Why archive the Internet?

Two reasons: one is to give historians and scholars an idea of what happened at the birth of this medium. The other reason, which we think is more important, is to provide services that can thrive from having this much information on the popular voice, services that can make for a deeper understanding of what we were thinking about.

What kind of services?

An archive is mainly about gathering and storing, which has its own value, but unless that information is put to use, then it's little more than storage. AltaVista is an interesting example of a service. It gathers, stores and indexes what's currently available. What we will have is not only the current presence of the Web, but also past copies of the Web, as well as the graphics. Whereas most people use AltaVista to find their way around, people will use the archive when they want to consult a deeper collection.

What does it mean to say the Internet is disappearing?

One study showed that a document lives an average of 44 days on the Internet. As a result, what the Web looked like a year ago is basically unknown at this point. We have anecdotal evidence and people's memories about what was here a year ago, but we don't know how it's changed on a concrete basis. For instance, the presidential campaign Web sites are almost all turned off at this point. Once a candidate drops out, his page disappears. If the Internet does become an important medium, then this record is important to preserve.

What challenges do you face in creating the archive?

The biggest challenge is to make it useful, but there are also technical and legal/social challenges toward building the archive. The technical challenges revolve around gathering information from hundreds of thousands of locations, storing them effectively, and then offering access in some timely way. Those are relatively surmountable. The social and legal issues get into murky areas about what is legal, and what's right. Questions about copyright, privacy, import/export, pornography all come up when you start to gather this much information.

You've said the archive is not just for posterity, but for profit. Where does the profit come in?

One way is if organizations can make use of the data to provide services that are useful. The other way is if we can learn to do this type of project--to build (data gathering and access) technologies that are useful in other domains. . .small companies may have terabytes of information they want to serve on Wide Area Networks. So there are two different directions: one, developing a digital library of popular voices, and the other, of building technology to manipulate terabytes of data on Wide Area Networks.

When historians look at the early Net, what will they see?

This early Internet has attracted a great deal of enthusiasm and optimism. People are trying to solve problems they've had for decades in distributing information through the existing power structure. . .people think it's bringing democracy and opportunity. They are dreaming this to be the solution to all sorts of problems. . .they are experimenting and trying all sorts of different services. In five years, it'll largely be settled. It'll be ad-based or pay-per-view or however different segments work out. But it will be largely understood and be an ëof course.' Right now, it's experimentation and exploration.

Do you remember when you first realized the Net would be something more than the next video game platform?

It was driven home concretely when (in 1989) I proposed a test project called WAIS to Dow Jones, Apple and KPMG Peat Marwick. All those players said, ëYes, we're interested. Let's do it.' That was a turning point for me in understanding that this would be a major industry.

Where do you see the Net going?

(Long pause) That's a hard question. (Long pause)

That's a valid answer. Anything more?

Wow. I don't have a good answer for that. I guess there are some limitations that need to be fixed if the Net's going to fulfill the promise of being a new-style publishing medium. Three limitations come to mind. There's the World Wide Wait, the perception that the Net is unreliable and also that it's impossible to find information that's appropriate.
I hope that the caching infrastructure that's being developed will help with the Wait and with bandwidth. The latter two--reliability and navigation--can hopefully be augmented with the Internet Archive.

You've said the Web might not be the Net medium of the future. What did you mean?

We will see news services that are more server-push oriented. . .the information will flow over desktops based on our own private [needs]. The infrastructure may be a new set of protocols. But I do believe the Internet infrastructure (in general) is doing quite well at a data transport level. In the future, most people will not search for information. They'll watch what comes over the desktop, and throw it away at relatively different speeds.

What's your favorite Web page?

www.firefly.com--in May of 1996. They had a wonderful visionary essay on where this technology can lead. But now that (Web page) is gone as the company is focused more on commercial concerns.

Given your early adoption of the Net and the fact you devoted so much education to it, are you gratified to see mass adoption of the technology?

This is a wonderful time to be alive. Very few generations get to see birth of a new medium--whether printing press, or telephone, or Internet. It's a chance to see changes in how companies are structured and even how families relate to each other. The opportunity to see not just how it will unfold, but to participate and to see the technology developments is more than a person can ask for.