Transcript of Remarks by Steve Ballmer, CEO, Microsoft Corporation
Santa Clara, California
May 9, 2007
STEVE BALLMER: Well, thanks! MR was trying so hard to sound like me, they forgot to dial back the volume on the sound system.
Thanks very much. It’s an honor and privilege to have a chance to be here with you today. And certainly I appreciate the chance to speak at the conference.
I have to say this is the hardest speaking assignment I’ve ever been given. I think the audience is perhaps as broad and varied and interesting an audience as any I’ve talked to. So, we wrote a first draft of the speech, and I said, OK, that’s great for small startups. Then we did a second draft and a third draft and a fourth draft. So, we’ll see. What I tried to do was mix in a few things a little bit for a lot of the very different kinds of viewpoints represented here today, but I thought the right place to start is actually with a perspective on where in some senses the future of software is heading.
I arrived at Microsoft in 1980, and already by that time essentially software had become really what I would call a standard packaged business, and somehow something got put on a CD, or at that time a floppy disk or a tape even when I first came to Microsoft, and delivered to a customer. And whether those are enterprise customers or consumer customers, for many years that was the nature of our business.
Microsoft CEO Steve Ballmer keynoting the Software 2007 conference. Santa Clara, Calif., May 9, 2007.
And yet software is going through now the fundamental transformation that the Internet actually is bringing to many industries that are having to say how do we fundamentally not just deliver our products using the Internet, but how do we redefine the very nature of what we do. And in some senses we have an evolution occurring in the model of computation and user experience that people expect coming from software, if you will as a commodity.
At Microsoft we talk about that not as just Web 2.0, although it encompasses that, not as software as a service, but as the evolution to software plus services.
And it’s clear our industry will evolve. With the Internet, the expectation customers will have on all of us in the software industry to provide ongoing value, up state, services, to any software experience, to be able to expect to click and run any piece of software, no matter where it lives in the world, to have that kind of auto-magically come down and work, not necessarily to get installed but to get roamed and used, but still use the rich processing power in my phone, in my laptop, in my desktop, in my television, whatever the case may be.
I think that once you can assume the Internet connection, the way in which value delivered is enhanced, I think the user experience morphs because we expect much more in terms of community and collaboration in everything that happens with software.
People love to focus on this in the consumer world, but even when we’re using line of business applications, our sense that says we ought to be able to collaborate in an ad hoc way with other people to resolve a question or an issue that a customer has on an invoice, that will become fundamental to the nature of what we do.
Our company grew up as a platform company. We build a platform for the client, we build a platform for the server, we’ve built a platform for mobile devices, and currently we’re in the process of building the Windows Live platform, with computation and storage, networking, directory, rendezvous transport services, all running out in the Internet cloud to facilitate not just for Microsoft but over time for many, many ISVs this transformation to software as a service.
And everything we do will evolve to this model, whether in our case it’s the fundamental Windows and Office experience, business applications, entertainment applications, and we won’t think about Web sites and Web servers and software completely separately; these things will become completely intertwined. And for those of us in the room who are either software companies or software users, we are Web-site builders, we are whatever we are, this fundamental transformation is very important.
At the same time, in some senses we have evolving models in our industry for how software gets packaged up and delivered. We grew up in the era of the PC, and our fundamental model was, if you will, a so-called desktop model. IBM, Oracle, Sun, those are companies that grew up with an enterprise model. And there are differences, some differences in technologies, but even more differences in selling model, support model, service model, pricing models, licensing model.
With the move to software as a service, there are also two new important business models for software that have developed. One is kind of an online and an advertising oriented model. And advertising is a very different way to think about getting paid for our work, but probably pretty important, particularly in the consumer market. And also we see an evolution to what I would call a consumer electronics model. The old world of consumer electronics was a lot of hardware with a little software. The new consumer electronics industry is a lot of software with a little hardware. And as we look and see devices like the Motorola Q phone or the Samsung Blackjack, devices like the iPod, those are really software defined experiences, and yet the monetization model for the software really winds up very much being hardware based.
So, as we evolve to software plus service, we’ll also continue to see evolution not only in the way we build our software and deliver it, but the ways in which we monetize it.
I think there’s a variety of different things that people think about in this world when it’s time to build applications. Sometimes the most important thing as you’re building out a new Web site cum software product is that you’re ubiquitous. The most important thing is touching the most number of people. And you get into issues like cross-platform and making sure that everybody on the planet can participate in your software experience.
We started out with just good old HTML, a little bit now we have things like AJAX as a client-side technology to deliver. We announced our Silverlight technology. Adobe has Flash. But people want the richness of the client experience but with the ubiquity that you get for some of these Web-site software-plus-service experiences.
Some applications, the most important thing is actually a rich user experience. When it comes time to design a plane, to participate in an instant messaging conversation, the notion of being able to have richness in the software plus service experience is very important. We call those experience first applications, because more than anything else, having a rich user and presentation experience is fundamental to what you will do to differentiate your software.
The third model of the world in this software plus service world is what we call the business solutions. In the business solution you may want the richness that comes from a rich client, but you almost certainly want the richness of integration with the tools and applications that users already use like Microsoft Office and SharePoint. And one of the themes we’re going to strike today is how important that class of application is.
If you take a look at the world of enterprise and business solutions, in some senses with the mainframe and the minicomputer the world grew up on the right of this chart. The world grew up with line of business applications and a line of business application platform. Applications tended to be very centralized. Presentation was very dumb; it was done through terminals. Along came the PC, and the stuff on the left-hand side of the board grew up: personal productivity, all of the intelligence in the client, none of it in the server, and infrastructure that spanned these and helped you managed and secure was there.
Today, there’s almost still a large gap between these two. When it comes time to analyze the information that’s stored in a line of business application, when it comes time to deal at an ad hoc way, collaboratively with something that is an exception to formal business process, there’s almost white space in this area of exceptions and analysis, teamwork and collaboration. And in the world of software plus service, there’s a great expectation than ever that these white space gaps will be bridged and interoperability will be established.
As we launched our Windows Vista and Office 2007 products, a lot of focus was on those as individual tools for personal productivity. Yet with what we’re doing in our product line with SharePoint, with Exchange, with our Office Communications Server technologies, we’re trying to establish this notion of Office as a front-end to line of business applications. And those can be instances as services, but in some senses what it lets you do is take advantage of software that is already installed and feed more and more information.
Even in terms of the healthy competition, let’s say, that exists between us and Salesforce.com — and I know you get a chance to hear from Mark, and Mark has a high rhetoric way of expressing what’s going on. At the end of the day we want to make sure, and he wants to make sure that Microsoft Office can be used as a front-end to the back-end service being delivered through something like Salesforce.com, because salespeople are comfortable in tools like PowerPoint and Outlook. A few salespeople are even comfortable with things like Excel for business analysis.
I’m a salesman, so I can say, of course, PowerPoint is the only mission-critical tool, but sometimes a little bit of numeric analysis is a good thing.
So, we see a real opportunity for these business solution applications to marry software and service with popular applications like Office that run on the desktop.
What that caused us to do in the Office 2007 timeframe is really try to birth a new class not only of our own applications but a new class of opportunity for software vendors. We refer to those as Office Business Applications. And they are really designed to allow you to bridge that gap between line of business solutions and personal productivity, to help you manage the world of structured process and information with unstructured collaboration and analysis at the desktop.
We are working with SAP on a product called Duet, which is precisely one of these Office Business Applications. It is something that is designed to let people peer inside MySAP and get information and bring it back to the comfort of Excel to analyze, to be able to take, say, an invoice that has a problem and collaborate on it, and then reinsert it in the SAP workflow through Office, through Outlook, through SharePoint; and to facilitate not only on the SAP side, but we’re also trying to lead in the same dimension with the connection between Microsoft Office and our Dynamics line of ERP products. And no place do I think you see that more clearly than perhaps in our own CRM offering.
But this is a general approach, and for everybody in the room who either builds or user business applications, we now have a programming infrastructure that allows you to extend those and project those and participate in business analysis, in workflow, in document management all the way out to Office as a front-end to these applications which increasingly you will render as services.
We have I think some real critical mass of momentum with independent software vendors on this opportunity, folks like Dassault Systems, who you’ll hear from in a minute, obviously SAP, CorusWorks, Epicor, Approva, Bentley, our own Dynamics product, Artesis, and many, many, many others. There’s a little Goldman report here that talks about how they see the opportunity.
What we’re trying to do is drive deployment of licenses that people own, and essentially for all of our enterprise and all of your enterprise customers deliver more value for the — you know, if you take a look at it, 80 or 90 percent of the cost is in putting in a line of business application in the desktop infrastructure. And yet when you actually talk to businesspeople, for another 3 or 4 percent of the cost, the value grows dramatically.
I was telling somebody earlier today that we have a CEO conference every year, and we had this CEO of a large U.S. financial services institution with us a few years ago. And this was not a technical guy, but he was talking, we were talking about IT and its role in business, and he says, “I love my IT group.” And everybody gathers around: “Why do you love your IT group? Is it because they’re really low-cost?” No. And he says, “I’m not even really sure how to measure whether they’re low-cost or not.” So then people fire him another question: “Is it because your business processes are more sophisticated and complicated and mature than anybody else’s?” And this guy, who was running really a very large financial services company, says, “I don’t know. I’ve never worked in one of our branches. I can’t tell you for sure.”
And they said, “Well, why, why do you know you have great IT?” He says, “Let me show you this. And he takes them to a PC and he shows them a screen from just a simple — what we would say in the technology sense — a simple business intelligence application to be able to actually walk through profitability by customer, activity by customer, flip, flip, flip, do a little analysis.
And it turns out the things that we are most able to get businesspeople excited about are precisely the things that fall between Office and the line of business applications. And I think as you take a look at this demonstration that John Squire and Tarun Tejpal from Dassault Systems are going to do, you’ll get something of a sense on how we can, with marginal additional investment from ISVs, from Microsoft, and from enterprise accounts, through Office Business Applications bring a whole lot more value to the enterprise.
So, please join me in welcoming on stage John Squire and Tarun Tejpal. (Applause.)
JOHN SQUIRE: Thank you, Steve. I’d also like to thank you for helping us reach those additional users you talked about. I mean, one of the problems we have — I mean, as you know, Dassault Systems is the largest provider of product lifecycle management solutions. So we manage the product lifecycle from initial design on through retirement. And our customers include people like Toyota, like Boeing, like Daimler Chrysler, Ford. These are people with heavy engineering content.
Now, despite all of that, we can only touch about 10 to 15 percent of the employees in those companies, and we see the Microsoft Office Business Applications as helping us dive into that other 90 percent. And that’s what we’ll show you in the demo.
STEVE BALLMER: Great.
JOHN SQUIRE: OK. So, this application is actually an aircraft maintenance application, and it’s pretty typical in aerospace, but there are several aspects of the implementation that are not typical, and I’ll point those out as we go through.
In addition, I should mention —
STEVE BALLMER: I think we should probably get people to just switch the machines.
JOHN SQUIRE: Well, we’re going to switch in a minute, but in addition this is a real application. It’s a prototype for a system that we’re implementing at Falcon Jet right now.
So, let’s go. We’ll start off actually on board a Falcon jet, as you see here. And the pilot here receives an error message. There’s a problem with one of the fuel pumps. So, he’ll log the error message, and upon landing he’ll report this to the service technician.
Now, the service technician is equipped with a Tablet PC, and will use the Vista mobility features to go into the cockpit and download the diagnostics. He then uses InfoPath, along with the Office Communications Server, to log the problem, capture the problem description accurately, and communicate with the maintenance support center. Now, taking care of this at the source adds a lot of value and increases our accuracy greatly.
Now we’re back at the support center. And as you can see here, our problem becomes one of thousands of problems within the SharePoint repository. And you can see here we’ve got our cumulative data displayed as a SharePoint portal, and we use this as our support dashboard. And you can see it’s what you were talking about, it’s fairly simple. I mean, there’s an Excel graph that shows the most frequent problems, there’s the actual error messages as they were downloaded in the cockpit, there’s the service report that the maintenance technicians have filled out, and there’s the 3-D product model for those service reports; all the products and all the parts that are mentioned in those service reports are highlighted in this model. So, it’s easy to see the part that’s new, and that’s the 3-D.
Now, that 3-D is displayed as 3-D XML. It’s a lightweight open format that Dassault Systems has developed, working very closely with the Vista team, to ensure compatibility with XAML. And XAML, if you don’t know, is the XML format used by Vista.
So, thanks to the data management capabilities within SharePoint, we don’t have to worry, we know this is accurate data, as captured at the source.
STEVE BALLMER: So, you’re saying basically any user, regardless of whether they know the whole Dassault Systems application experience, can nonetheless get a view of what’s going on right here just inside a SharePoint site?
JOHN SQUIRE: You’ve got it, because our problem is we don’t want the techies, we’ve got the techies, we’ve got the engineers in spades. We want the other guys. So, this is the manager at a support center. He would not typically be a user of this kind of application.
And you can see we’re clicking through these error messages. Here’s a problem with the landing gear. Now, it’s very easy to see the problem is with the actuator arm, which is highlighted here in red.
STEVE BALLMER: I’m glad it’s easy for you to see that. (Laughter.)
JOHN SQUIRE: Hang around engineers long enough and you’ll get it, too.
But 3-D greatly enhances this application. It’s what you were talking about, a simple thing to put together, a portal application, but I can tell you the productivity in the support center goes up dramatically.
So, let’s go back to our scenario. This is the problem pump within the context of this assembly. Now, the pumps are always redundant for backup, so you can see there’s two of them, but you can see which parts need to be replaced.
So, let’s actually go into that service report that was originally filled out. Now, you can see the work order that’s in there, and very easy to go through this. This is an InfoPath form. And it’s clear that he needs to order a replacement part, so here’s the replacement part order form.
Now, InfoPath has already pre-filled all the fields, and inserted the 3D model in the part order so there’s no discrepancy. Now, I’ll pause on this for a minute, because I’ve got to say this is a mix normally, I thought so. We’ve had forms fill applications for, I don’t know, 15 years. I mean, it’s not technology, it’s not leading edge technology necessarily. But I can’t tell you how many times in the field, when I’ve shown this to customers, they jump out of their chair, because despite all the technology, they’re still losing money on rekeying data.
STEVE BALLMER: And they’re jumping out of their chairs primarily because they can see the actual rendition of a part right there in the order form.
JOHN SQUIRE: Right. So, there’s no chance for an error.
STEVE BALLMER: That’s great.
JOHN SQUIRE: So, we’ll submit this and off the system we’ll take care of routing this to the next step in the workflow.
So, let’s go back to our scenario now. And because there have been so many problems with this pump, a support analyst will assess whether or not we need to make a design change.
Now, a support analyst is not typically a user of PLM systems. It’s the type of user we want to reach through Office Business Applications.
So, here we are, and the support analyst, simply by pointing and clicking, is navigating his way through the product. Now, we think this is great. In fact, we think this is revolutionary. And we’ll decompose the airplane here to its major subsystems, and you’ll be able to see he can navigate through those subsystems and find the part that he’s looking for. Now, he could also launch a search.
This is a new product that’s called 3-D Live. It’s a Vista-based product. And we developed this specifically so non-technical users could have access to the detailed technical information.
And what you’re seeing here is we’re actually using the 3-D product model as a guide. So, you point and click, here we’ve selected the subsystem for the fuel system, and you just simply point and click, and behind each of these parts and behind each of these assemblies is all the detailed product information. It could be the original specifications, it could be the tolerances, it could be NC machining instructions, but it’s very rich data contained behind these parts.
So, now we’ve zoomed in on our pump, and we’re going to initiate an instant collaboration session with the supplier who originally designed this pump. So, we use our contextual buddy list, and you can see here, working closely with Microsoft, we have developed a very unique heads up display. Now, at the bottom is room for instant messages, and at the top is a space that captures the 3-D field of view and transmit it using Office Communications Server as a 3-D instant message.
So, there is the supplier session in the lower right, and we can also initiate a co-review, what we call a co-review. So, Office Communications Server will synchronize the view of both users.
So, I think you can see how 3-D in this case, and this 3-D instant collaboration helps to greatly enhance the problem understanding, and speed the resolution.
STEVE BALLMER: So, essentially what we have is something that is an exception, because normally things don’t break. We have something where we want people collaborating, and the goal, of course, is to have full intelligence about what’s going on and be able to have users who are not themselves CAD users, who don’t themselves do the design of the airplane be able to participate in this process through kind of a SharePoint, Office frontend.
JOHN SQUIRE: That’s exactly right. I mean, you could do this other ways. Normally you’d have to go to an engineer, make a request, get the data, schedule a review session, and three weeks later you’d be getting to this point. So, we’re just speeding the process along by giving access directly to the users.
So, let’s go back to our scenario now, and let’s assume that we’ve made a design change with the engineer, and it’s time to issue a new technical service bulletin in tech post. OK, now this is not a new application. Here, Office System has directly queried the PLM database and pulled from it all the PLM information, including the 3D part model, and it’s inserted it into our document template.
But what we’ll do here is we’ll do one final step so you can see how it works. First, thanks to the extended XML support within Office System, all of our data remains live. So, this table on the right, that’s the downloaded XML data. It includes things like part numbers, engineering change numbers. We can access this and reuse it at any time.
STEVE BALLMER: So, I get additional viewing and capability, but I don’t actually have to be running your application inside Word; I’m running an XML rendition of it.
JOHN SQUIRE: Exactly. And it’s not our application, it’s the data from our application.
STEVE BALLMER: The data from your application.
JOHN SQUIRE: And I will say as an application vendor the extra XML support in Office System 2007, that’s a boon for any application vendor that has a lot of data they want to share with the masses. It’s really an effective way to get the information out there.
And I’ll show you a little bit more. So, here we’re going to do a manual step. We’ll manually find a 3-D part model for that pump, and we’ll insert it in the area of the document with the work instruction.
Now, thanks to Office 2007 this is much more than just a dumb image. I can associate a lot of information with this, such as the bill of materials for this assembly. And we display the bill of materials on the right. You can see we can click down through it. And as we select the part in that bill of materials, that part is highlighted in the 3-D model on the left.
And I can go further. I can take the information from the bill of materials, part names, part numbers, all boring stuff, but I can take it, insert it into the document. Then later when a reader clicks on that part name, it highlights the part for them.
So, you can see the result is complete, accurate, timely information distributed to the field.
So, that’s a quick example of Office 2007, working with Dassault Systems’ PLM solution in a real industrial setting to solve a business problem.
STEVE BALLMER: That’s great. Thanks so much, appreciate it, John. Thanks, Tejpal. (Applause.)
This new class of applications, which fits perfectly in the notion of software as a service model, I think you get something of a sense on how you can really add the kind of value in an enterprise, and again people working cross company borders, collaborating, where essentially you need very little on the client in order to have a dramatically extended experience across the applications.
We ran a contest that we kicked off about a year ago at our Tech•Ed conference where we asked a bunch of ISVs and software vendors to go out and work with nonprofit organizations. And we said we’d run a little contest. The nonprofit organization that had the most interesting and clever application of this form, an Office Business Application, we would reward with kind of a prize and some additional funding for that not-for-profit. And maybe we’ll just spend a minute, we’ll announce the winner here today, and I want to run a short video. There are over $160,000 in prizes for these nonprofits. There were many, many kinds of applications, fundraising, compliance, data management. The winners were Habitat for Humanity, Midnight Basketball, FUNDAMED, and the International Med Corps. And if you’d roll the video, it will give you more of a sense for Office Business Applications.
STEVE BALLMER: In a very short time I hope that gives you a little bit of a sense not only on some of the big trends we see in our software business, but one of the very clear opportunities in the enterprise around Office Business Applications.
I thank you for your time today. I’ll look forward to a rich dialogue up here with MR. It’s certainly been my pleasure to have a chance to address you today. Thanks. (Applause.)
MODERATOR: Should we be saying, OBA, OBA, OBA for O-B-A, Office Business Applications?
STEVE BALLMER: Absolutely, works by me.
MODERATOR: OK. All right, folks, OBA!
OK, that leads me to my first question on Vista itself. So, Windows 95 — look back: Windows 95, XP, Vista. How do the three things compare?
STEVE BALLMER: Well, each came at kind of a different time for the industry, different times for Microsoft. In a sense, Windows 95 was really the kickoff for the broad acceptance of the PC. And you can’t repeat that, since PCs are very popular.
On the other hand, in terms of kind of range and amount of innovation, speed with which customers have embraced the new products, certainly Vista is really in a class by itself.
MODERATOR: And we’ve heard a lot about Vista, obviously you’ve been marketing it a lot these last few months. How about Office 2007? We haven’t heard as much. And you’ve shown us some of the products here. What’s been the customers’ reaction to that? What’s been the uptake of the product?
STEVE BALLMER: I think there have been kind of three important reactions. The first reaction most people, of course, have to Office 2007 is the new user interface. And the new user interface, if you let somebody just play for four, five, six minutes, we generally get an overwhelmingly positive response, because one of the number one issue people have framed for us on Office is, hey, I know there’s a lot more stuff in there that I’d like to take advantage of and I can’t, and with the new so-called Ribbon user interface we’ve really changed that.
The second big reaction is to the XML support, and what that does in terms of openness and extensibility. And then number three is really what we’ve don with SharePoint.
And then, of course, every individual has got their new favorite features. Mine are in PowerPoint. I happen to think it’s impossible with the new PowerPoint to do an ugly presentation, but I won’t run a contest on that.
MODERATOR: A couple of things. Now that you’ve released Vista and Office 2007, and “Longhorn” I guess is coming up in a few months on the server side, all these have been large, monolithic releases in a sense. And you talked about software and software as a service and so forth. Those call for short, incremental releases, right?
STEVE BALLMER: No.
MODERATOR: No? OK. Tell me more.
STEVE BALLMER: No, I think the world of — if anybody thinks the world of the future of software is only innovations that can be done in a short time are worth doing, I think that’s actually wrong-minded. I think in a world of software plus service, every good software company is going to have, if I can borrow an analogy from sports, we’re going to have different twitch muscles. We’re going to have to have the ability to release value very regularly, three months, six months, whatever. We’re going to have significant, more significant releases every year, year and a half. And then there will still be long cycle innovations that require more effort, more testing, more comprehensive approach.
And so I don’t think anything migrates to all short twitch, and at the same time I think there will not be software products around that go a long timeframe without any improvement or enhancements.
MODERATOR: So, your Office Live releases would be far different than let’s say the next version of Office?
STEVE BALLMER: Well, Office and Office Live are separate, but in a sense they have a lot to do with each other. And so I think what an Office user will say is through Office Live I get a lot of value delivered to me regularly, and I get some things that are actually bigger and more comprehensive delivered to me less frequently, but importantly over time.
MODERATOR: So, looking at the vast empire you have at Microsoft, you have all these four divisions and multiple products in multiple areas. You know, Jack Welch was famous for having said, “I want to be number one or number two in each market I play; if not, maybe I shouldn’t be there.” What kind of criteria do you use at Microsoft in judging the success and the market share of the diverse set of products you have? And are you willing to be number six or seven or eight in the market for a long time, or do you strive to achieve for dominance in everything you play in?
STEVE BALLMER: Well, I don’t think any company ought to do anything where they don’t aspire to be number one. I mean, you may not get there, but it would be hard for me to run a company that didn’t aspire to be number one at everything that it did.
I think Jack Welch said he’d get out if he wasn’t number one or number two early in his career, and what he actually said at the end of his career was he missed a lot of important opportunities by not willing to be patient when they were number three or number four, so they got a chance to be number one or number two.
We’re a company that aspires to be number one in everything that we do, and I think that’s a phenomenal attribute. We’re also guys with a reasonable amount of patience. We’re persistent. Sometimes we get it right from the get-go. Sometimes the market is ready from the get-go, sometimes it’s not. Sometimes we’re first with a new innovation, sometimes we’ve got to come from behind, but we keep coming and coming and coming and coming.
So, sure, I always want to be not number one or number two, I want to be number one. And I come to work every day thinking that in the areas where we’re not number one, some day we will be. (Laughter, applause.)
MODERATOR: Great quote.
So, will you be number one just with organic growth, or will you be number one with acquisitions, too?
STEVE BALLMER: Well, we certainly do do acquisitions. I think in the last 12 months we bought 15 to 20 companies. And most — yet what we buy is typically — the companies we’ve bought are not typically large relative to our overall size. What that means is we buy things primarily to fit into the context of what we’re doing, or as platforms for whole new businesses. We bought a company about a year ago called Azyxxi, which took us into the healthcare software business. Six years ago, we bought Great Plains Software, which took us into ERP. But most of what we do are fill-in acquisitions to our product line, technology acquisitions.
I don’t think you should expect that most of our growth will come from buying huge companies and taking cost out. I think that’s more typical of other industries.
MODERATOR: So, would you pay $40 or $50 billion for any company out there? (Laughter.)
STEVE BALLMER: Anything is conceivable, but if your question is more specific, we, of course, don’t comment ever on any kind of speculation about any kind of acquisition. There could well be companies out there that are worth $20, $30, $40, $50 billion, and yet it is not our first — we have not by default opted for large acquisitions as part of our strategy, but we don’t rule them out either.
MODERATOR: So, you do have — for the right deal you do have the appetite to spend money?
STEVE BALLMER: We’re willing to do the things that are right for us to do relative to our ambition and our strategy, and in general those seem well served by smaller acquisitions. We’re open though to doing larger acquisitions.
MODERATOR: So, let me move to open source, another topic here that’s been discussed the past two days. You’ve forged a relationship with Novell, and you’ve kind of got at the open source space that way. Do we envision other relationships like that, and what’s the long term play for you with the open source market?
STEVE BALLMER: Well, obviously I won’t talk about open source in general because in some senses that’s very broad. There’s a lot of great open source development going on, on top of Windows. I think that’s exciting.
What we did concretely with Novell is forge a relationship around Linux, which is the most famous open source product, but by far not the only open source product. And we had basically two goals. Number one was to achieve a higher level of technical interoperability between Windows and Linux, and number two was to ensure that our intellectual property was properly licensed to at least Novell’s Linux users, and I think our relationship with Novell accomplished both goals.
Do we compete Windows versus Linux? Absolutely, do not be confused. If you come see us and say, how about the Novell deal, the first thing we’ll try to do is sell you some Windows. But for whatever set of reasons, if you decide that you need to use Linux, we want to make sure that you’re able to use it in a way that is properly licensed and interoperable, and we’re very excited about the Novell relationship in that sense, and we’d be delighted to extend that to other people in the kind of Linux distribution area.
MODERATOR: So, moving to the whole ERP space, we had Hasso Plattner yesterday in the morning, we had Marc Benioff in the afternoon, we have you today, and MBS has had a lot of changes. There have been management changes and some other things. So, what’s going on there, and how do you position against these other guys?
STEVE BALLMER: Well, we’re actually going gangbusters. I mean, we have the fastest growth rate of anybody in the business these days. We’re adding CRM customers, ERP customers at quite a nice rate. It took us a while to really understand the business, but we’ve been at it six years and I’d say some things we had to figure out, the first couple or three, but we really have pretty good momentum.
The truth is we’re only competing, say, with SAP in some ways. We have a product that supports about 3,000 tables. That’s kind of a description of how much business process description. SAP has got about 35,000 tables. Some enterprises need 35,000 tables, and some businesses don’t. And so in a sense I would say we offer a certain simplicity and ease and speed of configuration, and they offer kind of a massive opportunity for implementation of deep business complexity. We run into each other sometimes, but more frequently than not they are not a competitor when we add a customer, and we are not a competitor when they add a customer.
MODERATOR: So, what’s happening with the Duet product, and what’s the rollout, what’s the plan there?
STEVE BALLMER: Yeah, as I said, Duet really does fit in this category of an Office Business Application, and I’m more excited about the Duet concept and the partnership with SAP than even when we started, and yet I think — I was with Henning Kagerman about 10 days ago from SAP, and we agreed we probably haven’t made as much progress per unit time as both he and I want. So I see us continuing to push aggressively both companies to extend the value in that Duet offering.
MODERATOR: As you know, the next big area that everybody has been talking about recently is virtualization. Can you talk a little bit about what Microsoft sees in that area, and what your plans are?
STEVE BALLMER: Yeah. Virtualization is obviously a big theme, and one of the things we’re working hard on is integrating a virtualization layer into our operating system, client, server, as well as appropriate management technologies for managing virtual machines. We think that will be a fundamental part of all operating systems.
Virtualization happens at many levels. It will happen at the chip level, the operating system level, the application level, the presentation level. With our Terminal Services offering we’ve got presentation virtualization. With our SoftGrid technology we’ve got application virtualization. With the virtual machine layer that we’re building in we have it at the OS layer; and a great partnership really with Intel and AMD to take advantage of the virtualization at the hardware layer.
MODERATOR: But in terms of product rollout, from what I’ve read, it’s after Longhorn, there’s another timeframe of six plus months before —
STEVE BALLMER: Well, we’re in market today with our Virtual Server product and our Virtual PC product. We integrated the Virtual PC capabilities into our Windows Enterprise Edition. We will have an upgrade, and the integration of virtualization into the Windows Server post our “Longhorn” release, that’s right.
MODERATOR: So, the other area that was discussed this morning is with Ed Zander, who brought his Q, and who talked about Windows Mobile and so forth. So, what’s your take on that whole enterprise mobility space, and what are you doing there in terms of getting more products to customers?
STEVE BALLMER: Yeah. I think that’s an area frankly that’s going great. Windows Mobile in general we sell now close to 20 million Windows Mobile devices a year, which is above the numbers you’d find for Palm or BlackBerry or some of the other guys. We’ve had particularly good success in line of business enterprise applications, whether that’s the work we’ve done with the folks at Symbol, now part of Motorola, whether that’s the work that we’ve done with customers like UPS and FedEx and others. It’s been a real focal area for us, not only as a platform but also the application side.
MODERATOR: So, in terms of the ISVs who are out here in the audience, and enterprise customers, what’s the most exciting thing that Microsoft is going to deliver in the next year? What’s something exciting that you go, hey, this is going to knock your socks off?
STEVE BALLMER: Well, I think in the next year means I can’t take credit for anything in the last few months. I think the most significant thing is the work that we’re doing around .NET with Silverlight and Silverlight Extended, some of which we kind of went through and disclosed as part of our MIX conference two weeks ago. It will be shipping over the next several months. And if you’re trying to build a very rich software plus service Web style application, I guarantee you we’ve got incredibly exciting stuff coming this year.
MODERATOR: The early indications of that have been very positive from your conference recently, but what was not clear then is the business model. Are you just going to sell it as a product?
STEVE BALLMER: Well, Silverlight is a part of Visual Studio, and our Expression Design Suite. So, you pay for the tools, and then the runtime can get distributed by the ISV with their Web site cum application.
MODERATOR: Steve, this has been fantastic. Thanks for coming here. And we hope to see you maybe next year in Vegas.
STEVE BALLMER: Great. Thanks, all of you, appreciate the time. (Applause.)