CNET también está disponible en español.

Ir a español

Don't show this again

Tech Industry

Taking P2P to corporate America

Star developer Ray Ozzie is focusing on creating software based on peer-to-peer technology, an approach that some analysts believe can transform the way Web business gets done.

Ray Ozzie is in the all-time list of star software developers.

The inventor of Notes helped popularize the concept of groupware and was considered so important that IBM CEO Lou Gerstner paid a special visit to ensure Ozzie would remain after Big Blue launched a hostile takeover of Lotus Development in the summer of 1995.

Ozzie is now well into his second career--this time as the founder and CEO of start-up Groove Networks. In his new venture, Ozzie has focused his attention on creating software products based on peer-to-peer technology, a computing approach that some analysts believe can transform the way Internet business gets done.

But in the popular imagination, P2P has become inextricably connected with Napster--free publicity, to be sure, but bad news when you're trying to sell corporate America on the merits of a technology that challenges the concept of centralized control.

Ozzie recently sat down for an interview in San Francisco, where the star developer was speaking at a P2P conference sponsored by O'Reilly & Associates.

There's a lot of excitement about peer-to-peer computing's potential for transforming the way business gets done over the Internet. In fact, there's a lot of hype about the transforming aspects of P2P. But is there anything that you see on the horizon that might trip up P2P? Do you worry about a nightmare scenario?
There are a few like us who are doing collaboration stuff. And some are doing distributed computing, and they're serious about their niche. But then there's a whole bunch of people doing Napster follow-ons. If the Napster follow-on people pollute the reputation of what is--it's just a technology, just a stupid technology--but if the usage of it becomes so highly political that it polarizes people's attitudes about it, this just gives IT people more reason to be wary.

How do you get over that and bring P2P into the enterprise?
From a business perspective, you have to empathize with what IT has to do. If any software has a bad reputation for one reason or another, IT managers will stay away from it because if it comes back, they'll be left holding the bag in the end. I don't actually think this has happened. The whole thing that's gone on in the Napster trial is making all the people in that space so wary about talking about it that there's actually a resurgence on the other side of, "Oh, what are other ways that we can talk about this technology?"

Do IT executives balk when they hear about a technology allowing people inside the company to communicate with people outside the corporate firewall?
You have to come back and say, "I get it. I understand your fear. So, let's talk about your e-mail system and talk about how you manage your e-mail system and how you manage the flow of information in and out of the organization." The only industry that has that under control is the financial services industry. They really do a good job at watching the flow of what goes in and out. But most companies don't.

There are a lot of interesting ideas out there. But if 
somebody's got dreams of making money by bringing it to the enterprise, it's better beforehand to know what you have to do to get in.In customers' minds, has Napster become synonymous with P2P?
You can get past the Napster thing in about five minutes. If you leave it lurking, it might keep coming up. If you start right off and say, "Here's Napster, and here's what it does. Now that we know that, let me tell you what we're all about and how we're differentiated," and then work from there, it's not an issue. The big thing for the IT people is that the technology doesn't harm their networks and that it allows them to be able to manage it, consistent with how they manage everything else.

Is that a regular refrain that gets raised when you try to introduce the idea of a P2P-based solution?
It's obviously always an issue, but we have yet to find too many instances where it's such a showstopper because of religious principles.

What about the issue of centralization vs. decentralization that goes along with any discussion about P2P with IT managers? How do you overcome that obstacle?
By making sure that we don't have a problem with the centralized vs. decentralized issue. We have to show IT very clearly that it is the intent that applications developed in the (P2P) platform connect with apps developed at the center. So, if you have an SAP installation, an i2 (Technologies) installation, or something like that, so that the apps aren't off on their own, that they're integrated--you leverage that investment as opposed to duplicating it.

In that scenario, what's the advantage of P2P?
We're way better at controlling the flow of intellectual property securely between people who need to work together--way more than e-mail. Look, I think there's a major disaster waiting to happen--an insider trading scandal with some ISP administrator who's been watching the e-mail flowing by. The corporation's administrative management has good intentions, but people at individual keyboards have amazing access to the information that's flowing by.

This is crying out for a virtual private network that matches the actual group that's collaborating and has the spontaneity of e-mail. We won't succeed unless we end up with the usability or spontaneity of getting that secure space set up. Once IT understands the technology and sees that it improves the situation they've already got, I think they're going to be a lot more open to (P2P).

But don't let me get away with making this sound like a piece of cake. It's not. You need to have the answers to valid questions, and each IT person has them.

There are lots of young people who I'm on panels with who are in 
their 20s and working on some cool technology that they believe in. There's 
just this whole buzz around the whole thing.Groove's public unveiling came last October at Internet World, but you're still considered a start-up. Are the majority of your sales still coming from direct sales, or has word-of-mouth changed that breakdown?
At the very beginning, it was via Rolodex. It was, "Who do we know?" But since October, it started to take a turn. The people who we were working with since midsummer, the integrators and the VARs (value-added resellers), were starting to pitch their own clients on it. From that moment on, we were basically doing joint sales calls. At this point, roughly 30 percent of our sales are generated by us; about 20 percent of the people are coming to us; and the remaining half are coming from partners. And it's changing quarter by quarter.

Put this into perspective. How does the acceptance of Groove compare with the acceptance rate of Notes?
It's like night and day, absolutely night and day--even though there's a similarity in the pitch, and there is some overlap in the value proposition of what we might have been pitching in that era. But today it's a whole different world. In those days, we didn't know how to pitch the product. We didn't know the value proposition of Notes. Also, we weren't good communicators--even when we finally did learn how to do it. That was true all the way up to the time when I left Lotus; the partners just pitched Notes a whole lot better. If you took any random good VAR who was making $5 million to $10 million a year from Notes, they could pitch the product better than 90 percent of our own people.

That's surprising? Why do you think that was so?
I don't know. Some companies like Microsoft are incredibly good at getting the message out to their own field people. In this company, we started from day one realizing how much communications was going to be a big deal. That's why we brought on (former Lotus communications vice president) Rich Eckel and some other people who worked in that realm, because internal communications is every bit as important as external communications in terms of getting people to understand what you're doing.

OK, but with a new product or technology like Notes or what's offered by Groove, don't you again run into the same challenge of making that value proposition about something that hasn't yet become conventional wisdom in the market and easily understood?
When we were pitching Notes in those days, the concept of a technology could help facilitate lowering the cost of collaboration or lowering the time to decision--we were on the defensive to prove the return on investment. We were also trying to get people to install networks and to put graphical user interfaces on their computers, and it was really tough. The best thing that ever happened to Notes was (former PricewaterhouseCoopers Technology Chief) Sheldon Laube, who was the first big customer. He was a great communicator and able to espouse what he was using the product for far better than we could.

Nowadays, most good IT people and line-of-business people understand that a critical success factor for their companies is in strategic technologies. How can they use technology--whatever the palette of technology that's out there--on an ongoing basis? How can they use it to gain a competitive advantage and reduce their costs? They're in that mode of being open to hearing how it might affect them.

Are they skeptical? Rightly or wrongly, Napster and P2P have become almost synonymous in the public discussion.
Obviously the skepticism level is still high, as it is with any new technology. But in terms of understanding the principles of dynamic adaptive systems, understanding principles of cross-company collaboration and strategic outsourcing and how things are different nowadays, the need is there; the openness is there. Again, it's not a piece of cake, but there's a receptiveness that wasn't there before.

There is a lot of buzz about what's happening in P2P. Something along the early days after people discovered the Mosaic browser.
I find this space interesting. There is a real appetite out there for P2P. There are lots of young people who I'm on panels with who are in their 20s and working on some cool technology that they believe in. There's just this whole buzz around the whole thing. I feel like a wet blanket sometimes talking about this enterprise stuff, but that's because it's hard. And nobody wants to hear how hard it is to accomplish what they want to accomplish. A number of these companies will die. There are a lot of interesting ideas out there. But if somebody's got dreams of making money by bringing it to the enterprise, it's better beforehand to know what you have to do to get in.

Do you prefer the operational aspects of running your own company?
I think I do what I do because I enjoy complexity, and that complexity rears itself in a number of different dimensions--for instance, strategic planning for channels and marketing. I never thought I'd enjoy that, but I am enjoying it more than I thought I would because the problems to be solved are interesting, nonobvious. Managing a lot of people is not easy--ever. So from that standpoint, I could use help. I have a belief that you always hire people who are better than you are. So even though I dabble in this, and I dabble in that, I still have to keep reminding myself that my core competency is in product design and development.

Even Bill Gates brought in a Steve Ballmer to take over part of the load after he started Microsoft.
And Ballmer and Gates are a great pair. What a great complement to each other.

If P2P takes off, would that reduce or eliminate the need for the middleman that you have in business-to-business exchanges?
That's a really good question. It depends on the goods and services that's being exchanged and the purpose of the exchange. I would be willing to bet if the exchange is necessary for creating a market in something, I think it will be healthy. To the extent that it is an aggregator of commodity goods, I think it will be OK. To the extent that there's any negotiation or collaborative design and relationship that has to happen, I believe that's going to happen with P2P. So in those instances, I don't really know what the long-term prognosis is for business exchanges...In the future, I think the exchange points will still be there, but wherever a high-quality level of service is needed within enterprises, you'll definitely peer. So they might actually complement each other. You may have the default meeting place being the exchange.

Might that take a bite out of some companies' sales?
I wouldn't be really optimistic about the revenue flowing through the exchanges because once people start the relationships--I mean look at Dell--once they pick four or five partners in a particular area, they're going to throw a lot of business at those four or five partners. But they don't want 500, and so they'll just peer with those guys.