It is impractical to think that CIOs are going to deploy that kind of broad infrastructure on a many-times-a-year basis. It just doesn't happen. Secondly, people have begun to understand the absolute criticality of the infrastructure. That kind of quality and reliability risk simply isn't worth the upside. I think that phenomenon is being eclipsed by more mature understanding as the market evolves.

"That was fine for reasonably simplistic technology like browsers, that at the end of the day were not mission critical."

SCHINDLER: I have seen the daily revision culture embraced. It has been called a tight feedback loop, which gives the user exactly the software he or she wants, and fast bug fixes.
PAPOWS: I think there was a lot of excitement generated around this Wild West atmosphere of the Internet, and this constant progression of revolving releases. What you're describing was very consistent with some of the emotive responses to it. I don't think there's a lot of value there. I think more and more people understand the downside there. I don't think it is a concern limited to enterprises. I think small and medium-size businesses understand it as well.

Netscape has tried to reinvigorate this model around this notion of free client software and Internet-dispersed source code. That is a recipe for a lot of risk in the development cycle that isn't going to benefit the users.

That doesn't mean you don't need an institutionalized way to get user feedback. There are a lot more ways to get that than having everybody with a keyboard out there making changes to the source code that's going to support your mission-critical systems, and throwing it back over a firewall.

I think it is an overly simplistic view of it. I don't personally subscribe a lot of value to it.

C O N T I N U E D . . . 2 of 2
SCHINDLER: The Open Source movement is gathering momentum. They say there will always be more of them than us, the theory being that open source provides tens of thousands of developers, a staff level you can't match internally. Is this a fallacy?

"There's a lot of hype around the notion of Free Source. There's not a whole lot of incremental value that's obvious to me when we're dealing with mission-critical systems."

PAPOWS: I think there is a bit of a fallacy. I don't want to pooh-pooh the idea totally. To me this is nothing different than the kind of broad beta cycles that mature commercial products like Lotus Notes have gone through for years.

In a few weeks we're about to post the first beta of version 5 of Lotus Notes. That will go to thousands of beta sites. There is obviously tremendous value to user feedback, to gathering iterative responses, both to UI [user interface] conventions and code quality from that kind of release process.

Candidly, that's nothing new. Lotus didn't invent that. It has been part of the taxonomy of software development for a decade.

There's a lot of hype around the notion of Free Source. There's not a whole lot of incremental value that's obvious to me when we're dealing with mission-critical systems. I think it is fine in the categories of software where you can afford to be exposed in those dimensions. I think having the user feedback and participation is great. It has to be institutionalized in ways that are manageable.

SCHINDLER: Mitch Kapor would like to see the Internet appeal more to higher-intelligence functions such as group decision making. Do you see your Web-based products headed for the higher functions?
PAPOWS: The Internet is one of the few pieces of computing infrastructure that has been concurrently over-hyped and under-valued. That would seem like an oxymoron. It is accurate in this instance.

For all the hyperbole surrounding the Internet, until the last 18 months, most of what it was used for was static publishing, what I call 'brochureware.' We put up a series of Web pages on our intranet and we consume it, or we create an electronic presence in the form of a website.

We are getting into a category of what I refer to as market-facing systems, or interactive, more dynamic, knowledge-based applications which deal with things to Mitch's point, like collaboration. Codifying tacit information in actionable ways in our employee populace.

That's a much more exciting, a much higher-return equation than the static publishing models. Clearly it is all trending in that direction.