and ATs -- with no formal training or education. I'd had my own computers since 1982, used my first modem in 1980, had been crawling all over the BBS's, learning networking on my own, and all telephony, but while I started taking actual formal classes at a nearby major university, I was also expanding my familiarity with additional UNIX flavors and thinking of starting my own business revolving around this fairly new thing called the WWW which all the suckers were being told they "had to be on" and since there wasn't much competition back then in that line of work, you could charge a freaking fortune and they'd pay. But in order to do that, most of the time I had to teach these clients WHAT the Internet was (not the WWW of course), how it worked (thus a more formal need for comprehensive understanding of the TCP/IP stack, and thus this book), different types of networking, security issues, how to use basic apps for email, ftp, Usenet news, browsers of course, and then well, everything. And because these clients were geographically spread out all over the country and back then there really was no consumer broadband, I had to get accounts with 25+ ISPs, which all used various POPs of course, which was also "awesome" to try to educate newbies on, as well as modems and everything, but I had to get those accounts and test them regularly so I could advise a San Francisco client which of several had the most local POPs, most dependable re up time, etc., as well as Knoxville, Wilmington DE, etc. While I knew much on my own from nearly 15 years of experimenting and even though I had taken many programing classes, as well as recently some UNIX classes, I needed the formal, comprehensive, structured knowledge of the various protocols in order to best serve my customers while continuing to grow professionally myself. I have several "classics" back then that I've always felt I couldn't have made it without, and this is one of them. So very recommended, even though outdated, because of its historical implications and because the foundations it was built on still exist and thus this remains relevant. (One of my other -- essential -- favorites from that period remains Roger L. Freeman's Telecommunication System Engineering [Wiley, 1996] and I can't recommend that enough, although at over 1,000 pages, busy people may be hard pressed to find the time necessary to make it all the way through. For anyone interested, while there is some redundant crossover, I always viewed this as fairly complementary to the Stevens book here, and I gave both five stars and both are in my Favorites bookshelf, so again, anyone whose interest lies in this field could do no better than those two books, IMO). Both strongly recommended!
• 
  Sean S

  The Protocols a.k.a "Fantastic Protocols and How to Find Bugs in Them"
  
  This blast from the past is awesome for the following reasons:
  * It is a snapshot into a bygone era of the first stirrings of the Internet, and how people "normally" shared files and information during this time
  * The author is exceptionally thorough and logical in his analysis of each feature, and his testing methodology is reasonable and reproducible
  * Some of the information in this book (a fair bit) is quite outdated, but the fundamentals are not, especially the common protocol information, and especially the TCP information
  * Systems during this time were slow enough to expose the state machine of some of the protocols in an easy fashion
  * Insights and exposure into how these protocols were implemented, despite RFC standard (computing truly was like the wild west), usually resulting from hardware limitations, processing overhead restrictions, or just because they wanted to.
  
  Excellent read, and a good reference for the fundamentals and history alike.
• 
  Drobinsky Alexander

  Clear simple explanation of IP protocols / Ethernet essentials / BOOTP / TFTP / DNS / UDP and of course TCP protocols. It would be nice if author would perform cleaning e.g. remove obsolete protocols & outdated versions of TCP implementations while dedicate some time to review of newer protocols including routing protocols such as MPLS.
• 
  Michael

  One of the classics.
• 
  Peter Wang

  A real page turner. Was on the edge of my seat for all 600 pages.
• 
  Sandra MS

  It’s dry, but very informative.
• 
  Sean Rich

  This book was hard to get through, but I sure as hell know TCP/IP now. Wouldn’t really read it again (too long) but a fantastic read for anyone trying to get into networking
• 
  Kai Weber

  Another example of a book about computer technology that is outdated and a good read at once. The middle layers of network technology are described in minute detail, at just the right level of abstraction. This book (originally published in around 1994 I think) predates the advent of something like search engines for the WWW by years, and yet all the technologies described are still relevant today - or at least still working under the hood of other things. ifconfig, tcpdump, arp are commands that my Linux shell still understands today.
  The thread binding the whole book together is the author's intent to not just describe, but to actually show what's happening when you're running this or that (inter)network protocol. Examples are first described, then explained. This didactical impetus makes this book timeless - at least as long as we are still relying on TCP, UDP, IP, DHCP and their brothers and sisters
  
• 
  Ivan Idris

  This is a fantastic textbook about different network protocols. The protocols are illustrated using the output of tcpdump and other utilities. Protocols discussed include:
  
  IP
  ARP
  RARP
  ICMP
  UDP
  IGMP
  TFTP
  BOOTP
  TCP
  SNMP
  FTP
  SMTP
  
  Beside the protocols several diagnostic tools such as ping and traceroute are covered in detail. The book contains lots of diagrams, that illustrate the protocols even further. Each chapter ends with exercises. The solutions of some of those exercises can be found in the book as well. I give this book 4 stars out of 5.
• 
  Jon Swanson

  Best technical writing I've ever read.
  
  Stevens starts at the bottom of the networking stack and works the way up, explaining each component in a way that is both detailed and interesting. Each chapter is filled with lucid examples and makes you understand how all of the pieces come together.
  
  It's an older book, but still very relevant. Have yet to find a better text on the fundamentals of networking.
• 
  Plamen

  The book gives good understanding about basic network protocols (especially TCP). It is quite old and there are many new working protocols today which are not included.
• 
  Kam Yung Soh

  You want to know about TCP/IP? Get this book.
• 
  Vipul Ved Prakash

  A very rewarding book to read.
• 
  Kurt

  compulsory reading to work on the internet.
• 
  David

  The classic text. It both stands the test of time, and is also out of date (yeah, I know that doesn't make sense). I recommend Kozierok's TCP/IP Guide as a worthy successor.