O'Reilly has had a hard time with their computer security lineup since they started expanding it a few years ago. While they have tried to focus on tight subjects with short volumes (this book at about 220 pages is no exception), these books often wind up being cursory treatments of the subjects, and in some cases downright wrong.
Sadly, Internet Forensics is not an exception to this rule. While I like this book more than some of the other recent O'Reilly security books, that isn't saying much. We've come to expect clear, authoritative books or inspired tricks and tips type martial from O'Reilly's authors, and instead we're given unfocused, incomplete pages.
To be fair, the topic of Internet Forensics is broad, not very well focused, and no one has written a good book on the subject. It's coming into the foreground, especially in this past year, as threat analysis has become popular. This is a new, wide open field, covering a broad range of malware, spam, phishing, and malicious website analysis coupled to tracking the origin and leading to takedown of the materials. However, this book doesn't really do a good job of much of that. And, at the end of 2006, some of the material feels positively quaint (even though it came out in late 2005). Although the author has defined his target audience in the introduction (infosec professionals, and software developers and IT operations people), I don't think they're well served with this offering.
Chapter 1, an introduction to the book, is short and scattershot. Nothing promised in the preface is really delivered (no overview of spam, phishing, or other threats). Instead, it's just some writing with little focus. This tone carries throughout the book. Chapter 2 covers the basics of IP addressing (what the heck?! if you don't assume your readership knows this, they're in the wrong place), and then talks about DNS lookups with dig and whois. The people reading this should know how to use these tools already, where are the suggested requirements for the reader? Sadly, no tips on disambiguating whois results (p 22) are given, not an unexpected finding in this book. And we start with the inefficient Perl scripts, too. All in all, we're not off to a good start.
Chapter 3 covers email, and sadly we waste time on the basics of email headers, and then go into making very good use of them. The coverage here is inconsistent and again, unfocused. By the time you finish chapter three with "is it really spam?", you're left wondering what the heck the author wanted you to learn. Chapter 4 is slightly better, focusing on on URL obfuscation. Sadly, none of the techniques given really hold up all that well any more. Again, we start with some basics and try and get somewhere, but along the way we're distracted and we've never really gotten a good sense of what's the objective.
Chapter 5 on websites tries to cover some ground, but again, it's too unfocused. We talk about mirroring a site (why "wget -r" isn't listed, which is a common way of getting a malicious phishing site or directory, I don't know) and we even talk about SQL injection, but I don't know what the author is really after. It feels like random observations thrown in with no overall goals. Chapter 6 talks about web servers, and we talk about headers and redirection, and then delve into Netcraft stats (why?) and honestly I'm not clear what was useful here. This felt more like introductory material than anything useful. If the readers are infosec professionals, they should know what a web server header looks like and how to properly fingerprint the server.
Chapter 7 is the complement to that, and talks about your browser. Again, some useful info, but it's incomplete. No real discussions about why you want to alter things other than some basic concepts. Chapter 8 talks about file contents, and there's some interesting basics on examining Word docs (track changes, strings, etc) but aside from some basics, there's not much great there. Sadly, no discussions on how to un-redact a PDF are given, just that it's been a problem.
Chapter 9, which is a nice departure from solid technical materials, comes up short. It's incomplete and disappointing. Chapter 10 talks about pattern detection and signature creation, but again, this could have been beefier.
Chapter 11, "case studies", is OK, but some better treatment to tie the lessons learned (or hopefully imparted) would have been nice. Finally, Chapter 12, "taking action", isn't very useful. No real great info or insight is here, and if you think that you'll be calling police departments about every phishing site, you're in for a sad wake up call -- there's just no way you can do that. One of the comments made in this chapter, specifically wanting to see a community response, tells me that the author (Jones) isn't well connected to the community that actually does track and respond to these threats.
Internet Forensics is a poor attempt at this broad subject. While I appreciate the scope of what the author is trying to do, the execution is weak and suffers from a lack of focus or discipline. A book twice this size covering a fraction of the material, well executed, would have been a better offering. If you feel you must get this book, make sure you get it at a steep discount.