WebSphere Journal recently interviewed Doug Wilson, distinguished engineer and CTO of IBM's Lotus Software Division in Westford, Massachusetts. Part 2 of this interview addresses Eclipse and other topics, with concluding remarks by WebSphere Portal IT architect Richard Gornitsky about future WebSphere Portal enhancements.

WebSphere Journal: Okay, we were talking about vertical markets. One question we have is how is RFID driving things in retail? Are you seeing any thing of note in that area? And are there any other vertical market issues that you're facing that are worth noting?

Doug Wilson: RFID is, of course, a hot area right now. So far, we don't have RFID tags on people, so the interface between that and collaborative systems is still a bit early. But what is interesting about the relationship between the two is that RFID focuses attention in an area that we label sensors and actuators, i.e., the ability to have stations throughout a warehouse or a shipping organization or a point-of-sale that are sensitive to materials coming through them.

They can react to those materials, monitor things as well, and drive an infrastructure for pervasive computing: small, purpose-built devices and appliances that are "sometimes connected" or "often connected" or "fully connected."

We spend a lot of time and energy at the infrastructure level to support those things. That same technology is coming into the collaboration portfolio through Workplace Client Technologies and supports the mobility of users and workers. One of the thrusts of the Workplace initiatives is to make sure that employees can get access to and act on and react on information, whatever their mode of connection is, whether they're sitting in the office, at an airport kiosk, on their handheld or phone device.

WJ: More or less adjusting their needs when and where they need their information.

DW: Absolutely. And again, I stress that it's not just a one-way push of information, but it's the ability of remote employees to take action on that information that separates some of this.

WJ: A virtual office, so to speak.

DW: Yes.

WJ: Earlier you mentioned Lotus supporting new technologies. One of those, of course, is Eclipse. How does everything we've been talking about so far tie in to the underlying Eclipse framework and into the broader focus of moving some of the Lotus product line into the J2EE environment itself?

DW: Well, that's a great question. Building up from this notion about pervasive computing and "sometimes attached" and all the modalities that it introduces, comes the question: How do you manage those devices and how do you manage that platform? And by the way, is the desktop system that we think of today on the laptop really any different from that standpoint than any other device that might be connected to the network?

WJ: There is a merge going on between devices. Each device seems to become pretty much a complete solution....

DW: Or at least capable of operating complete solutions.

WJ: Right.

DW: The problem then becomes: How do you manage all of it? Our philosophy is to drive the management of the capabilities of those devices from an information model at a central location or server location based on the employees' role and identity, and, of course, their role with respect to the business processes that they have to engage in. That management model and the corresponding aggregation of capabilities gives rise to the need for a desktop system capable of being managed remotely, capable of providing high-quality user-interface interactions and also a programming model for applications to be built in that space. For us, that's Eclipse.

We've been working with the Eclipse.org group in a project called Equinox to drive into the Eclipse foundation the capabilities of remote management, the capabilities and extensions of the environment that will use it as a base application platform. That core technology, as well as being the core technology behind IBM's tooling portfolio, is also the core technology behind the Workplace client technology.

WJ: Are security capabilities included in various aspects of the Lotus software and tools?

DW: Absolutely. For example, the local file stores or the local document stores that we provide in Workplace client technology are always encrypted. That encryption is managed with public key certificates or interfaced with the security management system that a customer may have on-premises already.

So, an interesting difference between the engineering that we're doing now in the portfolio, and that perhaps was current at the origin of Notes and over the '90s, is that in those days there were no standards or widely available public capabilities for implementing secure systems. And so by necessity those capabilities were, to use the "P" word, proprietary. That is, while they were based on open and well understood algorithms, the key management infrastructures and the like were necessarily unique to each application.

That world has changed. And now there are many more open standards and systems available for us, and protocols and well-understood norms in the industry for managing the essentials of user identity and security, and the art form in the new portfolio is to take advantage of those without turning back the clock on what the security capabilities are.

WJ: How do you view the competitive landscape with, say, Microsoft Exchange or Novell Groupware, and what sort of market share do you envision in the short- and long-term for IBM and Lotus Notes and Domino products?

DW: Well, I think probably as far as I can go is to say an increasing share. In the space of mail and messaging the market is pretty mature. There isn't likely to be a lot of transition one way or another. Our Notes customers are well serviced in that space and we assure them that we are not going anywhere with the Notes Domino portfolio that they don't want us to go. It's a strong, stable investment, and we are committed to taking them forward without causing them to have to rip and replace the infrastructures that they have invested in over the years, and the programs and customizations and behavior they've invested in.

On Achieving markets share... I think it is important to bring forward a new idea. And our new idea is based in a new notion of collaboration that we call activity-centric computing. In a sort of short sentence, activity-centric computing is about managing the business activities and the thread of events in a business activity, rather than concentrating on managing the artifacts of that business activity.

Systems to date have typically concentrated on managing those artifacts, which is fine and important and we need to do that, but that doesn't bring enough value to the party. In traditional systems, the user is still responsible for the correlation of all of those artifacts with their business problem. They have to know that the contract is in this document and that the financials are in this document and that the chat we had yesterday and the three mail messages you sent me are all relevant to a particular business operation.