The Tim O’Reilly interview

I’d been hoping to inter­view Tim O’Reilly since starting work on the book. As the person widely recog­nised as having coined the expres­sion ‘Web 2.0′, I wanted to know more about what he thought of the way it was all going. He’s a nice guy to talk to, by the way. He’s better humoured, but also grumpier than a lot of people that talk to journ­al­ists. In my book, that’s a good thing. What tends to happen is that the people you talk to are so “on message” that you can’t see a per­son­ality behind that glazed smile. He’s even older than me, too, which always goes down well.

Did you invent Web 2.0 or discover it?

Neither! It’s a name attempting to point people at some­thing that existed. It wasn’t even me who came up with the expres­sion. However, it’s an idea that I’ve been pursuing since 1997. I started talking about ‘infoware’, which is much the same thing, at the same con­fer­ence [Linux Kongress, May 1997] that Eric Raymond started talking about The Cathedral and the Bazaar.

Many applic­a­tions and services that use the web as a platform (e.g. Writely) seem very dif­ferent from those that use the alleged wisdom of crowds (e.g. digg). Hasn’t it been mis­leading to call them the same thing?

Web 2.0 is a catch-​​all term, for sure. But when I talk about the web as platform, we’re talking about using the network as a platform. And that does include the examples you point to, albeit with dif­ferent emphases. This means a com­pletely dif­ferent approach to software devel­op­ment and to dis­trib­uting that software. We’re still getting used to that and adjusting.

In the same way, the original PC applic­a­tions were very much like main­frame applic­a­tions. It took a long time before we arrived at the idea of shrink-​​wrapped software you can buy in a regular store. In the same way, we’re still getting used to the idea of what Web 2.0 means. The people who realise where the leverage points are will win. There’s a shift in power from software APIs to big data­bases. The people who own the data­bases will win. Those data­bases might be records of people, or it might be devices, beha­viours or geo­graph­ical information.

A lot of critics of Web 2.0 ventures point to flakey business models built on CPC advert­ising. Is this a fair characterisation?

Focusing on the failure of com­panies and ventures is always a big mistake. It stops people making real progress and draws atten­tion away from what is suc­cessful. However, Web 2.0 is not about these bubble com­panies, it’s about the new approaches we are trying.

Most of the exper­i­ment­a­tion hap­pening now is wrong. But by having those exper­i­ments it means we are learning what dis­tin­guishes the sur­vivors. These new paradigms mean that there is a lower barrier to innov­a­tion. I think maybe the top ten of the Web 2.0 exper­i­ments that are big now will survive.

In any case, I think bubbles are a good thing. That’s how you get capital redeployed.

There’s a lot of con­tro­versy at the moment about paying the users of, or con­trib­utors to, Web 2.0 applic­a­tions. What’s your take on that?

It’s what we ulti­mately have to figure out. The applic­a­tions have to give the users a payback of some kind, whether that be in the exper­i­ence or the outputs they get from them, If the applic­a­tions aren’t working well in that way for users, then they’ll want to get paid in cash. There’s more than one answer.

To what extent do you think that Web 2.0 prin­ciples like com­munities, social net­working, openness and software-​​as-​​a-​​service will become a per­manent feature of the internet?

Communities and social net­working have always been with us on the internet and they always will be. However, I think that other things will change. It will become harder and more closed. It’s like when the internet first started, everyone was equal, then barriers started to appear. Access to data will become more guarded in Web 2.0, I think, and so there’ll be fewer, more powerful players as time goes by. That’s not so true of the software, where I think openness is a lot more important to success.

How long do you think the term Web 2.0 will last before we start talking about some­thing different?

I ori­gin­ally thought is was good for a couple of years. Now, I think it’s probably got another four to five years in it. There’s still a lot to talk about and learn.

Is that some­thing dif­ferent the semantic web?

Hmm. Before we had the web, there was going to be some­thing called Open Systems Interconnect (OSI). It had been researched by all the top aca­demics and was mandated by the gov­ern­ment. It was a lot more com­pre­hensive and clever. There wouldn’t have been things like 404 errors or out-​​of-​​date pages.

Then came along this crappy thing called the internet. And, as it turned out, though it was inferior to the OSI in many respects, it was good enough for most people, and as we know it’s never looked back. I think the aca­demics think way too hard about these things. In a lot of ways, worse is better.

That’s not to say that I think the semantic web people have got it wrong. They have a lot of ideas that are right. However, I believe that Web 2.0 is already the semantic web. We are building meaning into the pages. Ultimately, people will solve the problems that need solving and ignore the little things that don’t bother anyone. Only the solu­tions that offer value to lots of people will be propagated.

As a pub­lisher, doesn’t this boom in self-​​publishing make you uneasy?

Inasmuch as it does threaten what we do, so that has to change. People buy much fewer ref­er­ence books than they used to, so we don’t publish as many. But we’re also inter­acting with this movement. We’re doing more to build inter­activity into our books, even building Web 2.0 apps ourselves to extend the book exper­i­ence onto the internet. Make magazine is closest to our new model. That was put together by people we found on the internet who had some­thing new and dif­ferent to share. We can help them do that. We’re also exper­i­menting with using internet wikis as a way of putting together books.

The role of the pub­lisher is in selecting and adding value to inform­a­tion, and the need for that won’t go away. Our business is changing the world by spreading the know­ledge of innov­ators. However, as with all this stuff, there’s going to be a period of upset and dis­rup­tion before we discover the new rules.

Share this post:

Possibly related:

• The Robert Scoble interview
• Blogging and Drinking
• The new media interview