Articles Weblogs Books School Short Cuts Podcasts  

What's Up at Uprizer?

by Richard Koman

In mid-April, Uprizer, the company founded by Freenet inventor Ian Clarke, took a $4 million investment from Intel. Terms of the deal included the development of Uprizer technology for the Intel platform. Just what is that technology? The Uprizer web site doesn't provide a clue. We spoke to Ian to find out what Uprizer is up to.

Richard Koman: Tell me about the technology you're working on.

Ian Clarke: Well, our initial application is a content distribution application, so the aim with that is, first, to prove our technology. This application is designed to generate bandwidth savings for customers.

The Internet, as you know, was never designed to be a broadcast medium, and consequently it's very expensive to distribute information to large numbers of people. Your costs generally end up being proportional to the number of people you're distributing to, and so we're looking at using the technology we're developing to dramatically reduce those costs. And obviously, in this economic climate, cost savings are a very attractive proposition to many content producers. And by content, I don't just mean music or video. That would include software -- really any information which might be distributed over the Internet.

Q: And I assume you're building on Freenet concepts to distribute information more efficiently.

Related Articles:

Free Radical: Ian Clarke has Big Plans for the Internet

Gnutella and Freenet Represent True Technological Innovation

A: That's correct. Well, Freenet "concepts" being the underlying word.

Q: Concepts, as opposed to code.

A: We are not building on the Freenet source code base, the reason being that Freenet has a number of important qualities. Probably the most important quality from Freenet's point of view is that it provides anonymity to both producers and consumers of information. Freenet, however, has some other interesting qualities. It efficiently distributes information. It's highly scalable. And it's really these qualities that Uprizer is building upon. And by removing the anonymity requirement, we can actually achieve much greater efficiency and much greater speed in the technology that we are developing.

Q: So the technical aspects of Freenet that create an opportunity for you are the scalability ...

A: Well, the scalability is very important. Obviously, an architecture which is not scalable will work really, really well for 100 users, but not so well for 100,000, and so we would end up hoping that we would not be successful, which is not particularly healthy. So scalability is really a prerequisite. The benefit that we are leveraging for the initial content distribution application is Freenet's efficient use of network bandwidth. Freenet does move information closer to demand, and it replicates information in proportion to its popularity. These are the characteristics we are looking at primarily for the first application, but for future applications we will be taking advantage of other benefits of the architecture.

Q: Where do you see your market opportunity for a product like this?

A: Well, really, anybody who is distributing large amounts of information on the Internet, so examples would be content producers, obviously people distributing audio or video over the Internet, people distributing software over the Internet, really anyone -- people who will be distributing games, any type of content that would benefit from this type of product.

Q: How would this interact with the Web, or would this be like Freenet, sort of a separate network?

A: Well, it would be a separate network to a degree, although it would play very nicely with existing protocols such as HTTP.

Q: Do you have any ongoing role with Freenet, or are you completely out of the Freenet?

A: No, I am still the Freenet project coordinator, and I have no intention of giving up that role.

Q: So I'm wondering, between your role as project leader of Freenet with your initial interest in censorship-free networks, and your role in Uprizer focusing on how to exploit these characteristics for commercial use -- where does your interest in censorship-free systems sit right now?

A: Well, it's in precisely the same place it was two years ago when I first started to work on Freenet. From the beginning with Freenet, it was my hobby. I was doing it in my spare time. The only thing that has now changed is instead of working for somebody else, I'm working for my own company during the day, but really nothing has changed. There's no contradiction -- any more than people pursue their political interests in their spare time and that doesn't inhibit their ability to go to work between 9 and 5 and do their job.

Q: And removing anonymity helps a lot in terms of making things more efficient?

A: Well, it has a number of benefits. I mean the anonymity in Freenet is a really strict requirement, and it forced us to use very, very heavy levels of encryption, which slow things down. It's like building a system with one hand tied behind your back. And so removing that restriction gives a number of benefits ... let's say you're a software distributor, and you want to distribute your software, it's very important that you know how many people are downloading it, and maybe some information about who those people are. So a software distributor is very unlikely to use a system that is anonymous because it would prevent them from knowing how many people had copies of their software.

Q: So your business opportunity is to charge content distributors to use the software to distribute more efficiently?

A: Yes. That would certainly be one potential revenue stream for our initial application.

Q: Are you providing some kind of mechanism for the content provider to set a price for the content ...

A: Oh, you're talking about a DRM solution. Our technology is kind of lower level from that. We moved the bits around. We're like UPS. If the content, if the content provider wanted to use a DRM solution, then our software would not prevent ... I mean our software basically interfaces with the client computer using HTTP. So if somebody wanted to put a DRM technology between our software and the end user, then they would be free to do that.

Q: Okay. So, and so you're fairly agnostic about --

A: We're very --

A: We're pretty much lower level than that. It's not really our concern.

Q: So you're not offering a full solution there in terms of, in terms of the transaction for, you know, if you want to get this movie it costs five bucks?

A: No. We move the bits efficiently.

Q: I saw one thing I wanted to ask you about in an InfoWorld story. We may have actually covered this, but the very last line says, "Clarke said he plans to license the software to third parties in favor of developing, at least initially, an organized development program for software developers." Can you talk about your plans for third-party developers?

A: Well, I mean, our initial application -- we're developing a platform, a library which is very, very flexible. But, of course, the challenge when you develop a generically applicable piece of technology is proving it. And so the point of our initial application is, first, to derive some revenue streams because these days profit is actually important. And second, to prove that our core technology actually works; and once we've done that it would certainly be our intention to license that to third parties. There's so many potential applications for it that we could never hope to create applications for all of them. So, we will be able to license our technology to third parties who will be able to leverage it into whatever areas they happen to specialize in. ... But our business is about creating great applications.

Q: OK, a platform. Can you describe that platform?

A: I think that's probably premature, at the moment, to discuss that. We're really just focused on our applications at the moment.

Q: Okay. But I should understand that what Uprizer is doing is more than building this one application?

A: Yes, that is our initial application, and we have further applications down the line, but this is our initial way to derive revenue and also our initial way to prove our technology, but there will be a diverse range of applications coming after this.

Richard Koman is a freelancer writer and editor based in Sonoma County, California. He works on SiliconValleyWatcher, ZDNet blogs, and is a regular contributor to the O'Reilly Network.

Return to