Saturday, August 13, 2011

Migrating Applications to IPv6: A Review

Paperback: 50 pages
Publisher: O'Reilly Media (July 7, 2011)
ISBN-10: 1449307876
ISBN-13: 978-1449307875

This wasn't what I expected.

I have an interest in migration from IPv4 to IPv6 for reasons I can't really explain right now, but I thought that reviewing Dan York's book Migrating Applications to IPv6 would hone my thinking and expose me to a different way of looking at IPv6. After all, most of us think of implementing IPv6 on an internetwork. We think of the expanded IP address space and the changes in DNS records, how routing mechanisms will change in IPv6 implementations and so on. We don't really think about how the change will affect applications. You might not have thought that IPv6 would change anything about how you write applications. Dan York's book gets you to think again.

But in only 40 pages?

Actually, the content of the book is only 34 pages. That's like a short chapter in just about any other IT book I've ever read or reviewed. Something made me decide to check around at other books on IPv6 recently published by O'Reilly. There's a trend.

Cricket Liu is something of a mainstay at O'Reilly in terms of her DNS and BIND book (in its 5th edition as of June 2006). Her DNS and BIND on IPv6 book was published at the end of May this year. It's only 52 pages long with about 37 pages being actual content. Planning for IPv6 by Silvia Hagen will be published at the end of August and it's a total of 82 pages long. If you put these three books together, you might have the start of an actual IPv6 book based on the most recent IETF RFCs for IPv6 implementation, but I'd certainly expect more to round out the subject.

Each of these three books runs a little over $25.00 each, but you could also buy Hagen's IPv6 Essentials book (published in 2006), with a full 448 pages of information, for only about ten dollars more. I looked through the front matter of York's book trying to find some reason for it's "mini-format". I couldn't find anything. I've never seen a complete book from O'Reilly before that was so thin on content. I had expected an in-depth treatment of the topic but what I found was a "pamphlet" of introductory ideas, tips, and tricks.

York's book does address, in a very compressed space, a number of issues that you will need to consider as an applications developer, and they all have to do with how your application accepts, manages, and stores IP addresses and DNS records. If your app requires that an IP address be input, does it do so allowing enough space for IPv6 addresses? If your app manages DNS records, will it be able to accomodate both A and AAAA records (IPv4 and IPv6 respectively)? Does your app expose any APIs that have an IP address format dependency? As a developer, if you haven't asked yourself those questions before, this book will help you find the answers.

Is the information in this book useful? Yes. I found the content, what there was of it, well written, insightful, and knowledgable. Is it worth 25 bucks? That's hard to say. If I wanted information on application migration, DNS and BIND, and general IPv4 to IPv6 migration planning, would I pay over $75.00 (for the three books I just mentioned) for a total page count of only 174 pages, no matter how well written or useful, when a full-sized book of almost 450 pages is less than half that cost?

Sorry, but I'm having a tough time trying to figure out what O'Reilly's strategy is here. When I requested a review copy of York's book from O'Reilly, I didn't give the page count a single thought. Now, it's all I can think about.

What happened, O'Reilly?


  1. Brevity is a virtue? :-)

    Seriously, though, as an app developer myself, I wouldn't expect a whole lot here. Most apps use a standard library of sorts, usually shielding us from manually storing or parsing addresses. Seems to me this might be more fit for operating systems or other low-level programming.

  2. I can see your point Judah, but according to the rep at O'Reilly, I'm not the only one to register surprise.

    As it turns out there is a strategy involved. This book is part of a series that attempts to get information to market as quickly as possible. I know as an author how many months it can take to write, edit, and finally publish a "full size" book, so I can see how creating a work only 30 or 40 pages long would accelerate the process.


Please make comments.