Remarks by Steve Ballmer, CEO, Microsoft Corporation
Gartner Fall Symposium 2003
October 21, 2003
ANNOUNCER: Ladies and gentlemen, please welcome Vice President and Gartner Fellow David Smith. (Applause.)
DAVID SMITH: Good morning. I’d like to welcome you all to day two of the symposium. We have a big day planned for you and we’re starting off with a very big event, a mastermind keynote interview of Steve Ballmer, CEO of Microsoft, but first I’d like to introduce my colleague, Research Vice President Tom Bitman. (Applause.) And the guest of honor, CEO of Microsoft, Steve Ballmer. (Applause.)
Welcome, Steve, and thank you very much for joining us. I think this is the fifth year in a row you’ve been here and we’re very pleased to have you.
STEVE BALLMER: Pleasure to be here.
DAVID SMITH: Steve, as we do every year, as you’ve seen us do, we ask our attendees what questions they’d like to ask you and maybe what comments they have. Could we roll that tape?
INTERVIEWER: Who was that guy?
STEVE BALLMER: Get rid of the last guy.
INTERVIEWER: Who was that? (Laughter.)
I see a whole bunch of questions and as we had to kind of frame what we’re going to talk about, security, Linux, costs and the future, we had to summarize it in four areas that we’re going to talk about, let’s talk about costs first. Anybody would have to be on another planet to not know that costs are first and foremost on most people’s minds these days. What specifically is Microsoft doing to help enterprises lower costs?
STEVE BALLMER: I think the important opportunity for our company, for our industry is really focusing in on total cost of ownership, with primary emphasis, frankly, around all things required to build, deploy, operate and manage applications. That cycle, if you look at where the bulk of IT cost is, it’s involved in that cycle.
We’re very focused today, of course, on some of the specific costs of deployment and management of security, particularly the deployment of patches; it’s a big issue. But if you take any part of that process from build through operate, then we see lots of opportunity to take costs out of the system by improving the manageability.
We embarked on a project, oh, a little over a year ago I guess now that we call our Dynamic Systems Initiative to sort of go in and re-plumb the fundamental manageability of the product. We’ve done a lot of things short term in our Systems Management Server, the work we’re doing with our Software Update Services to provide well designed automation and deployment of patches; so a lot of work, I think, that can sort of use automation to cut total cost of ownership, and that’s a real priority.
INTERVIEWER: Well, obviously we’ve said many times total cost of ownership is really important and the initial cost is not the most important, but the initial cost is a component of it. And recently you had to do some special deals in some emerging countries such as Thailand where your real competition we know is piracy and it’s really a consumer issue not an enterprise issue; we know those things. But you have lowered the price down to about 10 percent of what people in the U.S. pay. How do you explain that kind of price differential to folks in the U.S., folks like here in the audience, why they should pay so much more?
STEVE BALLMER: We did an experiment in a very poor country to see whether or not actually having a low price on software with a very low-end, not a very good PC would actually stimulate consumer interest and uptake in a very poor country. The data on it is actually very mixed. The uptake on the offer from the Thai government has been quite low, but we’ve been pushed for years on this question of would it help in very poor countries to really work somehow on this aspect. The experiment, as I said, has not been a great success and I’m not clear we’ll continue it.
There are, of course, markets where we do have better prices. The education market is an important one. Our education customers do get lower prices from us than anybody else, and we recently announced a program for schools in poor countries or poor schools in richer countries where they can get Office and Windows at very, very low prices, just recognizing the fact that the budgets are simply too low if you look at the schools in poorer parts of the United States or Namibia, countries of that ilk.
INTERVIEWER: We understand that you have an announcement today. We’re all under NDA on it so we can’t talk about it. (Laughter.) But we thought you might have a couple of words you wanted to say.
STEVE BALLMER: Yeah, we’re announcing the new version of Office as well as a suite of surrounding products we call the Office System 2003. I’m pretty excited about it because I think it is the most fundamental release of Office we’ve done in a long time. Whether your perspective is the individual who simply wants to get more personal productivity, where there are improvements we’ve made in Outlook or PowerPoint; teams of people, I think, is where the thing really shines, the way we built in collaboration or frankly the great XML support we’ve built in that really lets Office participate in and be the, if you will, intelligent client to XML Web service line of business applications in the enterprise, I think it will be a big deal kind of across the board. Certainly it’s the most significant release of Office we’ve done in a number of years.
INTERVIEWER: Well, why should people upgrade?
STEVE BALLMER: Individuals I think are going to want to upgrade for, if nothing else, if you’re not running the new Outlook you’ll want to run the new Outlook; I guarantee it. That in and of itself for most end users will justify the upgrade cost.
From a corporate perspective the ways in which we built in support for collaboration I think will really revolutionize the way people collaborate on business plans, marketing plans, product proposals, financing proposals, you name it. And I think that will be a powerful motivator for the organization, if you will.
I was telling some folks we’re working on applications, for example, when we’ve worked on the U.S. Air Force where they’ve actually taken the time down from five hours to two hours that it takes to get a fighter pilot off the ground once he receives his mission by using Office System 2003 as an XML Web service client to information, and that kind of productivity shows enterprise payback quite quickly.
INTERVIEWER: Let’s take one of those questions we had from one of the questioners — I’m not sure who it was — but about Office for Linux. Would you ever consider that kind of pursuit?
STEVE BALLMER: You never say never; it makes no sense to do that. We have no current plans and we don’t really see it as a very interesting opportunity for a variety of reasons, and let me kind of give you a few of those.
Number one, the uptake of Linux on the client is not very large. Linux is still much smaller on the client, for example, than the Mac, and I can tell you how big our Mac Office business is; it’s a nice small business but it’s a small business. So if somebody says you have a new opportunity, you’ve found a platform that’s less popular than the Macintosh and you can go write a new application for it, I’m clear that’s a good starting point. (Laughter.)
Number two, nobody pays for software on top of Linux. People who use Linux will also tend to use other applications, middleware, et cetera that has the same price point. I can argue till I’m blue in the face about why I think there are great reasons somebody should use Windows instead of Linux on the client, but the person who’s made that decision is probably not a good candidate to come along then and pay over 100 bucks to us for a copy of Microsoft Office, further reducing the size of the opportunity for Office on Linux. (Laughter.)
And at the end of the day, for our customers as well as for us this isn’t about religion; it’s about business. These people have something they need to get done. We need to try to figure out what they need to get done and what they’re willing to pay for, and that’s what’s going to really make a healthy environment in which we can innovate and provide our best work to our customers.
INTERVIEWER: So if Linux expanded beyond the role it’s been popular in today, Web servers, very specific utility servers and desktops, if it really did become more general purpose and there is a market for software to be sold on it, then you might be interested?
STEVE BALLMER: If, if and if; that’s why I never say never.
INTERVIEWER: Never say never. (Laughter.)
STEVE BALLMER: Exactly.
INTERVIEWER: So are Linux and Microsoft two things like oil and water that would never go together or can you foresee some kind of a strategy like you’ve done with other things like embrace and extend, or some opportunities to actually use Linux and Open Source?
STEVE BALLMER: Well, I would say again a couple comments. First of all, it’s clear we compete with Linux. That doesn’t mean you don’t embrace your competition. Excel competed with 1, 2, 3 and part of the method of competition was by embracing the 1, 2, 3 user base. But Windows and Linux compete and we think part of our job is to make sure we have more innovation, better total cost of ownership, better overall value proposition than Linux has.
Part of that will involve making sure that we have good interoperability with other environments, including UNIX based environments. And if you take a look at the work that we’ve been trying to do participating in standards bodies, in the WS-I with IBM around XML, we’re trying to make sure we have a good interoperability story with UNIX, with Linux, with anything else that comes down the road that moves in that direction.
Third, I actually want to make Windows the best platform to port a UNIX application. We’ve told the engineers — and there are some cases where I think we’re there today and some cases where we’re not — but we’ve told our people let’s make Windows a better target platform for UNIX applications, whether they come from Linux or other forms of UNIX, than any other platform on the planet. So we do want to have that kind of strong interoperability migration story with all UNIX environments.
INTERVIEWER: Let’s pursue this Office on Linux a different way, a bit of a different way; not Office on Linux but Microsoft has a strong business model around Windows, Windows on the server and so forth, but could you see a time — and don’t say, never say never — could you see a time where your business of software built on top of Windows becomes more important and perhaps you even charge less for Windows, Windows becomes the commodity operating system, a base for Microsoft software as a way to compete against Linux and other commodity operating systems?
STEVE BALLMER: In the software business this is a question that’s hard to answer, because we do have the ways the products get defined we have choices. We had a choice, for example, to put our application server, make it separate from Windows, charge more money for it and essentially put the value outside Windows. We made the choice instead to extend and enhance the definition of Windows to include an application server.
And so when you ask me the question you’re sort of asking me not do we expect to do innovative value-add work — of course we do; the question is do we expect to package it outside of Windows or with Windows. It’s our basic view that our customers would prefer to see it packaged and integrated with Windows so that, if you will, you can get the simplest, cleanest design that we can possibly deliver to the market.
We don’t put everything into Windows but we look for opportunities to provide integrated innovation inside the Windows product that helps reduce, hopefully, the cost and complexity that our customers face.
And every time we come up with a new feature we take a look, should it be in, should it be out, is it part of the definition or not, and if we think it adds fundamental value we’ll put it into the basic definition of the Windows product, and I think that’s led to a lot of good things.
INTERVIEWER: Hasn’t it also led to a situation where you have an awful lot of stuff going into one code base, making it difficult to get new releases out?
STEVE BALLMER: Does it really make any difference if you happen to put things on two CDs, charge two prices and have two installation routines? The software that gets delivered has to work together. The pressure from our customers, whether things happen to come on two CDs with two prices or one CD with one price, the pressure to get our software to work together is equally as strong.
One of the great advantages we think we offer to our customers versus anything else in the market is the ability to essentially take cost and complexity out of the system by having us think through these relationships between these pieces of software as opposed to having to force our customers to cobble them together themselves. That is part of the Open Source world, the customer puts things together. We think part of our value proposition has to be we have to take a lot of that effort out.
Certainly that kind of integration has offered value. Nobody doubts today that it was a good idea to make a TCPIP stack part of Windows. It was controversial at the time it was done; it’s not controversial today. I think that integrated innovation value proposition is one of the key things that we bring go the market relative to any other supplier in the business.
INTERVIEWER: You mentioned Open Source, and some of your competitors have embarked upon strategies where they’ve put some of their intellectual property into Open Source. Can you see any scenarios where you might do that with your technology?
STEVE BALLMER: Certainly we have, in fact, there are many, many scenarios where we will choose to share our source code with our customers, with our partners.
INTERVIEWER: That’s not the same.
STEVE BALLMER: It’s not the same, but that’s what we do. If you’re a large enterprise we’ll share our source code with you. We do not have the option of publishing our software at no fee. What does Open Source mean? It means everybody can look at the source code. We say we can’t do that. We can give our customers the ability to look at our source code and it means there’s no cost of acquisition; inconsistent with the $6.9 billion that we spend on R & D to turn around the work of our engineers and make it available by and large at not cost.
Most people who are putting their software under Open Source are doing so because it hasn’t been very successful when it was sold. And if something is not very successful sold, sure, why not make it free. That’s not where we come from. We’re trying to build software that actually delivers value.
INTERVIEWER: Well, along those lines, the questions on Open Source aren’t just about free versus not free; it’s also about the process. Why do you believe a vendor-driven development process is going to produce more quality code or better quality code than an Open Source process?
STEVE BALLMER: I’ll give you my reasons, but in a sense it doesn’t matter what I say or what I think; the test of time will prove out what’s the case.
We get a lot of questions, people say we have the security problems, let’s fix it, what’s your roadmap for this, what about this, what about this, what about this. There’s no roadmap for Linux. There’s nobody to hold accountable for security issues with Linux. There’s nobody sort of, so to speak, rear end on the line for issues; it may or may not be an issue.
We think it is an advantage that a commercial company can bring. We stand behind the products. We indemnify for the intellectual property that’s in the product. We provide that product with a roadmap. If there’s problems and people do have security issues, I’m SteveB@microsoft.com, they know where to send e-mail and give somebody a hard time about it, and to the very best of our ability to get a response. None of that is true in the other world.
Will there be better code or not better code? Is the model a better model? So far I think our model works pretty well.
Somebody will say, well there’s a scenario where Linux has good penetration, to which my response will be, okay, I know that scenario, I’ve got engineers working on that scenario, we’re going to innovate in that area, we’re going to take costs out, we’re going to work and work, exactly what people should want us to do is to push ourselves and work harder.
Should there be a reason to believe that code that comes from, how do I say this, a variety of people unknown around the world somehow will be of higher quality than people who get paid to do it professionally? There’s no reason to believe it will be higher quality. I’m not going to claim it necessarily will be worse quality. Bu why should code that may get written randomly by some hacker in China and contributed to some Open Source project, why is its pedigree by definition somehow better than the pedigree of something that is written in a controlled fashion? I don’t buy that.
INTERVIEWER: It may not be obvious but a lot of people believe that Open Source is by almost definition more secure, and a lot of the reasoning is that they are not beholden to one company to fix things, they can do it themselves if they want to or they can go to a third party to do that. So how do you answer that particular concern?
STEVE BALLMER: The data doesn’t jive with that. If you just take a look —
INTERVIEWER: But that’s what they believe.
STEVE BALLMER: That’s right, so let me give you the data and then maybe they’ll believe something different; let’s see. (Laughter.) And I encourage everybody to go to the cert Web site.
In the first 150 days, and this is the kind of stuff we track, the first 150 days of Windows 2000 we had 17 critical vulnerabilities. The first 150 days of Windows 2003 we had four critical vulnerabilities. The first 150 days of Red Hat 6, go check the number, just go check the number. It’s five to ten times higher than what we are showing.
The vulnerabilities are there. The fact that somebody in the middle of the night in China who you don’t know, quote,
it and you don’t know the quality of that, I mean, there’s nothing per se that says that there should be integrity that come out of that process.
At the end of the day it’s people who write software, we have a methodology, we have an approach, we have a testing process that we know can lead to a sustained and predictable level of quality.
Imperfect; our customers are not happy with four critical vulnerabilities in the first 150 days of Windows 2003. So I’m not saying, hey, boy, aren’t we the cat’s meow here, we’ve accomplished it all. No. All I’m saying is I think it is absolutely not good reasoning to think you’re going to get better security out of Linux and certainly the data is not there.
INTERVIEWER: Well, speaking of your efforts in security, about a year and a half, almost two years ago now Microsoft embarked on this Trustworthy Computing initiative, and you’ve started to take it very seriously and I think people will give you a lot of credit for that, yet the results haven’t been optimal. Are you surprised by the results so far? Did you expect to be further along?
STEVE BALLMER: I think we’ve made great progress and I think we’ve learned a heck of a lot about what all is encompassed.
You know, if you take a look and say how are we doing in reducing the vulnerabilities in products released since we embarked on what I might call the trustworthy computing release process, we’ve made dramatic strides; maybe not good enough, four critical vulnerabilities, still not good enough, but we’ve made dramatic strides. You see it in Windows Server 2003, you see it in Exchange, you see it in the SQL Server releases, you can see it across the board, dramatic strides.
We put a lot of effort and energy into improving our patching process, probably later than we should have and now we’re just gaining incredible speed. Our patching process needs to be more predictable, people want smaller patches, we need one simple installation process for patches, which we haven’t had, we need rollback on patches, we need a more consistent patch policy, people want more predictability about when they come out, people want better patch management tools, we’ve responded on that, et cetera, et cetera, et cetera. I mean, there’s a whole set of things that people absolutely want and we’ve been raising our game.
The third bit of learning, and this is more recent, and you could say shame on us or you could say thank goodness we got started on that trustworthy computing initiative so that our eyes were wide open; a lot of people are saying, okay, we know we need to patch but we want to be able to patch on our own schedule, we don’t want to have to patch on the hacker’s schedule, so you need to give us better technology that helps us shield our system or systems. If it’s a corporation you may want to shield the whole system. If you’re an individual you may want to shield an individual system. Many of our customers will say that the most recent vulnerability primarily came from laptops which got infected and then reintroduced into the environment or from people VPN-ing in from home. You’ve got to give us tools that will help inspect systems that come into our network and to make sure that they are up to snuff with the security standards and approaches inside our business.
So this notion of these inspection and shield technologies and all we need to do with distributed firewall, that’s an area where we’re really cranking up. We announced that we’ll do a new release of Windows XP and a new release of the Windows Server 2003 to incorporate some of these technologies. You’ll be able to use those to shield down level systems in the environment.
So I know we need to do better but we’re in this very challenging position where in a sense the hacker only needs to find one vulnerability; we need to keep them all out. And we’re really in high gear on this and we know our customers have very, very high expectations.
One more thing; I know you’d like to move on, but one more thing I think is important. We’re also trying to work carefully with law enforcement. At the end of the day this is an imperfect — whether it’s with us, whether it’s with competition, there will be security vulnerabilities for the foreseeable future and there needs to be more deterrent, criminal deterrent to the kind of malicious acts that people are involved in.
INTERVIEWER: So code quality, patch management and shielding are three major things you’re focusing on. In six months or in one year are we going to have the number of complaints about security that we have now?
STEVE BALLMER: I hope not, I think not and we are sort of committed, if you will, as priority number one to make sure that’s not the case.
INTERVIEWER: Okay, we’ll keep the tape for next year and we’ll show that.
STEVE BALLMER: Play it next year and at least, well, I’ll be able to recite everything we’ve done and I think it will have had a major impact.
INTERVIEWER: One of the challenges I think you face right now in this area is confidence. Because people do know that you’re trying, they give you credit for that, and they know that they’ve watched over the last few years, that a year and a half, two years ago you started to take it very seriously but because you’re trying they feel that perhaps you don’t really have a handle on it. So what can you say to make people have a little more confidence that you’re going to be able to really make some progress?
STEVE BALLMER: We rarely fail at something that is our top priority. This is absolutely our top priority. This is tougher though. It’s tougher. It’s not a game of horseshoes; you don’t come close. One vulnerability can still cause a variety of problems for the folks in the audience. But absolutely we’ve got our best brains on it. We’ve told people anything we need to do — acquiring new technologies, people, approaches — we should put our heads down and go get that stuff done. And we’re not going to let anything stand in the way.
We understand this is an issue of customer satisfaction. It could slow down progress on IT for the whole industry. We realize it’s a competitive issue and in some senses our ability to respond and perform well is one of the key issues that could differentiate us from competition.
So in every sense it’s kind of a defining moment issue for us. We see it that way and we’re committed to address it in that way. We’ll have gobs of detail. We’re going to try to train 500,000 people on the best ways to secure Windows systems over the course of the next nine months or so. And we’re just super committed to getting people where they need to be.
INTERVIEWER: But just in terms of code quality and bugs that are causing security problems, a lot of the problems, most of the problems, probably 95 percent of the problems we’re seeing are from code that was written six, seven, eight years ago. Does this mean that you’re just going to have to rewrite some of this code over time and maybe incrementally or maybe start over in a few years?
STEVE BALLMER: No, there is code, absolutely, that will need to be redone and we have done that and we’ll continue to do that as we find pieces of code that need to be redone. The issue with redoing something is compatibility. We had this issue some when we shipped Windows Server 2003. We put Internet Explorer into kind of a lock-down mode by default. There were a set of compatibility issues. So, we’re trying to work and carefully gauge how do we redo, how do we manage the other issues that that’s going to cause for customers. In a long-term sense, we have some research projects, probably others do, to say, if you were starting a new operating system design in the year 2003, assuming the Internet, you would design operating systems differently. You would assume XML, you would assume the different kind of security models because the presence of ubiquitous connectivity. So there are some things that, in the 20-year time horizon, I’m sure we will redo, and perhaps others will as well.
INTERVIEWER: We talked a little bit about security, about open source. We’ve talked about the cost. We haven’t really talked about the future. Let’s talk a little bit about the future? So one of the themes of our conference here is business process fusion, and real-time enterprise, and how we see the economy starting to improve, more spending on innovation, not just on cost control. We’ve seen strategies from various centers on demand, and adaptive agility, utility, all these words popping up. What’s Microsoft’s play in this changing world of innovating, and building agility into business? How are you going to play there?
STEVE BALLMER: Well, in a sense, we’ve been hitting on this theme for many, many years. If you go back, I think
INTERVIEWER: You have great advertising, no question.
STEVE BALLMER: Well, in some senses, that’s what all these slogans are, they’re great advertisements. That’s what on demand is, that’s what adaptive enterprise is, that’s what software for the agile business is, that’s what information at your fingertips was when we rolled it out, whenever that would have been, blah-de-blah, 10-plus years ago. The truth is, everybody, every major player in the industry sees essentially the same issues, how do you give people and businesses a chance to have access to the information that they need to really drive their business in a more intelligent fashion. How do you help open up the communication channels between businesses and their customers? How do you develop those solutions at reasonable costs, and manage them in a very flexible, low cost way.
At the end of the day, that set of needs does not distinguish Microsoft from Hewlett Packard, from IBM, from Intel, from everybody in this room. Every customer expresses that same need. The real question is, how do you get after it? How do you get there? From a Microsoft perspective, we start from two heritages, and we’re trying to add a third. Our heritages are, one, with the client and the information worker, the view of information, the tools to manipulate information, to collaborate on information, the way most end users end with information is through products like Windows and Office. And how do we connect that up to really make information at your fingertips, or the agile enterprise, or the on demand enterprise, how does that really happen? Big focal point for us, and I think we are absolutely out in front on that.
The second heritage for us is software development. At the end of the day, one of the things that’s distinguished us and our platform, more people use our tools and more people use our platforms for writing applications than anything else out there. And that’s still very, very much the case today. So, when you look, how are people going to build the next generation of XML Web service applications, the kind that can be integrated, that can deliver the benefits of the real-time enterprise, the process, reengineering approach that
say the slogan?
STEVE BALLMER: The business process fusion that you talk about. If you take a look at what we’re doing with .NET inside Visual Studio, inside Windows Server, inside BizTalk, again, I think we have historic strength, we have a big customer base, and we’ve been real leaders in getting out in front.
The third area is the area of management back end, so that folks who develop and deploy these systems for their end users can manage them at reasonable cost. That has not been our area of historical strength. That’s the area where we’ve really made an effort. We’ve recruited many of the best thinkers in the management area to come join us from sort of the places in the industry where these people live. We launched our Dynamic Systems Initiative to really focus in on the next generation of manageability.
And we know that it’s kind of in the sweet spot of those three things, end user innovation, application developmental and management that we have an opportunity to really distinguish ourselves. And we feel we are way out front on two of those three areas. And, frankly, I think in the management area there’s nobody who really is even close to what our customers are expressing as their needs in that area.
INTERVIEWER: There’s a big difference, though, between Microsoft and IBM and HP and other vendors who are promoting business change using IT. And that big difference is that you deliver technology, you deliver parts, and pieces, and your partners make it all work. Do you think that’s going to work for you? Do you need to be more engaged in actual hands-on consulting, helping customers become agile enterprises?
STEVE BALLMER: We’ve got about 4,000 consultants, and I’m fully prepared to grow that number. We have a large number of partners in the world, very large number of partners, probably employing over a million people in aggregate around the world. All that is very important. But if you really think about, as we do, our fundamental value proposition, what we’re trying to do is engineer our products so that out of the box they’re simpler to deploy, simpler to build applications in, and the cost and complexity come out of the process with the software, as opposed to us having to move in, you know, 1,000 people, or one of our partners having to move in a 1,000 people, or our customers having to mobilize 1,000 people.
I think the services-driven approach to IT is not the right long-term approach. There will always be a services opportunity, but the right approach is to try to re-engineer the software and the hardware so that there is less total cost necessary between our customers and our service partners. There’s less total cost necessary, less total man-hours necessary five years from now than there is today.
INTERVIEWER: Related to the future, a lot of our clients are very interested in where Microsoft is going in the world of business applications. You’ve made some acquisitions. There’s been a lot of other acquisitions in the field. Is there anybody left that you’d like to acquire, and where are you going with this?
STEVE BALLMER: That’s such an open-ended question, Dave, I don’t know where to start. We see great opportunity in business applications, particularly in smaller and medium-sized enterprises on up to some decent size enterprises, and that’s sort of where this is always a tricky conversation. We do start with a design point that says, we need to provide the back-end software for the General Motors of the world. We said, let’s take a design point to small and mediums, and let’s be able to really scale it as far as it goes, but never lose our heritage in the simplicity that’s necessary to serve that small and medium business.
We’re happy with our market. We know there are guys who do a great job, like SAP and Siebel, partners of ours in the enterprise. We don’t think we have to be in that business. We think we’ve got to do a great job in the small and medium space. There’s no obvious thing I feel like we need to acquire. But with that said, you shouldn’t be surprised if we do an occasional acquisition to fill in some important functionality.
INTERVIEWER: Speaking of important functionality, one of these areas that people think of as important is management. And you’ve been a leader in Web services. But when it comes to Web services management, we haven’t seen very much from Microsoft there, to the point where people might think it’s an afterthought, Microsoft is once again treating management as an afterthought. When are we going to see a strategy?
STEVE BALLMER: I think it is an important area. It’s always a tricky thing to know how to weave together an approach to developing new applications and managing them. And, you’re right, the right way to do it is at the same time. If we tended to have a little bit of a bias, as I said earlier, it’s probably to get the development right first, that sometimes makes it hard to get the management right afterwards. The thing we’re really focused in on right now is to give ourselves and our customers the tools to design applications that are manageable from the get-go. And the work that we’ve done as part of the Dynamic Systems Initiative around the SDM, or the systems-design model, is all about building the capability into the Visual Studio workbench, so that a developer can describe how his or her application can be managed when they’re finally deployed. And we need to do that in the XML Web services area, and while I’m not making any specific announcement about an acquisition, I think you should stand by for news.
INTERVIEWER: Along those lines, though, with the dynamics systems initiative and SBM, we’re expecting to see improvements there. It’s an interesting strategy to go after making operationally aware applications, but it’s going to take time. There’s a culture issue getting developers to think in terms of operation. Is there something that we can expect sooner to help users with management soon? Again, in the next year, are they going to say things have improved quite a bit?
STEVE BALLMER: Within the next year I actually think we have a lot of great stuff coming. So let me kind of summarize it. The first I’ll put on people’s radar screen is the new Software Update Services, which will be available the first part of next year. The Software Update Services is a technology that is free, it sort of comes with a Windows license, if you will, and it helps do patch deployment and management inside the environment. It’s designed to address the kinds of problems people were talking about on the tape. I think it’s a very important piece of management technology. For people who want a more comprehensive management and deployment suite, we’ll have our systems management server, that’s a new release in 2003, coming to the market, and I think it’s a very good product. It’s a product that was troubled in its early releases, but I think it’s now a very good tool for management. And in the operations area, we’re bringing to market a bunch of knowledge packs, you could call them, for the Management Operations Manager, or MOM, which I think will be essential for people who want to operate Active Directory, Exchange Servers, file servers, SQL Servers. The extra knowledge about those applications that are built into the MOM knowledge pack I think will be important advances in management for people who do have Microsoft environments.
INTERVIEWER: Outside of just security, and outside of general manageability, actually the number one issue our clients talk about with servers has to do with consolidation. I deploy a new box for every application, I’ve got hundreds, I’ve got thousands of boxes, and most of them are heavily underutilized, are you going to help with that in the near future, or is this more of a long-term?
STEVE BALLMER: I’d say we have some technologies that will be available within the next year or so that are helpful, very helpful, but probably incomplete, virtual server technology, which will be made available as part of Windows, so it’s easier to consolidate applications, even if they would look otherwise inconsistent or incompatible on a single box. The automated deployment service technologies that we’ll bring to market. They’re all helpful, and yet I know that we will be pushed by our customers to make them more integrated and transparent for server consolidation scenarios.
INTERVIEWER: You talked a lot about agility, we all talk about it. How does Microsoft stay so agile? It’s such a big company, now arguably a mature company, yet you’re able to respond pretty well. How do you plan to keep that up as you get even bigger?
STEVE BALLMER: Well, we’ve done a lot of work on the company over the last few years, to try to both decentralize and recentralize a few things, so enable us to stay agile. People have written a lot about the fact that we’ve got seven business units, what’s that about? That’s about having groups that can really focus in on a given customer set and market opportunities, really hear what people are saying, and staying nimble, and staying agile. At the same time, in a sense we’ve recentralized our approach to some of the core technologies, because we know that our customers actually want better consistency and integration in what we do. If I get asked two years from now, what’s your workflow strategy, it looks different between the Office group and the BizTalk group, I’ll be chagrined, shall we say. If I have people ask me two years from now, I’m not sure which of these different things is really your data-access strategy, I will be chagrined. So agility will come both by decentralizing, the way we’re doing a lot of R & D and focus on the customers, and recentralizing some of the R & D so that we get greater consistency and integration, and coherence in some areas that our customers want to see it. And we’ve worked on our organization to permit both of those.
INTERVIEWER: But, in the area of your business, you’re a software company, you’re not going to become a services company, if you grow your consulting business it’s really only in support of helping your customers use your software. We all agree on that, right?
STEVE BALLMER: Yes, absolutely. We don’t foresee any situation over the next several years where a financial analyst will ask us, how much did services grow last quarter, because it’s fundamental to some profit model they have for the company. We will grow our services capacity, so that we can put skin in the game with our customers on their important projects. But, if we go from 4,000 to 10,000 consultants, in the grand scheme of Microsoft revenue and profit, that will still not be
that is a group of people who will be in support of our basic business of having software that helps empower our customers.
INTERVIEWER: Some might say that $50 billion, which you have on hand, might by a lot of agility.
STEVE BALLMER: Do what with our agility? Buy some agility, okay.
INTERVIEWER: We did a few more video clips, sort of the outtakes, where we asked people what did they think that you might want to do with that $50 billion. We should roll the clip.
STEVE BALLMER: We’ve got financial advisors.
INTERVIEWER: Let’s close with one last question. What do you think we’ll be talking about next year, any predictions?
STEVE BALLMER: Yes, I think this time next year there will be a big focus on a couple of things. One, I agree with Tom. If you take a look just at where we are in the cost-value pendulum with our customers, things have swung from cost of acquisition is the only thing to talk about, to total cost of ownership, and I think the dialogue a year from now will be on new scenarios. Cost of ownership won’t have disappeared, but new scenarios will really be much bigger. What we’ve spent a lot of our time today talking about is essentially things where we need to respond as an industry leader to pain points our customers are having, pain of security, pain of costs, et cetera. We’re not talking about opportunity, we’re talking about pain points.
I think a year from now we’re going to be talking about new scenarios, things that are enabled. A year from now, certainly from a Microsoft perspective, we’ll be talking a lot about not only the things people have accomplished with the new version of Office, which I think will be amazing in terms of some of the new scenarios for the enterprise, but we’ll be talking about the next, upcoming version of Windows. We have a professional developers conference next week. We’re not going to launch the next version
the next major version. We have a security-oriented release at the beginning of next year. But, the next major version is a couple of years out, and we’re going to start talking about it to software developers next year, and I think by the time we get together you’ll be ready to grill me a little bit on that 12 months from now.
INTERVIEWER: So we’ll see you next year?
STEVE BALLMER: If I’m invited back I will make every opportunity to attend.
INTERVIEWER: Great. Thank you very much.
STEVE BALLMER: Thanks, everybody.