Wednesday, November 15, 2006

Transcript of Dana Gardner's BriefingsDirect SOA Insights Edition Vol. 4 Podcast on Oracle OpenWorld and Oracle's Linux and SOA Support

Edited transcript of weekly BriefingsDirect[TM] SOA Insights Edition, recorded Oct. 27, 2006.

Listen to the podcast here. If you'd like to learn more about BriefingsDirect B2B informational podcasts or to become a sponsor of this or other B2B podcasts, contact Dana Gardner at 603-528-2435.

Dana Gardner: Hello, and welcome to the latest Briefings Direct SOA Insights Edition, a weekly discussion and dissection of services oriented architecture (SOA)-related news and industry events with a panel of independent IT industry analysts. I’m your host and moderator, Dana Gardner, principal analyst at Interarbor Solutions.

Our panel this week consists of a return by Neil Macehiter. He is research director at Macehiter Ward-Dutton based in the U.K. Also returning is Steve Garone, a former program vice president at IDC, and a founder of The AlignIT Group, and now an independent industry analyst. And joining us for the first time, we’d like to welcome Joe McKendrick. He is an independent research consultant, and a contributing editor and columnist for Database Trends, as well as a SOA blogger for ZDNet and ebizQ. Welcome, Joe.

Joe McKendrick: Thanks, Dana.

Gardner: We’re going to talk this week -- which is the week of Oct. 23, 2006 -- primarily about Oracle OpenWorld. This large trade show, focused on Oracle’s ecology and products, was held this week in San Francisco, and some interesting news came out of this event, primarily around SOA, development, middleware, and even Linux. So let’s pop over first to Joe in metro Philadelphia. Joe, tell us what you saw as interesting or impactful from the Oracle OpenWorld Show, particularly vis-à-vis SOA?

McKendrick: Sure, Dana. Well, it was definitely a big show, and they really made a big splash. Just let me say, Oracle has been confusing the heck out of the market over the past year or so with its double Fusion initiatives. There’s the Fusion Middleware, of course, and then there’s the fusion of Oracle’s acquired companies, such as PeopleSoft and Siebel. Are the two fusions connected? Who knows?

I do research for two of the leading Oracle user groups, the Independent Oracle Users Group, IOUG and OAUG, The Oracle Application Users Group, and I find few end-user customers know what to make of Fusion, or how it’s going to impact them. I’m not quite sure if OracleWorld brought total clarity. I think it might have brought some clarity to the question. What we want to talk about here, of course, is Oracle’s Fusion middleware, which I equate to IBM’s WebSphere on some levels.

In a survey I conducted and put together for IOUG earlier in the year, only a small handful of Oracle database administrators and users could see how Oracle Fusion middleware could help them, or lower their deployment costs. So Oracle has a lot of user education to do, and it looks like this week’s OracleWorld was a good start, but there’s plenty of work ahead.

Gardner: It seem to me that Oracle at this event really put a stake in the ground and said, “We’re not just a database company, and we’re not just a database company plus business apps.” I think they were really saying, “We’re going to be a full-service software company from soup to nuts, and development, deployment, and lifecycle.” So, that really puts them up into the IBM arena. Would you agree with that?

McKendrick: I agree with that, and it's interesting, because looking at what’s been coming out of OpenWorld, you not only have IBM, but you have two of the largest vendors in the space -- Oracle and Microsoft -- having made recent announcements that are diametrically opposed in their proposed approaches to SOA. Microsoft had their own SOA confab a couple of weeks back, on the one side saying, “Go slow, build incrementally, and don’t spend a lot of money on your SOA as you roll it out.”

Then at the other end, you have Oracle, which is all about big SOA, saying “Take your huge, 10-terabyte database, take your huge ERP system, and move some of that functionality out to a service layer, preferably Oracle’s.” Plus, Oracle was talking a lot about mainframes, recognizing that IBM has been quite successful in reenergizing this big server platform, and Oracle wants a piece of that action as well -- more big SOA.

Gardner: I also did some field research recently that came up with a pretty solid number. It’s a general number, but of the SOA services that are in production and in use that we could get some insight on, fully two-thirds of them were of a services data layer nature. That is to say, they were all about making data available from a variety of sources, repositories, and applications to then be used more generally. Therefore, the whole role of data seems to be that the horse sits in front of the cart when it comes to using SOA, which of course would put Oracle in a pretty good position.

McKendrick: Absolutely. They’re also strong in applications as well, but it's interesting, Dana, as you point out, that there’s a convergence in place between the world of enterprise data management, and the world of application management or SOA. The two have been in separate silos, so to speak, with different disciplines, different types of professionals working on each, but inevitably they’re going to come together, and Oracle sees that on the horizon.

Gardner: Let’s go next to Neil, tell us what you took away from the announcements emanating from Oracle OpenWorld?

Neil Macehiter: A couple of things. First, back to your comment around the two-thirds of the SOA initiatives focusing on the data layer. I must admit that doesn’t surprise me, because the proliferation of information, and the variety of information sources that organizations have, and the challenges they’re facing around things like compliance and getting better insight into business operations absolutely depend on getting access to information.

I think that a SOA approach to that is something that can help them address those challenges. And interestingly, given where Oracle comes from, and its strength in the database; it’s actually an aspect of the service oriented strategy that they haven’t talked up significantly if you compare it, for example, to IBM around its whole information server strategy. That would be my observation there.

In terms of the broader recent announcements of Oracle, clearly they’re upping the ante in terms of their credibility in the SOA space, with one eye on IBM, another on Microsoft, and their third eye -- or their eye in the back of their head -- on SAP with NetWeaver. It was interesting, looking at the announcements they made particularly around the SOA suite, which is their bundling of their BPL Process Manager, Business Activity Monitoring, JDeveloper, Business Rules Engine, where they were really talking up the governance aspects of their offering there.

This is something that historically Oracle has not spoken greatly about. They didn’t make a big push there, and that is no surprise given that governance has been such a hot issue recently. I was also interested in the announcement around their WebCenter, which is the coming together of their portal capabilities with more Web 2.0 style technologies around RSS, Wikis, etc. This is still in its comparatively early phases, and is primarily oriented toward the application developer community, in terms of being able to build these mashup style applications that merge content ... to enterprise applications, and support what some people refer to as situational processes. That’s something that IBM has been talking a lot about around its QED Wiki technologies.

Gardner: What about Joe’s point that Fusion is been sort of fuzzy? Now, some people were making a hypothesis that the reason that Fusion was fuzzy, or confusing, was because Oracle was going to make some acquisition. Perhaps, they'd acquire JBoss, perhaps BEA, and bring in this capability for their middleware and development and deployment power from an outside perspective. I guess that’s still possible. Did you get a better sense -- and you can pipe in here too, Joe -- that they’re defining Fusion as the end-all, be-all for their middleware and development?

Macehiter: Well, I think Oracle is clearly positioning the Fusion middleware as their core SOA infrastructure offering, to allow organizations that don’t have Oracle applications to build services-oriented style solutions -- or if they do have Oracle applications to extend those in the same way that SAP, for example, is pushing NetWeaver. Fusion middleware is an enabler of Fusion’s application convergence.

My own view is that Oracle, with its notion of "hot-pluggable" -- and looking at the way it’s made acquisitions, and to my mind some pretty smart acquisitions, around things like, security and identity management, and Web services management -- this is the core platform. They will look either for tactical acquisitions to plug gaps, for example, in the areas around governance or extending some of the management capabilities, but I can’t see them going beyond that. The alternative will be this "hot-pluggable" architecture, which is an attempt to allow them to hook into other capabilities, be they open-source type offerings or solutions from other vendors.

Fusion is the way forward for Oracle now, and that’s where they’re investing their effort. They need that as well, given that it’s going to be the foundation of the unification of Siebel, PeopleSoft, JD Edwards, and Oracle E-Business Suite.

Steve Garone: If I can just jump in for a moment to add a couple of short comments. What I just heard makes perfect sense. I may be a bit of a contrarian here, but I’ve never actually found Fusion, either as a strategy or a product set, very confusing. It was very clear to me that this was Oracle’s way forward in terms of infrastructure and applications. The only confusing part of it in the past has been, a sort of a late incoming message from Oracle that their middleware is going to be “hot-pluggable,” meaning you can come in with elements, tools, and so forth, from other sources, plug them in, and they will work -- and everything will be integrated in a nice fashion.

That was not an approach that Oracle talked about three or four years ago and, in fact, they sent out the opposite message. This is really where the confusion lies. They pretty much cleared that up. The only issue is, do their actions speak as loud as their words do, and is in fact the Fusion middleware stack as “hot-pluggable” as they say it is.

Gardner: They did certainly embrace Eclipse and moved toward open source tooling.

Garone: Begrudgingly at the beginning, I believe.

Gardner: Well, it’s against their nature, I suppose, but everyone has had to do that. It sets us up for this Unbreakable Linux announcement -- a bit of a shocker toward the end of the week-- where [Oracle Chairman and CEO] Larry Ellison got up there and said that they’re going to create their own distribution equivalent to the Red Hat Enterprise Linux distribution. He said they’re going to support it for a lot less money then you might have to pay Red Hat, and they’re going to tell you that it's configured and it’s indemnified, and they’re going to back that up.

So, that puts Oracle in a much larger position as a platform support vendor. It changes the business model to one of subscription and maintenance support, and puts them in a position to start being a bigger partner now, for a variety of different configurations of stack, be it partially open source, partially Oracle -- or for other partners. It seems to me that they’ve really taken a big step toward a much larger software provider role. Any reactions to that, Joe?

McKendrick: Well, Oracle’s been trying to get into this operating system space for a long time. In the past decade, they were one of the aggressive promoters of the whole network computing concept, where theoretically you wouldn’t be relying on a fat client, so to speak, meaning Microsoft Windows. You could access all your applications via some type of thin client or network computing device. They very aggressively were at the forefront of those efforts, as well as other servers, I guess you could call them, preformatted packaged database servers, where you wouldn’t need to worry about the operating system. This is an interesting move, and an interesting end-run around a major Linux provider.

Also, what I find interesting from OpenWorld is their agreement with IBM to support implementations of the Oracle database on an IBM mainframe. Now, this is very historic, in a sense.

Gardner: Yeah, co-opetion in the best sense, right?

McKendrick: Exactly, no one in the mainframe world could have imagined that someday Oracle would be running on the mainframe. It’ll be running within a Linux partition in the z/OS Operating System, or rather the Linux portion of the zSeries.

Gardner: I think the logic there for IBM is, “Listen, if we can’t get DB2 in the account, at least we’re going to get the hardware, and we’re going to get people locked into the zSeries for a long time. We’re going to make some good maintenance money, because, heaven knows, once you get a large, honking relational database in a mainframe it sits there for a long period of time.” I can see the logic of it from both sides, but it is interesting that they’d actually extend hands across the table, and make it happen.

Macehiter: I think it will be interesting if you look internally within IBM to see from where that relationship with Oracle was driven, and my guess would be that it was primarily driven by global services. They’ve been working incredibly hard to increase the usage of mainframe MIPS, by partnering with a variety of vendors that are in direct competition with IBM.

Gardner: Probably not out of the DB2 group, right? Do you think we can read in the tea leaves here -- a little bit by what Oracle has done and by what IBM is doing as well -- that, the heyday of the relational database as a core lock-in, and as a high-margin, expensive aspect of a data center and an IT architecture is coming to a close?

Is Oracle saying, “Yeah, we recognized that lock-in doesn’t work, and so we need to provide the low-cost, best-configured approach, and our real new value proposition is in our business applications, and our ability for people to create data-driven applications quickly, easily, and cheaply?” And, does that mean that at some point in the not too distant future that the cost for relational database, be it an Oracle, a Sybase, an IBM DB2, and -- because of the market forces -- the Microsoft SQL Server, are going to plummet?

And, that perhaps an open-source subscription model for databases is in the offing, particularly as we get into virtualization and multi-core, which makes it very difficult to charge for these things on an upfront license basis to begin with. So, let’s go around the table here. Are we at the end of the heyday of the relational database?

Macehiter: Well, I spent six and a half years of my career working for Oracle, and two and a half at Sybase, in the heyday when the database was absolutely the control point. If you look at what’s happened in the market, the industry has consolidated massively.

Sybase is now really more in the mobile space than it is in the core database space. Ingres is now open source. You’ve really only got Oracle, IBM, and Microsoft with SQL Server vying for dominance in terms of the commercial database. So, you’ve had significant consolidation. You’ve got the entry of, the likes of, MySQL, which now is offering much more of an enterprise offering with its support network.

I think all the signs are there that it’s becoming commodity technology, but that should naturally result in a driving down of cost. I think the organizations also depend more and more on unstructured information that doesn’t exist inside of a relational database. So, everything is there that points to there being a drive down in cost, and it's interesting in that regard, and also in regard to the Oracle Unbreakable Linux, where Oracle makes the vast majority of its profit in its maintenance services. The control point is shifting, and it’s much more about how you get the information out and harness it in the business processes and activities that support employees, customers, and partners in their operations. I think absolutely the writing is on the wall.

Gardner: So, it’s not so much about the object model. It’s about the business model, right?

Macehiter: Indeed.

Garone: Looking forward, we’re going to see a lot of emphasis put on Fusion by Oracle, which really brings together the leverage point of its product strategy, which is applications and middleware. I agree with everything Neil said, I think it’s important to know that what’s going on here, and to some extent it's mimicking what is going on with a lot of other software-type products or markets that have seen open-source offerings come in and have their effect on what you can charge, and how you license, and so on.

So, I don’t think it’s unusual in that regard. It’s a little bit difficult for me to describe database technologies in companies like Oracle as commodities. I don’t really see them that way, because “commodities” implies there’s really not a lot of differentiation and value-add. You’re just kind of getting the same thing from everybody at a bare-bones price. I don’t really think that reflects the database market today, or even the database market down the road. I do think that there are, and will continue to be, other competitive forces that are going to drive Oracle to look elsewhere for its high profits.

Macehiter: I would. Looking at Microsoft’s recently announced quarterly results, if you look at the grades of SQL server from Microsoft, those numbers were phenomenal, again.

Garone: Look at what happened with Oracle and IBM’s recent financials as well. They’re making a ton of money, and it’s a high growth, in the very area that we’re talking about being commoditized. This hasn’t happened yet, but I certainly see these companies setting themselves up for that.

Gardner: What do you see, Joe?

McKendrick: I think Neil and Steve have definitely hit on two important things -- essentially a double-whammy -- that’s impacting the industries. The open-source databases, in particular open-source middleware, are coming into play. Fusion, the whole SOA phenomena, is still getting off the ground, but I think there’s some long-term implications. The companies are making money off of these products, but I think they’re looking long-term down the road.

If you look at what’s happening in both the database and the ERP world, Oracle’s bread is databases, and its butter is ERP applications as E-Business Suite. Many of the functions of the database are, or will be, abstracted out to a service layer, and eventually the same thing may happen with ERP systems, where many of those functions get abstracted to some type of what we call the SOA layer. Granted, it’s going to take a few years -- we’re probably talking about five or ten years down the road -- but these are large ships, and they take a long time to turn around, and they have to look at what’s happening in the space.

Oracle rightly recognizes that the future lies somewhere more out in the middle, not in the big complicated mega-databases, and mega-ERP systems. Recently, in my ZDNet blog I quoted Bruce Richardson of AMR Research, who predicted that, within a few years, probably by 2010, SAP and Oracle customers may actually stop buying applications from their ERP vendors. Some folks have said Bruce may have gone a little too far with his predictions, but I think there’s some credence to it. I’ve actually spoken to customers who have been able to deploy SOA techniques as workarounds to more expensive, painful upgrades to their ERP systems.

Gardner: We’ve already had a little bit of a discussion around this, and I refer our listeners back to an earlier Briefing Direct SOA Insights Edition Vol. 2 for the discussion about SOA and ERP, and how they shake out.

Returning to OpenWorld, one of the things that jumped out at me too was this new warm, fuzzy Oracle: “Oracle’s your buddy.“ That wasn’t necessarily the case in many peoples' minds. We saw [Sun Microsystems CEO] Jonathan Schwartz, [AMD Chairman and CEO] Hector Ruiz, and [Dell Chairman and CEO] Michael Dell on stage. We had HP [Chairman and CEO] Mark Hurd on stage -- so a lot more of this big ecology, fuzzy, warm thing.

Also, quite a bit about Web 2.0 as a catalyst in Oracle’s Fusion. I guess when we talk about Web 2.0, we’re thinking about the use of rich media, social media, the links, and viral approach to reaching, and communicating and collaborating, including RSS, Wikis, blogging, and the whole idea of open rich platforms and suites and features to support that. And, that becomes an element of development, and even becomes a development of how services can be consumed and used.

I’ve heard people say that we really shouldn’t have this SOA repository-register approach, that we should have a wiki, and that people will actually go in and be able to, as business analysts with proper rights, be able to decide, and have input to how services will be used and consumed. And that to try to make services governance it too control-and-command and rigid is the wrong way -- that we should make it open and fluid. Like you would with the definition of an encyclopedia entry, as in Wikipedia, you should be able to do that with the definitions and usage of a service.

Any reaction out there to this one, warm, fuzzy community-approach from Oracle, and then secondarily the role of Web 2.0 capabilities?

Macehiter: I think about the Web 2.0 thing in couple of ways. One, is that it was interesting to see that Oracle has moved away from talking about SOA 2.0 which is very fortunate indeed. In terms of Web 2.0 style technologies and approach and their impact on the enterprise, as with many things, this is going to be a mix of the very structured, highly controlled, registry-repository governed approach to service orientation, where it's needed.

For example, in some processes, subject to regulation, those processes need to be highly structured and highly monitored. Equally, what we’re now seeing is really a recognition that organizations work in fluid ways as well. They are highly collaborative, and involved in a lot of interactions, much of which has gone on in organizations in the past, but out of the purview of IT, because IT hasn’t been able really to provide the support through systems for the way that people really work.

I think that’s where the role of these Web 2.0 style technologies can have a role to play in actually enabling IT to better support the way that people often work, particularly dealing with exceptions to these well-structured processes. I think the challenge there is about culture, and getting the frame of mind and reference within the users, so that they can actually utilize this stuff, and not be afraid to use these capabilities.

So I think it's going to be a mix of both, and what we’re seeing is -- with the Oracle’s announcement of this WebCenter technology, what IBM is trying to do around the QED Wiki, and what Microsoft’s doing around things like SharePoint and the Atlas Framework -- you’ve got the dominant providers of IT for the enterprise actually mixing these two worlds around common platforms, be it .NET, Fusion, or WebSphere in IBM’s SOA foundation. So, I think it's going to be a mix-and-match of these two capabilities.

Gardner: What do you think, Steve? Have we got oil and water here when it comes to SOA and Web 2.0? How do you see them relating?

Garone: I agree with what Neil says, and some of it's too early to call. But I think there is a lot of synergy between the SOA approach and Web 2.0. I think there will be an effort to keep things as simple and low-complexity as possible. I see a lot of synergy there, and we’ll see how it goes down the road. It’s just really early in the adoption to putting stakes on the ground at this point.

Gardner: And how about you, Joe, as a blogger? Do you think this is a fad or here to stay?

McKendrick: It’s going to percolate up. When I think of SOA and Web 2.0, I think of that Apple commercial. You see the two guys, one guy’s an up-tight, straight-laced guy in a suit…

Gardner: "I’m a PC," yeah.

McKendrick: And the other guy is the cool Apple dude, hip and laid back. The guy in the suit could be SOA, and the cool hip guy, Apple guy, could be Web 2.0. And they’re trying to understand each other, at least the SOA guy is trying to understand what the Web 2.0 guy is up to, and why is it that he’s so cool and hip, and can’t he be more like him.

Gardner: So, maybe if they were in therapy together, they might actually work together. It might be kind of cool.

McKendrick: Exactly.

Macehiter: I’m sitting here in front of a Dell laptop, and a Mac Mini, so I’m obviously a bit schizophrenic.

Garone: That’s a really interesting analogy though. The commercial that you just described does point out that the two heads have to get together somehow, and see how they’re going to make that shake out, but I think one’s going to support the other. I really do.

Gardner: And, talk about commoditization, I should think that as companies like Oracle and others latch on to these capabilities that the whole idea of blogging as a function, and having an RSS capability as a function, is just going to be immediately embedded within the background here. But it’s the power of the people and the process that rises above it, and so there’s no need for lock-in, there’s no discussion about who has got the better word processor for blogging. It’s just going to be accepted and embedded into these processes up and down the whole development and deployment lifecycle.

Garone: Embedded as services.

Gardner: As services -- you don’t even have to own this stuff yourself. You can just use Blogger off of Google servers if you want.

Macehiter: Absolutely. It comes down to the sort of environment that you can provide, which is where the likes of IBM, Microsoft, and Oracle are trying to drive things, to make it easy to compose and exploit those capabilities in a way that reflects the pace of change, in terms of the way these activities occur within organizations, and how employees really work. You can’t afford to now throw those requirements over to IT, wait six months, and get the solution back.

The individuals need to be empowered to make the changes, and that’s where Oracle, for example, is trying to drive things around this WebCenter, and the competition framework they’re talking about. IBM has similar capabilities around its QED initiatives. It’s around the competition environment, and you’re not going to compete on whether you’ve got RSS support or not, or it’s your RSS, instead of anyone else. That doesn’t make any sense.

Gardner: As someone mentioned earlier, Oracle’s chief competitors are IBM, Microsoft, and SAP, and in doing these things, and with their Linux move, they take the oxygen out of Red Hat. Red Hat’s ability to price with impunity in the market, which it’s had for a bit now, is quickly eroding.

Novell came in and probably tried to do something about it, but didn’t have sufficient clout, at least in the North American market, to do that. Oracle could very well have an impact on the pricing of a Linux support license. With this Web 2.0 stuff, they really are taking a bit out of Microsoft’s quiver, in terms of their set of features and functions.

If you don’t need Microsoft products installed for doing Web 2.0 and SOA then what do you need them for?

Macehiter: Absolutely! That’s the way it’s driving. Interestingly, in terms of the Red Hat thing, it also does raise the question of what IBM is going to do now. How is IBM going to react to that, given that it provides a fair amount of support around Red Hat?

Gardner: Good point.

Garone: From what I’ve read, a good part of Oracle’s rationale for doing this is they don’t talk about it in terms of, “We’ll give you a better price than Red Hat can, but we will do things for you, if you are in fact an enterprise user. You have special needs in terms of upgrades and getting things fixed quickly and easily, and so on, that Red Hat isn't going to do for you, but we will.” So, they’re kind of positioning this as, “Red Hat’s not enterprising -- we are.” I think that’s interesting in terms of your question about IBM, because IBM, obviously, can make the same claim as well.

Gardner: Do you think we can perhaps look in hindsight, with Red Hat buying JBoss, that that was a mistake, that it allowed them to then be viewed as a competitor rather than, people like IBM and Oracle saying, “Okay, we’ll live with that. We’ll let them control that domain. We’ll cede onto them this Linux licensing thing.” But as soon as they got JBoss, then, boom, they got a big red bull’s eye on their back, right?

Macehiter: I think it was either the day after, or a couple of days after, the Red Hat JBoss announcement that Larry Ellison was interviewed in the Financial Times, and it became pretty clear that he called out what has now ultimately happened. He was making comments about once open source gets good enough, Oracle will just take hold of it, and they did that with Apache. Oracle had its own web server embedded within the Oracle application server, and then as soon as Apache became good enough, he just ripped out his own Web Server and put in Apache.

I think Ellison flagged this and I don’t think it’s coincidental that he flagged it about the time that Red Hat acquired JBoss. It changed the rules in terms of the relationship Oracle had with Red Hat.

Gardner: And IBM has done the same with tools, vis-à-vis Eclipse, and then also in buying Geronimo, going with Glucode, rather coming out with a lower-end approach to open source. I think IBM has hedged its bets very well, and can either turn up the spigot, or turn it off, in terms of which direction and which balance to go open source in. Of course, they have a hardware business to fall back on -- which Oracle does not, nor does Microsoft -- and they have professional services revenue. So, I think they’re in a pretty good position to manage this pretty well.

Garone: Just to declare though, IBM’s acquisition of Geronimo took place significantly before the JBoss acquisition. So I know they weren’t responding to that, unless they suspected it happening down the road.

Gardner: No, I think they were responding to the fact that they need to have both an open source and a commercial strategy.

Garone: Correct. It’s a good strategy, and the only gotcha there -- which IBM has been very clear it’s committed to doing -- is making sure that compatibility exists up and down the WebSphere line, which includes the WebSphere products that are based on Geronimo. I’m not convinced that has happened yet, but IBM has stated a very strong commitment to it, and they’re usually pretty good about fulfilling those commitments.

Gardner: As we’ve discussed in the past, these vendors have been able to follow a certain maturity of commercial products that then has a following counterpart in open source. What’s unique and interesting about SOA is that we’re getting not just a following of a commercial process, but we’re getting an open-source process, almost from the get-go.

So how these companies react to SOA componentry as open source is still an interesting conundrum, I suppose. And we’ve discussed that in the past. Joe, you haven’t had a chance to chime in on this. Do you have any thoughts -- and we’ll make this sort of our closing topic: open source and SOA vis-à-vis what we’re seeing from these large vendors.

McKendrick: Some good points raised here. I think middleware is becoming the battleground. We talked about the abstraction of database functionality and ERP, large application functionality into the service layer. And this is where the battleground is going to be in.

Oracle sees this with JBoss as a huge threat. Oracle talks about making its Fusion components “hot-pluggable” with other software from other vendors, but, make no mistake about it, Oracle wants Oracle customers doing SOA the Oracle way. Oracle has this huge built-in customer base, and it has expanded it greatly over the past year, and its likely many will eventually move to Fusion. But Oracle really has to make a good case as to why customers should consider sticking with Oracle up and down through the stack, and not go on with a commodity or an open-source middleware type of solution. That’s a big challenge all the large infrastructure vendors face.

Gardner: Well, there it is. I would like to thank everyone for joining us. We forgot to do this last week, and that was because we were running out of time, and I was afraid our hour-long conference call was going to end and cut us off. What I would like to do at this point is to go around the table and ask everyone to give us a disclosure. Tell us what companies you’ve been working with, so that we are on the up and up here. Neil, why don’t you start?

Macehiter: Yeah, sure. In terms of the vendors that I’ve mentioned today, Microsoft, IBM are customers, or have been customers of Macehiter Ward-Dutton.

Gardner: And how about you, Steve?

Garone: IBM, Oracle and Microsoft.

Gardner: And Joe.

McKendrick: IBM and Oracle.

Gardner: And for me it’s going to be Eclipse, HP, and IONA, and that’s it for this particular show.

All right, thanks everyone. Also for our listeners, if you’re interested in learning more about our BriefingsDirect business-to-business informational podcasts, or this particular show, the SOA Insight’s Edition, to become a sponsor or to learn more about it feel free to call me at 603-528-2435. Thank you. This is your producer, host, and moderator, Dana Gardner, principal analyst at Interarbor Solutions. Thanks to our guests, and come back and listen again next week.

Listen to the podcast here.

Transcript of Dana Gardner’s BriefingsDirect SOA Insights Edition, Vol. 4. Copyright Interarbor Solutions, LLC, 2005-2006. All rights reserved.