Beyond File Sharing: An Interview with Sharman Networks CTO Phil Morle
The legal and social debate over the legitimacy of peer-to-peer (P2P) applications has made headlines since the days of Napster and Hotline -- and continues to be a major issue today. While there are certainly other hotspots in the tech industry, the P2P debate serves as a focal point for some of the most important issues of the day, such as security, personal privacy and copyright law.
On one side of this debate are applications like LimeWire, Morpheus, Bearshare and Kazaa, whose makers say their products have many uses and that they promote the legitimate ones and are not responsible for any illegitimate activities. On the other side are entertainment companies that see those products as taking revenue from music labels and artists. The two sides have waged an ongoing war in the courtroom, which continues unabated and in which consumers now find themselves as participants.
One can argue about the legality of such applications, but there is no debating their popularity. Sharman Networks' Kazaa P2P application has been downloaded more than 315 million times, making it -- according to Download.com -- the most-downloaded software in the world. Clearly, demand for these applications is still strong, despite the Recording Industry Association of America's push to squelch P2P file-sharing activities.
TechNewsWorld turned to Phil Morle, CTO of Sharman Networks, for an exclusive interview to talk about the future of file sharing and the latest developments in the world of P2P.
TechNewsWorld: Please tell us a little about Sharman Networks and what your role is there.
Phil Morle: Peer-to-peer software is an evolutionary development and not a revolutionary one. It is a very natural direction for technology to take. Every day, more people move more files, and these files get larger. Simultaneously, the same people own increasingly powerful computers that are underutilized.
Network resources are like any other resource that is needed by society, but if the Net were our water supply, our reservoirs would be depleted more quickly than we could build them while the entire population would have enormous, unused tanks of water in their backyards that stagnated. Sharman was formed to connect these water tanks to ensure that no one goes thirsty and the cost of water remains low.
Our main priority has been on solutions for media distribution. This area has seen explosive growth in file sizes and the demand for files. As media companies put their digital catalogues online, they are going to need the resource capacity to meet the demand. By using peer-to-peer for both streamed media and direct downloads, online distributors can save about 90 percent of the cost of traditional methods -- and in many cases can offer more robust, faster delivery through the intelligence of the peer-to-peer network.
We have been proving the model for the past two years, demonstrating that bands can make it big, movies can be released and game publishers can do excellent business over P2P. My role as CTO is to collaborate with our international technology group and the other areas of our business to develop Kazaa technology to meet Sharman's vision for P2P.
TNW: What would you say the biggest technical challenges are right now in the world of file sharing?
Morle: The benefits of distributed peer-to-peer networks create unique challenges that are not present in centralized environments. Users of the Kazaa application itself -- and not Sharman Networks -- manage the files on the P2P network, and our challenge is to give them the tools to manage it effectively. The network can only be created, organized and made safe by the users.
For example, Kazaa is the only file-sharing application with integrated antivirus technology. Using P2P technology, we push the latest virus definitions to users at an incredible rate. Users effectively give the definitions to each other. Automatically, they protect each other from viruses and clean the experience for other users at the same time.
We are making good progress on the security challenges. We are also looking at collaborative filtering techniques to empower users to manage the files that are available so that they are collectively responsible for removing bad files and categorizing and labeling good files effectively.
An example here is the metadata in digital photographs. There are billions of photos out there waiting to be shared by their creators, but they are lacking detailed metadata. We are looking at ways to facilitate fun and easy management of metadata by all users. The more complete the metadata, the better the search experience.
TNW: What about the most exciting technologies right now?
Morle: I am most excited about magnet links at the moment. Read more about these at magnetlink.org. Magnet links are an open standard that we have adopted to allow bloggers and other Web site creators to add P2P links to their Web sites or as links in an e-mail.
For example, if a blogger creates a short film but has a very limited bandwidth allowance, he or she can add a magnet link enabling consumers of the blog to download it from other peers using a P2P application like Kazaa instead of downloading directly from blogger.com. This breaks out file sharing beyond the application into the Internet at large.
When we released magnet links, we also released Kapsules. We are only just beginning to imagine the possibilities of these. They are basically ways of collecting distributed files together. When a user downloads a Kapsule, they have all the information they need to download all the files referenced in that Kapsule automatically across multiple sources. A client could try to search Kazaa for each file and -- if it fails -- try another P2P network, and if that fails then fall back to a Web server. With Kapsules, a world dominated by individual files gains the concept of collections of files, presented in context. We are considering making the Kapsule file format an open standard.
From this we can extend the application of Kapsules into a way of digitally packaging a collection with a multimedia presentation layer and value-add functionality, distributed playlists, distributed slide-shows and so forth. On a larger scale, we are also seeing the emergence of the next generation of P2P network layers and really exciting products that will use those layers.
I think 2004 will be the year that P2P becomes more than file sharing. Skype is a superb example of a P2P technology of this kind being applied to create tremendous value to users and businesses. Skype is a completely distributed VoIP solution built completely on P2P technology from lessons learned building Kazaa.
TNW: Would you say that Morpheus is your biggest competitor right now? Are there other clients, like LimeWire, that you admire or are partnering with for future developments?
Morle: Morpheus ceased to be our main competitor, or indeed a trusted solution for users, a long time ago. eDonkey/Overnet is a strong competitor. They have cool technology and millions of users. I am consistently impressed with their releases that show a passion for P2P and what it can accomplish.
I am also surprised that Shareaza stays under the radar. Mike, who develops it, is a very ambitious young man, and I have to wonder when he sleeps. Shareaza is a well-made application that is a pleasure to use. His singular vision shines in Shareaza. He is also in Australia, demonstrating that the country is a center of P2P innovation today.
These talented groups have their sights firmly set on the file-sharing phenomenon as it is today -- but we are past that. We are focused on the future of file-sharing and the extension of P2P into next-generation consumer Internet tools. I am confident that we will continue to out-innovate these great companies in 2004.
TNW: While the latest Kazaa client declares that it is free of spyware, the client has been criticized for including spyware in the past. And now that the current client includes software that would be better described as adware, has Sharman Networks been listening to users about the spyware issue? And do you anticipate a time when the client will be adware-free?
Morle: We listen to users in a big way on this issue. Spyware is like the bogeyman of the Internet. The phrase gets thrown out there to strike fear into the hearts of users, but it is just a phantom -- a character in scary stories. It exists, but it is rare and certainly not distributed with KMD. Nothing goes into KMD without going through a very detailed acceptance testing procedure that ensures that the integrity of a user's privacy remains intact. No way will we allow anything that spies on users into our installer.
Ads are necessary for us to run our business -- as necessary as they are in modern television, radio and print. Viewing the ads is the payment for the free software users enjoy. We do install some advertising software, or adware, with Kazaa, but we are careful to keep this to a minimum. We install GAIN (which displays contextual advertising) and PerfectNav (which suggests Web sites to look at if a Web site cannot be found) only. eDonkey, as an example of another mainstream P2P application, installs Webhancer, TopText, New.Net and nCase.
Altnet is sometimes incorrectly cited as adware. This software is the P2P software that powers our premium content delivery. Altnet TopSearch displays premium content to users in search results in the form of gold icons, and Altnet PeerPoints gives users redeemable points (like frequent flyer points) for sharing premium files. Over time we are migrating from adware to other revenue streams as those revenue streams mature. We have so few adware products because revenue generated through content distribution is growing very well.
TNW: What is your general feeling about the RIAA?
Morle: Mostly disappointment. I have the benefit of many years thinking about P2P and its potential, and they have not got past the stage of fear. I am naturally disappointed that we are all wasting time and money on futile legal battles when everyone could be growing their businesses and benefiting users today. P2P will endure. We at Sharman have great confidence in this, and in the future we will look back and wonder what all the fuss was about, as we do today with videocassettes.
TNW: How about personal privacy on P2P networks? Is Sharman Networks planning any privacy-protection measures like those found in some Kazaa Lite modules?
Morle: Where I differ from some other file-sharing CTOs is that I don't believe users have anything to hide. Creating default "darknets" through technology defines the purpose of the network as illegal and encourages a certain kind of activity. This is not what we are here for because P2P is an enabling technology for all and not somewhere to hide. The modules found in Kazaa Lite that claim to bring privacy to the user can cause problems with other Internet applications, decrease the performance of the software and are ultimately possible to work around.
Kazaa users are only identifiable by IP number in exactly the same way that Outlook users and Mozilla users are. The rights of an organization to subpoena ISPs for the identity of the user connected to that IP has just been rightfully overturned. This, of course, is a great victory for a user's right to privacy. The only privacy issue in Kazaa was the capacity to view the entire contents of a user's shared folder. This functionality is now disabled by default to prevent abuse and has been disabled for some time. We want to put the users in charge of their own privacy.
TNW: What do you see as the biggest threat to P2P technology?
Morle: I don't think there is any great threat to P2P technology. There is so much value in the technology and so many groups and individuals making great P2P software, I am not sure what force could stop it. The greatest threat would be that the application of P2P technology does not meet its potential because it is resisted by some.
TNW: What do you think the best use of P2P technology is?
Morle: Today I would say it is highly optimized for the delivery of media files at a price that's right for end users. We will see more and more games, movies and music collections distributed across the P2P nets. Recently, we have become very good at distributing movies. In the next 10 years, we will see P2P becoming more pervasive, used for many different P2P applications, even built into phones and TVs.
The great thing about P2P is its simplicity. This is what makes it so democratic. Anybody can just drag a file into a special folder and make it available to the world. Written a book? Just drag and drop. Launching your band? Just copy to your shared folder.
TNW: The age of grid computing is upon us. P2P applications like Kazaa have demonstrated the power of distributed computing. What do you see as the role of Kazaa in the future of distributed computing?
Morle: The file-sharing era has given us the opportunity to get the resource-sharing relationship right with millions of users. It is actually quite difficult for users to understand that their resources could be used when they participate in a grid.
This is something that I don't think many of the more theoretical grid-computing projects have put enough effort into considering. It needs to be developed with respect for users with the tools in place for them to be in complete control. They must also feel that they are getting back from the grid at least as much as they are putting in -- meaning costs are reduced or extra value can be had.
P2P is more than file sharing, and we are looking at many applications for it. We have the practical experience of making software that millions of users connect to at the same time. I expect we will continue to take a leadership position as the market blossoms.