IBM's Collaboration Chief Talks Domino, Workplace Game Plan

Mike Rhodin, IBM Software's general manager for Workplace, Portal and Collaboration software, sat down this week with CRN Industry Editor Barbara Darrow to talk about Project Hannover, the next release of Lotus Notes/Domino, Workplace confusion, and Lotusphere game plans. The annual Lotusphere confab, in Orlando in late January, will serve as the public launch pad for Workplace 2.6, and the first public demonstrations of Project Hannover.

CRN : First, can you give an update on Project Hannover, the next release of Notes/Domino?

Rhodin: Sure. Some of this is a tit-for-tat game with Microsoft. Every time we ship a release, they send out announcement that it's the last release. Four straight times they've sent that out since I've been here and this time we pre-empted them. We talked about Hannover right before we shipped 7 so they couldn't say that. The [Notes/Domino] 7 release was primarily focused on [total cost of ownership] TCO, security, scalability, administration capabilities, what our IT shops are telling us needed to be done.

With Hannover, we looked square in the eye of the audience and said this release is about end users, all about end users. We're coming out with stuff IT shops but we wanted to get that out in front of people. the focus has squarely shifted to end user.

CRN : You're talking about both the rich client and the thinner client?

Rhodin: Absolutely. At the time we said the first public unveiling of running code would be at Lotusphere. We showed slideware. Our intention is to take the covers off of it. Our intent was to kind of get into beta in mid year and that's still the plan. We have been playing with running code for about four months now and had an analyst event a couple weeks ago, and we showed a sneak peak of what we're going to show at Lotusphere.

The right signals are there. We've been running focus groups with customers as design partners to help guide us and feedback's been pretty good.

CRN : Do you bring in any integrators or VARs into this process?

Rhodin: I wasn't in the meeting so don't know who was there, but we typically do. Hit all the constituencies at different points.

The key thing, particularly with ISVs and partners, is we've been very, very clear that we're not breaking any of the APIs and all the apps they've invested in over the years will continue to run in Hannover in contrast with what Microsoft's doing in E12. Where they announced all the APIs are changing and oh by the way you have to change the hardware too to 64-bit.

CRN: You're not narrowing support to 64-bit servers?

Rhodin: No. We've been doing 64-bit for awhile but we continue with 32-bit support.

CRN : Their argument is that you can't really even find a 32-bit server anymore.

Rhodin: A new one. If you're trying to drive a hardware upgrade that's the statement you'd make and the reality is there are a lot of 32-bit servers out there running both Microsoft and Lotus software and I'm not going to be standing here saying you have to buy new hardware just because you want to go to my next release of software. That's a choice you make on your operations, your administration. We'll continue to deliver choice and flexibility on that.

CRN : Even though IBM has a hardware group and servers to sell and Microsoft doesn't. Does that make Microsoft a better partner to IBM hardware than IBM Software?

Rhodin: [Laughs.] I wouldn't go that far. Microsoft doesn't put any software on most of our servers. We have a lot of non-Intel servers. That's a huge business and we run on all of them. Microsoft doesn't. Even on our Intel servers right now we're pushing high performance Linux and last time I checked Exchange and Office don't really run on Linux.

CRN: Does Lotus break out share by operating system for Domino etc?

Rhodin: I'll give you the growth numbers. We've seen 200 percent growth year over year in domino and Linux and we're seeing it in multiple platforms. People trying it on Linux for Z[series mainframes], we see it on iSeries, and we see it in Intel Linux. Probably the most on Intel Linux. From the Notes and Domino view point, the 7 release provided astronomical gains on scalability and performance on Linux. Linux had been lagging performance wise and it was simply because the threading packages weren't open in the kernel for Linux at the time they came out with the latest kernel releases.

Now we've got Linux on par with the other platforms.

CRN: Is Windows still the largest share for you?

Rhodin: Windows is the largest installed base. I don't know specifics but I'd be hard pressed to say that Windows is growing as fast [as Linux] because it's such a large percentage of the base.

We are seeing phenomenal growth in Linux and that's encouraging. That means customers are seriously deploying Linux now as a server operating systems. And we're seeing that in portals as well. Because we sell our products on a disk and you install what you want, we have to wait for follow up research and support calls to track OS.

CRN: Can you talk of penetration of Workplace vs. Domino?

Rhodin: Well Domino has an astronomically huge installed base, over 60,000 customer sites, some 122 million seats of notesthat's over 20 years. Not something you do overnight.

CRN: Any idea how many are active?

Rhodin: Not off the top of my head. Depends on what you have for maintenance. We are seeing very rapid uptake release to release and version to version [of Domino]. Of customers we're in direct contact with 90 percent are on 6 or later, most have moved to 6.5, and we're seeing 7 move faster than we've ever seen. [Notes/Domino 7 shipped in September.]

Historically people would wait for a point-one release for production, but not with 7.

Inside IBM before we shipped had 55,000 mailboxes on 7 and pushing on full deployment on 7 out here in the next several months. Fastest rollout ever.

CRN: What OS are you running internally? Everything?

Rhodin: Yes. We ran a 15-month internal beta on 7 on all platforms before we shipped.

CRN: How about Workplace penetration specifically?

Rhodin: Workplace is an emerging product. Our business split into three segments, established, large Notes/Domino business, we have the portal business, the WebSphere portal business, which I classify high double digit growth. And the Workplace business is growing very rapidly but off a small base. Workplace is not a product, a set of products.

CRN: Which is part of the problem...

Rhodin: [Laughs] I understand my challenge on branding and communication. It's something we have to clarify. Our Web content management product is growing extremely well, Web Services Express, the SMB version that your readers re familiar with, is doing very well, and it's only been out for a year, running on a promotion giveaway for most of that. [It's] selling quite nicely now.

What I'm most pleased about is number of ISVs are starting to build very interesting applications, starting to exploit the new template-based programming model. That's a harbinger of things to come.

CRN: Are most of those net new to you, or are they Domino people moving over?

Rhodin: It's both. It's kind of a hybrid because WSE stuff is built atop the portal architecture; it's a hybrid between WebSphere portal world and Notes/Domino world. All of things are coming together. Not separate businesses, converging businesses. And so the type of applications on WSE are types of applications that would have traditionally been built on Notes/Domino but they're not necessarily Notes/Domino ISVs building them. People are putting collaborative software idea, and line-of-business apps atop a J2EE infrastructure which I think is a good sign.

CRN: There's always been a sort of syndrome, on the Microsoft side, people used to call 32-bit apps ISVs the beaten wives club, because Microsoft would come in and take their business. On the Domino side, ISVs were hurt by a bait and switch when [IBM Software General Manager] Steve Mills dropped a bomb by talking about replacing NFS with DB2 and etc. A lot of them still feel burned by that. What do you tell ISVs?

Rhodin: Dropped the bomb on what?

CRN: He talked about moving from NSF to DB2 then took it back. There was a lot of confusion.

Rhodin: I think statements were made and people read a lot of things into those statements. When we said we were going to put DB2 under Domino, we did, in 7. But we did it in a way that didn't break any apps. We mapped NSF on top of [DB2]... which people should look at and say 'Okay when IBM says they're going to start to introduce J2EE into collaboration, does that mean its rip and replace or a smooth evolution?' You'll look historically to see we drive evolution.

The architectural challenge that Microsoft put in front of their existing customers is a much bigger leap.

CRN: Well, they hate them too.

Rhodin: Since the day I got to Lotus, when I headed up the engineering teams, I made a promise to customers that I was not going to create a cliff that they had to jump off to get to the next thing. I was going to provide a smooth path forward and guarantee applications moving forward and I believe I've delivered on that promise.

CRN: So if you were talking to a traditional Notes/Domino ISV, say Percussion, what do you tell them to do going forward? Stick with Domino?

Rhodin: What you'll see unveiled at Lotusphere and in the coming year is how these things will start to converge. We've been really working hard with our customers to understand what seamless evolution means to you. And what we get back is it's choice and flexibility about when I do what. No forced dates or forced migrations. If your skills are in Domino applications, we'll carry those skills forward and those assets forward. Scripting will continue to work. New releases of Domino coming out.

Microsoft's claims that Domino ending are about as stupid as the claims that CICS was going to go away. When we announced WebSphere, there were stupid articles saying that CICS would go away. Well CICS is growing. The truth is as things come out it doesn't necessarily mean it's at the expense of old things. It's expand and grow. Things around Notes and Domino will expand. The vision of what Domino apps means as part of this whole SOA [service oriented architecture] approach.

If you look at specific things we did in Notes/Domino 7 about adding Web services into Domino so Domino apps can be published as Web services and consumed in an SOA architecture, not only by IBM's SOA but anyone's SOA. That's the approach we're taking.

If people want to keep their skills on Domino Designer building new apps, we're actually seeing more people building this year than last the previous year people are more comfortable with the longevity of the platform and no one's come up with a better solution for building apps faster than Domino Designer.

In order to bridge the skills gap for those moving toward J2EE technologies, we came out with Workplace Designer, which brought the skills of the Domino developer to a new tool that would be familiar in a couple of hours to build apps except what gets generated out of the bottom is J2EE components that fit into it without ever having to write any Java code.

We're bringing the applications forward, continuing to support new apps built on the platform, helping people who want to transition to J2EE do it in a seamless way. And for people who want to build hard-core Java applications, we've got the Rational toolset. We've tried to build this out in a way that hits the widest possible audience, especially in the ISV world.

We usually show off Rational stuff, at Lotusphere and the Workplace Designer stuff works as an adjunct to Rational tools if you want to work that way because both are built on Eclipse Technologies.

By building on common architectural fundamentals we've made it easier to build these things mix and match depending on what you want to do.

What you'll start to see is more and more composite development model unfold both from rational, WebSphere and Lotus viewpoints.

The visual composition capabilities we've added into Workplace are really starting to catch on. When I talk about WSE apps being built, that's what I mean. People are starting to figure out this template based stuff is pretty cool. I can build components, I can assemble components that get reused in different ways. I can use the points of variability of the templates arch to really customize these things on the fly for different roles in the organization. People are really starting to get it.

CRN: How does the Ajax methodology fit in here?

Rhodin: It fits right in. Ajax is a style of development but building Ajax components is non trivial. The really good Ajax applications in the world are handcrafted by experts. The thing that's been missing on the Ajax front is simple tooling. You're going to start seeing us start to extend Workplace designer with some Ajax capabilities. We already have some Ajax-style componentry. If you look at UI for WSE, the palettes that slide in and out, the drag and drop, that's all Ajax-style things and have been there for awhile.

We've built the arch so you can incrementally add. In the WSE case we added it into the themes. The portal arch has this concept of themes and they're enabled by Ajax. If you go back historically, one really, really large Ajax, style app was iNotes, not with the current definition of the emerging architecture, but with that style initially.not the current definition but the style of asynchronous JavaScript etc. That's becoming much more prevalent.

CRN: Will Ajax be the next boondoggle?

Rhodin: I think it's real. From our point of view, it's way to separate the data delivery and the UI delivery into a browser paradigm...to deliver a richer user experience with lower performance overhead. So we think there's big benefits to it but it's not a panacea. Only certain classes of applications are appropriate for Ajax.

CRN: Like?

Rhodin: E-maps, Google maps or document management capability, anywhere you want to do drag-and-drop desktop like operations, those operations are applicable for Ajax-style UIs. But we have to be careful as an industry not to go overboard, not to run from one side of the ship to the other, because it'll tip over. The idea, is we're gradually introducing Ajax-style components and widgets into our programming model so that you can do it in a sane manner.

CRN: So it's not like you'll need a point upgrade to Domino or Workplace Designer or whatever to start Ajax-style work?

Rhodin: Absolutely. Basically in Designer you can pull widgets and buttons off the side and some are already Ajax style buttons so you can start to add more and more of that over time. And we'll make it so the end user doesn't have to go 'oh my god I have to learn Ajax,' but just use the things they have on the palette to build a component.

CRN: There is a huge push among developers to learn Ajax, they think it'll make them more valuable, charge more money.

Rhodin: Yes. It's a handcrafted skill but I think it has a lot of promise and is real. And we'll add more and more of it but we've been doing it awhile in portlets, some Alphablox components have been Ajax style.

We're just trying to do it in away where you don't have to start over and rewrite everything. Google maps has a component that's Ajax, but not everything there is Ajax.

Think about it as a way to do intelligent UI for components. [You've got to ] pick your battles, the ones that will deliver performance and user experience benefit you're looking for. For example, if it's a content thing, a news blast, why? It's not worth the programming headache. But for mail, where you have a container and are shooting data in on regular basis, or an RSS reader, those kind of things you want the UI to go down once and Ajax keeps flowing.

CRN : A Microsoft source has said, looking five years out, mail will not be a differentiator anymore for any customer. More and more will outsource e-mail to someone and concentrate their resources on IT areas where they can differentiate. This is [someone] from the Exchange side of the house saying that not even Exchange or Lotus will be the ideal platform for such mail. Comments?

Rhodin: I view e-mail as a mission-critical commodity. It's both. Different companies based on regulatory posture have to do things different. Some organizations under heavy compliance issues ,will want to maintain control of the data. If you use gMail as corporate email what happens if it goes out of business, do you have access to the data/ Maybe, maybe not. But for some companies, where email is just a casual communications tool, outsourcing may be appropriate.

His point is right, e-mail is not the center of the universe five years from now. In fact, I don't think e-mail is the center of the universe today. It's shifting largely. Inside IBM the number 1 communications venue is instant messaging. The speed of business is at the point where e-mail is too slow.

CRN: And too full of crap.

Rhodin: The other piece that's fundamentally flawed is that e-mail is overloaded. Your point on crap, I will say "overloaded." It has more connotations than just spam. People use email as their to-do list. Very inappropriate because it's LIFO [last in, first out], as a LIFO stack, if you get a lot of email, the most important thing for you to see maybe on page four. Really where our research is and we started to tee-off on last year, is activity centric computing. We previewed Activity Explorer and will GA that as part of the 2.6 release shortly. That's the first step in getting people to rethink how they work. I don't think Activity Explorer is the end-all be-all, but the first part of the experiment in trying to shift the model so you're organizing your work, and all artifacts in an activity center. Organize artifacts around the project instead of organizing the project around the artifacts.

[Lotus first showed off Activity Explorer last year at a Future of Work event in Cambridge, then announced it as part of Workplace at Lotusphere and are now previewing it.]

Customers are trying it out, everyone gets that it's a paradigm shift. It can't replace email, it has to augment email initially. You can't force a shift but activities will be increasingly important.

In the early Hannover design, activities will start to be more and more prominent. Hannover is important signal to the market our customer base, that these workplace things are starting to come together. Hannover is a unifying point between the two architectures. All of the things people have known and loved about Notes area still there but there's additional stuff for the Eclipse community and new capabilities like activity-centric computing on Notes, not just the Workplace.

You start to see benefits of componentization architecture we've been on at the Software group. Let us mix and match capabilities into different offerings without having to go through major engineering projects.

The current Workplace is 2.5.1. [Workplace] 2.6 is imminent.

CRN: There are a bunch of startups in the collaboration space, Zimbra. Is this category too jam packed to support new entrants?

Rhodin: No. I think new entrants are bringing interesting innovations and trying new things. As an open vendor I embrace new vendors, it fosters competition. I plan on competing vigorously.

CRN: Are you going to buy them?

Rhodin: We are always doing acquisitions and not doing acquisitions. I think my PureEdge acquisition this summer is extremely strategic for us.

CRN: They were a big partner.

Rhodin: They were a big partner but were in a space that was a logical extension to what we dohigh"end electronic forms, pixel perfect. It was such a synergistic fit to what we're doing. The simplest way to put it, we've talked about electronic forms automation fro 25 years, it's finally starting to happen because we're finally agreeing on standards.

CRN : StreamServe is another.

Rhodin: Yes. StreamServe and Adobe which bought JetForms. And a number of players but the amount of standards and I think the market is heading towards xForms", that's the horse we've decided to ride.

Early deals PureEdge has done show tremendous return on investment for customers. [There is] a lot of interest around the world. We'll have internationalized versions of PureEdge in the next few months. A great partner play.

CRN : There's so much consolidation going on, it seems ISVs get into markets for the sole purpose of being bought. Is that healthy?

Rhodin: That is a common exit strategy for a lot of small companies. For this space, however, think of forms. We tend to invest in things that are horizontal in nature vs. vertical.

Think about forms, someone's going to build the healthcare forms application, the government application. Those are very domain-specific capabilities and huge opportunities for partners both on the SI and ISV side. Building the law office automation software based on a standard form processor seems like a logical thing to do as opposed to a proprietary forms processor that can't interoperate with any other company. Because law forms aggregate and consolidate too. They have to share information with governments and with businesses and clients.

The secret about xForms isn't about internal homogenization of forms architecture but external homogenization of the forms standard.. You have to exchange forms beyond corporate boundaries.

CRN : [Oracle CEO] Larry Ellison has been saying for years that there will be just a handful of big software companies left standing.

Rhodin: Well, he's personally trying to make that happen.

CRN: Having said that, it seems if there's creativity out there, it just gets sucked up by IBM, Oracle or Microsoft. Is that a good thing in a broader sense?

Rhodin: The industry always goes in waves. Historically there are consolidation/innovation cycles, usually driven by business cycles and the market.

We've been consolidation for three or four years and where big players tend to consolidate to move to next level. But every time we consolidate and move to next level, there's another round of innovation built atop that level.

We may be starting a new innovation phase. Particularly in what we're doing with Workplace, tremendous innovation can be done atop that programming model. The biggest shame around Workplace as it got misconstrued in the early days as the replacement for Notes and Domino vs. a new platform at a higher level that starts to extend the open standards world with new capabilities for people to build new classes of applications that up till now have been hard to build.

In order to help people get it I actually had Larry Bowden's team go build a couple. The BCR [Business Controls and Reporting] solution and BSE [Business Strategy Execution] Solution are examples of the class of application that are easy to build.

You have to have a lighthouse to lead people to the harbor. We see them as horizontal solutions vs. vertical solutions. As software and collaborative systems become more and more about business, having control and audit frameworks seems like a good idea to be part of the infrastructure.. Same for being able to manage an organization by objective. Good candidates for us but also examples of what you can build with this architecture.

CRN : Do you offer an online store for components?

Rhodin: We've had a portlet catalog for portals since day one. Long before SharePoint. Workplace stuff is being posted to that. And we put emerging stuff on Alphaworks.

Thinks like Atom were originally developed and delivered from there, tools for social networkingthere's a lot of interesting stuff.

Portlet catalog is gathering point for components we've built and pointers to partner components

CRN : Given the continued confusion around Workplace/Domino, will you pull back on the Workplace messaging at all?

Rhodin: We're going to continue to try to clarify it. The key thing is we believe the whole composite app model around workplace is fundamental to how SOAs are going to be built in the future. What we've focused on all along is positioning he portal and workplace stuff as the interaction surfaces for SOA. That's always been the design point.

Various analysts have written some pretty nutty stuff. I can't believe we were in the same meetings. They keep trying to spin it back into "This is just a new definition of e-mail, new definition of instant messaging."

My point is, no. Those things become services. They're commodities. No matter how you look at them, it's what you do with them that becomes interesting. And making those components available as part of this composite application model versus a separate e-mail system or separate IM system is what makes it interesting.

Just as when Notes came out 15 years ago, no one knew what groupware or collaboration was. It was the first set of applications that were built that started to show people the way. We're heading into that phase.

I'm comfortable with the progress we've made with Workplace. From my original point of inception till now it's been about two and a half years, I'm within six months of where I thought I'd be. I'd like to be where I am six months ago, but other than that I'm pretty happy. I think it's on track. You'll see us position the open standards-based, components-based composition model stuff as the front end to the SOA architecture IBM's bringing to the market. That will start to clarify things for people because it's where interaction and collaboration services meet business process through the SOA framework. When we started talking about Workplace being collaboration in the context of business process, people thought we'd start to implement business process things in workplace, and weren't' thinking about it as the front end to the business process stuff we're doing in another part of Software Group.

If you look at the SOA arch we laid out this fall, it clearly positions the role that Lotus plays, Tivoli plays, Rational plays , WebSphere plays and Information Management plays. They all play a role in this blueprint we started to build 2-3 years ago. Steve Mills used to give speeches on componentization, this is what he meant. Software becoming composable surfaces based on a component architecture. It's giving us speed and time to market we haven't seen before.

I'm okay with the naysayers going on about the current state of workplace. We have a very clear vision of where we want to go and think we're on track. And in 24 months from now when we're having the same discussion before Lotusphere, you'll be looking back saying 'yeah it worked out,' or you'll be talking to someone else. This thing is so real for me and so clear for me and I'm not clairvoyant so it must be happening.

CRN : In this move to software as a servicethe latest buzz, For VARs and Sis, there's worry about disintermediation, that big companies like IBM, Microsoft will be pumping all these services out directly to partners, I think a mistaken perception, but can you talk about that?

Rhodin: Software as a Service is an interesting topic, because it's more than one topic.

CRN : It's more than ASPs

Rhodin: Right. We've been there ASPs didn't quite work. You have to break software as a service into two fundamental models from a consumability point of view. Advertising-subsidized SAAS and more traditional, next-gen ASPs, delivering Web-based services, like Salesforce.com or Webex. The interesting thing is the dynamics between advertising subsidization and license subsidization and how those business models are going to evolve. That's the interesting debate.

Does software simply become a tool to drive advertising revenue vs. software having a value of its own. Then you have to get into that whole issue we talked aboutwhere's the data? Where does the data live? How do I get to it? Is it controlled? Is it auditable? Is it regulated? When you get into Software as a service those become real questions.

Early adopters who moved down hat model are ok with it. Webex is a transient model, if you lose it, it's a couple of days of data worth of data. Salesforce.com--sales still tends to be very tactical kind of thing. Recovery cycle would be different. But what if all your corporate documents were in some service and your company dealt with a service provider that went out of business? Where'd your data go? How do you handle that?

CRN : In ADP's case, Companies retain their data but ADP disperses the payroll I guess.

Rhodin: I don't know but it's probably a simply a production capability but if you're running your e-mail outside, where's your data?

The third model of Software as a service is outsourcing. We have seen a major shift towards people outsourcing a lot of infrastructure but still control the IP and the data. I think that space will require a lot more fleshing out in the market. My guess is there'll be some public failures along the way. There always are. Those are things we learn from and improve on.

CRN : What services do you offer now?

Rhodin: BCR is offered as a service now hosted by [IBM Global Services], you can buy a certain number of users.

CRN : Is there a broader IGS hosted email service offering?

Rhodin: We don't do a broad e-mail offering . we do outsource and manage email systems. Our strategic outsourcing group does a lot of Domino hosting and Exchange hosting for that matter. They're pretty much a Switzerland play. On their servers but all data is separately.

On Google model though it's on a user basis, not a company basis and segregation of data is an issue.

CRN : In terms of customization and verticalization even in Software as a service there would be a role for Sis and VARs?

Rhodin: Yes. But I don't think on-premises software is dead. That's not my current view of the market. I think we'll go through another experimentation phases. Remember when ASPs came out, it was the second coming.and then they crashed with the bubble no one made any money and people had to deal with business models. A few people learned from it and came out, Webex and Salesforce.com are examples.

Web conferencing is a commodity, like audio conferencing. Salesforce.com is something of a new model for outsourcing business process. As you say the back office stuff has been there for a long time like ADP but things that are more end user facing like salesforce.com is a new model.

An ecosystem has to exist, that's what drives sustainability. Anyone who tries to go it alone runs into trouble. That's why Microsoft trying to compete with ISVs is an interesting play because those are the partners who actually took them to market for years. When you compete with your channel you create a problem for yourself down the road. It'll see how they cross that bridge.

CRN : Any more things to look for at Lotusphere?

Rhodin: This will be one of the more announcement-rich Lotuspheres in history. Maybe the most. We have a lot keyed up.

CRN : I assume Ray Ozzie [now CTO of Microsoft] won't be there again?

Rhodin: [Laughs] Probably not. He's pretty busy in his new role. It's interesting, Ray's divorced himself from the Groove stuff.