An Interview with Tim O'Reilly

by Frank Carver

I'd like to introduce Tim O'Reilly. Tim O'Reilly is the founder and CEO of O'Reilly & Associates, thought by many to be the best computer book publisher in the world. O'Reilly also publishes online through the O'Reilly Network (www.oreillynet.com) and hosts conferences on technology topics. Tim is an activist for open source and open standards, and an opponent of software patents and other incursions of new intellectual property laws into the public domain.

I've caught up with Tim as he grabs some lunch before hitting the podium to talk to the geeks and eggheads at British Telecom's sprawling Adastral Park research site.

Let's start with a question on many of our minds here at the 'Ranch: How's the Java book market going, and in particular what about Kathy and Bert's "Head First Java"?

Over the last few years, many Java books such as "Java in a Nutshell" have been big sellers for us, but we have seen a drop-off in the current economic situation. Programming books have been especially hard hit, with a lot of the book dollars moving to other subjects -- new operating systems like Windows XP and Mac OS X, and new application areas like digital photography. For example, Java was the #1 book category for one of our UK distributors last year, but the #16 area for them for the same period this year. "Head First Java" is an amazing book, but it's too soon to tell how much of an impact it will have. I'm sure that if we'd released it a couple years ago it would have been the #1 computer book bestseller, which would have put it on everyone's radar right away. Now it will just have to settle for being the #1 Java book :-), which means that it will take a bit longer for this approach to take over the world.

What's exciting is that the Head First concept is applicable to far more than just Java. It's a whole new approach to teaching complex subjects in which mastering key concepts shapes subsequent learning. It can make very difficult subjects interesting and engaging, and can help even experienced developers to understand their tools more deeply. The question now, is what other areas can we give the "Head First" treatment. I'm open to suggestions! We're doing a lot of brainstorming with Kathy and Bert, because we believe they've come up with something really brilliant that can be used in so many ways.

How about other alternative approaches? The "hacks" series?

Oh, the "Hacks" books are doing really well. "Google Hacks" was a number one seller for something like four months. Linux Server Hacks and Mac OS X Hacks have also been bestsellers. We have a new wave coming out this fall, including Amazon Hacks, EBay Hacks, Wireless Hacks, and Windows XP Hacks.

For those of you who don't know them, the "hacks" books are full of short, fun, things to do, things that are both useful and catch the imagination. A book about Google might seem strange if you think of Google as just a web site. But it's not, its part of the new paradigm of software on the internet platform. You can't buy it or download it, but you probably use it every day, and there's a big difference between how beginners use it, and how experts use it. And that spread is what we thrive on.

Thanks. That leads quite neatly to another of my questions. How are you tackling the economic slowdown and the challenge of the internet? I've heard several people say that they don't buy tech books any more, it's all in Google!

Of course there's been a change in buying patterns, but I couldn't say how much of that's due to the economy or the internet. Maybe in the long term we'll be able to look back and see trends. Reference books are down, but tutorials are still selling. People still like to have tutorials you can read in the bath. And people are always looking for succinct collections of expert knowledge -- it takes a lot of searching to accumulate the kind of depth that we have in our books.

We are also at the forefront of technical publishing on the Internet, with both Safari and the O'Reilly Network. The O'Reilly Network is a free ad-supported network of online technical information sites, including www.oreillynet.com, xml.com, www.perl.com, and many others. More recently, we've started doing some "private label" online information sites partnering with corporate customers. The first of these, and most relevant to your audience, is java.net, which is sponsored by Sun. O'Reilly is producing the site, and developing all the content, as well as managing weblogs and other forms of user interaction. Over time, we may add some subscription-based sites to the O'Reilly Network as well.

Safari (safari.oreilly.com) is a subscription-based reference library online, containing well over a thousand books from O'Reilly, Addison-Wesley, Prentice-Hall, Peachpit, Cisco Press, Macromedia Press, Adobe Press, and many other leading publishers. Interestingly enough, I got Pearson (the UK company that's the parent of all those non-O'Reilly imprints listed above) interested in partnering with us on Safari using an argument based on the ideas of "crossing the chasm". You start by gaining a small beach-head in one area, dominate that market, and then grow out to other markets. And the market we originally targeted to dominate was online publishing for Java. What I saw specifically was that if you had O'Reilly, Addison Wesley, and Prentice Hall books on one online site, you'd effectively have almost all the Java books that mattered. So we hoped to make this a must-have service for Java developers. Of course, it grew beyond that, and now, between O'Reilly and Pearson, Safari pretty much covers the computer tech books field. Right now the service mirrors the structure of existing books, but what we've created is a huge database of high quality technical content. Over time, it will become possible to offer services based on that content, services that go way beyond just reproducing the books online.

Wow, So what's your opinion on the advantages of the screen vs paper? Can you see publishing moving on to the internet completely?

Well, there's no way that paper publishing will go away, at least not till we have "books" with "e-paper" that can be reloaded electronically with different texts. The user interface of the book is just too good for many types of task. That being said, though, what we really need to do is to ignore formats, and think about functions. A "book" is a form factor, but it covers a lot of different ground -- reference works, tutorials, entertainment. And translating those functions to a new medium will mean massive changes to how we think about electronic publishing. Most of the current eBook efforts are doomed to failure because all they do is try to reproduce what happens on the printed page. It's like claiming that the way to make a movie is to point a camera at a stage play. We know that's crazy. The medium has way more possibilities than that.

We're already well beyond eBooks for many of the functions formerly performed by books. MapQuest and sites like it have put a lot of pressure on Atlases and maps; google and the entirety of the web put encyclopedias out of contention. Amazon has replaced the specialized "Books in Print" as the definitive source of bibliographic information. And can you deny that Everquest is successor to the fantasy novel? One that's presented in an immersive way that makes use of what the network platform has to offer. It makes the idea of an e-book reader with the straight text of "Lord of the Rings", or "Harry Potter" seem very weak in comparison. And I'll note that forward-looking authors like Tom Clancy not only write books, they produce movies and have companies to build computer games based on their characters and stories. George Lucas gets novelised versions of his movies, and has games like Star Wars Galaxies to boot. All of that is why I'm saying that eBooks per se are too limited a concept, and that what really interests me about Safari is how much more we can do once the technical information database it represents gets rich enough.

I think we're running out of time here, but have you got any general comments?

If there's one thing you need to remember it's not to miss the important things by taking them for granted. A lot of people, even a lot of smart people, are stuck in the old paradigm of software as something that is delivered on a CD or on a personal computer. The real "killer apps" of the internet are not like that. For example, the fundamental paradigm shift represented by the internet is missed by many people in the open source software community. The killer apps of the internet are all built on top of Linux and use many other open source tools but are not themselves open source. A licence like the GPL which lays down what you must do when you distribute your software is simply never triggered. The code for Amazon is not open source, but it wouldn't be much use even if it was. The value in the system is not in the proprietary source code, but in the data and the network effect created by the large user base contributing to that data.

The applications that are winning in the internet arena are the ones that make use of open source software to drive their costs down, but then harness the power of a collaborative community to help build their product on top of that open source base. They make money and survive not by selling software but by selling user benefits and encouraging the users to help their system grow. Every time you browse from one book to another, or write a review, or any of the things you can do with Amazon, you increase the quality and usefulness of their site. Every time you add a link on your web site you add to the power of Google. And this is a constant process. Microsoft may be on a three-year release cycle, but Amazon and Google are updated every day.

Thanks very much, Tim. I'll let you get back to your lunch, now!

Thanks.