Tim O'Reilly is the founder and CEO of O'Reilly Media, Inc. O'Reilly also publishes online through the O'Reilly Network and hosts conferences on technology topics. Tim is an activist for open source and open standards. We recently caught up with Tim at the MySQL Users Conference in Santa Clara, California.

Edge: What are your thoughts on digital rights management, and how do you see DRM being managed in the near future?

Tim O'Reilly: There's a couple of different layers to that question: What do I think is actually going to happen? I have no clue. What I believe should happen and what the indicators suggest will happen is very different....I believe that loose DRM really works. I believe that a good DRM package will let people set any policy they want, but a wise policy is to make it as minimal as possible.

For example, I recently read a blog entry about Apple versus Sony. Sony's music DRM was very strong, and led to a lot of user restriction. With iTunes, Apple made DRM relatively invisible (there is DRM there, but it's not very hard to get around) and has ended up selling a billion dollars worth of songs.

Somebody responded to the blog entry, and wrote, "DRM is like taking your cat to the vet. It's not like taking a dog to the vet. When you take a dog to the vet you hold him tight, and when you take a cat you better hold him loosely or he'll claw you." That's a nice metaphor for how people need to think about DRM.

Edge: How does O'Reilly Media use DRM?

Tim O'Reilly: At O'Reilly, we have our Safari online books formatted to include the ability to download the books in PDF. They're watermarked, but they're not protected in any other way. We also have some anti-spidering protection to try and prevent people from sucking down a lot of the content. Some people do anyway. And then we receive an e-mail from a user saying, "Hey, I discovered copies of your book on this website in Russia or Romania, and I wanted to let you know about it," because that user has a relationship with us.

I think in the world of Web 2.0, that kind of relationship and respect between customer and vendor is really critical. If your customers value what you do, they're not going to rip you off.

Edge: What do you think about the controversy surrounding Google Book Search?

Tim O'Reilly: What's really interesting about Google Book Search is that it would make a lot of content available, content that would otherwise remain obscure and undiscovered. Google wants to build an index, and I say more power to them... I look at Google Book Search as a gift horse that the publishing industry is looking in the mouth. We really need to think hard about how to increase the liquidity of information markets, not decrease it.

Edge: Despite its downfall, would you say Napster helped create new market opportunities for online music services?

Tim O'Reilly: If Napster hadn't existed, iTunes wouldn't have existed either. The fact that it was really easy for people to rip their own CDs, I think is the reason why online music sharing and online music services have taken off... By contrast, look at eBooks. EBooks are something that's promised but never arrives. Why is that? Because it's expensive and hard to make digital copies. I think if it were as easy to rip a book as it is to rip a CD, we'd have a thriving market for online books.

I think we really have to figure out how to grease the wheels of free reading distribution, because ultimately, once people get accustomed to having that digital good, they say: "Wow, this is valuable. It's useful to me." And then they are willing to pay for it.

Edge: What do you think about Web 2.0 and the need for a universal runtime? Specifically, how important is the need for technology that operates across operating systems as well as devices?

Tim O'Reilly: There are two answers to that question. It's pretty clear that the web teaches us that we don't need a universal runtime. The PC and the browser combination provided enough of a framework that even though it was a lightweight runtime, it took the place of a universal runtime. However, when you get to devices, it's really clear that we need a universal runtime.

If we're ever going to see the potential of the web on phones and various kinds of handheld devices, we have to find a solution that's a) open, b) universal, and c) allows people to get outside the stranglehold of carriers.

Edge: Do you believe Flash® has the capabilities to become that universal runtime?

Tim O'Reilly: I think the challenges are less around whether Flash has the capabilities and more around whether, for example, cell providers will really open up the capabilities of that platform or whether they are going to try and create these walled gardens because they are concerned about their revenue model.

Right now, what the providers want to offer is gated by what they think they can charge for it, and so we don't have the kind of open innovation we've had on the Net. I think if we were able to get there, we'd get a whole new level of innovation on phones.

Hopefully, we'll get to a point where the carriers will get on board. The hidden ally for Flash and the universal runtime is called Voice over IP, because if the carriers don't get there, people are going to start routing around them. I think that's going to create some interesting pressure for the carriers to open up and for people to discover that value.

Edge: Do you feel reliant on your devices?

Tim O'Reilly: There are a lot of special focus devices that I rely on. I wish they'd talk to each other better. One of my favorite devices right now is very special-purpose. It's a little solid state traffic monitor; it's just for the Bay Area. It's always on. I pull it out of the glove box of my car. It has a very crude map of the Bay Area, and the highways are either clear, or they've got dashed lines on them, or they have flashing dashed lights on them. When they've got flashing dashed lights, then you find another route.

I also have a Garmin GPS in the car; I really like that. Before, when I went somewhere, I'd go to MapQuest or Maps.yahoo.com and print out a map. Now, I can just see it. So that's a handy device. But once again, I shouldn't need two separate devices to check maps and local traffic.

Edge: Do you think we're getting closer to having one general-purpose device?

Tim O'Reilly: What I think we're seeing right now is we have this vulcanization in the device world, and looking ahead, that's got to break down. There are going to be too many devices, and you see people walking around with big backpacks full of stuff. It doesn't make any sense, because really, we need a general-purpose software platform that lets all of the applications work on one device.

Edge: What do you think about San Francisco providing free Wi-Fi access through Google?

Tim O'Reilly: One of the most interesting things going on right now is the question of what business models will prevail. I think it's a real challenge for software companies because they're used to a model in which people pay for software, one way or another... Google is totally fueled by advertising, and they are delivering services that people used to pay for. They are working at all kinds of services that once might have been monetized through software purchases or software subscriptions, but now they may be free and monetized by advertising. This is the same in content.

I was talking with Sergey Brin about Google Book Search and telling him how I wanted Google Book Search to feed into our subscription-based service, and he was like, "Yes, but what if we could do it with advertising? All this material could be free for people in third-world countries who can't afford it." Maybe they can get advertising to work for books as well. It's certainly possible that many services that used to be something you paid for explicitly, you'll pay for indirectly.

Edge: What are people willing to pay for?

Tim O'Reilly: At our Web 2.0 conference last year, probably the biggest take-away that I had was something that Michael Powell, the former chairman of the FCC, said on a panel. He said: "I don't get it. Everybody says that they don't want to pay for music, but my kids are spending $40 a month on ring tones." His point being that a ring tone, which is like a little sample from a song, is now worth more to consumers than the song itself.

There's this wonderful essay, "The Birth of the Big, Beautiful, Art Market," by Dave Hickey, who's a MacArthur Fellow and an art critic. He describes how General Motors, after World War II, stopped selling cars for what they do and instead started selling them for what they mean. That's his definition of an art market. I thought about that, and immediately I thought of Apple and their "Think Different" and "1984" commercials. They've always sold their computers on the basis of what they mean: What does it mean to be an Apple user, as opposed to what's the benefit of using an Apple computer? It's not so much about functionality; it's more about design and brand.

There's a lot of evidence that people are willing to pay for design. Coming back around to the ring tone story, what I took away from that was: Oh, people are willing to pay for ring tones because they're about display; they're about showing off; they're about fashion. Ultimately, you're not paying for a song; you're paying for what it means to the people around you. That way of thinking is part of what's driving the market.

Edge: Now that Adobe and Macromedia have merged, what do you think Adobe's contribution to Web 2.0 will be?

Tim O'Reilly: Adobe has the opportunity to liberate the class of document that's normally created on paper — whether it's a book, or a magazine, or a research report, or product literature — and say, "How do we make that document become a more mobile part of the information economy?"

It goes back to our discussion on DRM earlier. I think Adobe has an opportunity to help people realize that creating liquidity in information markets actually increases value.

If I were Adobe, I'd be thinking hard about how to manage PDFs better. There's this enormous opportunity as people create things for print, to say, "Oh yes, we create them for online as well." How do we make that experience better for users, as opposed to making the experience safe?
Shawn Fanning unleashed the online music revolution by setting the default as "Make this shareable," as opposed to "Keep it private." I think there's a really interesting opportunity to say, "What does the economy look like as we start to create richer documents?" not just PDFs, but also thinking about Flash.

You think about multimedia content — Flash has this amazing rendering engine for video; it has the ability to be on more devices than anything else. What is the opportunity? How do you let people get into the content? How do you make it possible for people to index it? How do you make it possible for people to make a reference to a particular segment in a video? If you're thinking about PDF, yes, you can put a PDF up on the web, but you can't put up a URL that points to page 134 of a 300-page PDF. We need to start thinking about those things. How do you point into these documents, how do you point into these rich media types?

I think there's a real opportunity for Adobe to lead the way in taking these enclosed document types and figuring out how to give people more granular access to that information and also the ability to remix it.

Edge: What does ubiquity mean to you?

Tim O'Reilly: It means that you can take something for granted. If it's everywhere, you don't have to worry about it. Also, when I think of ubiquity, I think of standardization.

For example, let's look at cell phones. It's really surprising when you're driving on the freeway through San Jose — a place where you would expect a really strong signal. Why is the coverage so spotty? It's because the carriers don't want to share, and as a result, the user experience suffers. One of the things we see is that ubiquity requires interoperability. The technologies we have that are most ubiquitous are those that are most interoperable. Ubiquity is achieved when multiple providers play by the same rules.

Edge: Looking toward the future, what advice do you have for web developers to stay ahead of the curve?

Tim O'Reilly: The first thing I'd say: Developers need to be agile. What we're seeing on the cutting edge of application development is that people are getting very comfortable with running and gunning and writing applications quickly. Look at Cal Henderson and Flickr. He was pushing a new build out on the Internet every 30 minutes. Corporate IT development shops needs to wake up and smell the coffee.

A lot of IT shops are going to be out of business because there's going to be some kid who says, "I can plug together this web service and that web service and in a week have something that works." And behind door number two is some complex IT process that's going to take a year or two and five million bucks.

In fact, I was just talking to an entrepreneur the other day who's saying all those young web developers are just starting to wake up to corporate development opportunities. They're the equivalent of what Microsoft put in place with its channel of 800,000 ISVs.

There's an opportunity to build a new kind of ISV channel in this sort of lightweight development space. I think the first thing is to think agile and the skills that go with that — obviously web development is fairly central. That's not to say that hard-core software development is passé. At the end of the day, you still need to develop a deep infrastructure that supports all the lightweight stuff, but I think that that will be sort of a smaller part of our ecosystem.

Edge: What skills are becoming more essential for application developer?

Tim O'Reilly: First, as I mentioned above, agility is important. The other skill that I think is really important is data management. I had a conversation with Hal Varian, the noted Berkeley economist, who also does consulting for Google. He made a comment that stuck with me: "SQL is the new HTML." I think database competency is going to be important because so many of the applications that matter today, and will matter tomorrow, are about managing data.

Design also matters. At the end of the day, people value applications that are clean and stylish....I think Adobe has some interesting advantages in this area because there is so much design heritage in the company that, in some sense, I think there's an opportunity to bring some of that design expertise into the software and application development realm.

- Visit the O'Reilly Radar for the latest posts from Tim.

Julie Campagna manages newsletters for Adobe Developer Relations and serves as managing editor for the Edge newsletter.