Remarks by Microsoft CEO Steve Ballmer
CIGREF - Club Informatique des Grandes Entreprises Françaises
Elysees Palace 20r Quentin Bauchart Paris 8, France
Oct. 2, 2008
STEVE BALLMER: Thanks. It is a real privilege to have a chance to be here with you today. I want to thank CIGREF for co-organizing this meeting with us. It's a real honor to have this opportunity.
I thought what I would try to do is in 20 minutes lightly run through some of the biggest topics and trends and themes that we see in the industry, and that we see at Microsoft, that I think might be of interest to you, and then we'll go wherever you want to go during our question and answer and discussion session.
I will say in advance I'm sure we won't get to something that's on somebody's mind, so I give you my e-mail address: Steveb@microsoft.com. Feel free to send follow-up mail if there's something on your mind, either a question or, of course, any problem; you can feel free.
I thought -- yeah, Vincent also says thank you, but I'll give his e-mail address later. (Laughter.)
Maybe I'll spend just a minute on something that I think is on people's minds that's not sort of normal in my discussion, and for which I have no prepared PowerPoint, but this is kind of the economic situation.
Certainly in the United States we see the potential for major economic disruption. I was in the UK yesterday, and I felt the same sort of pressure. I was in Norway and Denmark earlier in the week. The pressure felt a little less. I talked to Eric Boustouller who runs our business here in France. He said, oh yeah, we feel a little bit of that concern also here in France.
I think it is important to remark. I think we will see more over the next few weeks exactly what happens with the central banks around the world, with the congress in the United States. I suspect that all of this credit crisis will have some impact on business, and some impact on the technology business. Since technology represents about 50 percent of capital spending in Western Europe and in the United States, it would be inevitable. But hopefully with the appropriate action by the banking community, the national banks, the central banks around the world, we'll see this as a small dip as opposed to something more severe.
But it does remind us and it does bring back the context that whenever we talk about information technology, we've got to have two themes in our minds: doing more with less, and I would say the pendulum swings between more emphasis on “with less” and more emphasis on “doing more,” and yet as CIOs I'm sure every day you feel both of those pressures as businesspeople want to change business process, get better insight, and as people are constantly putting pressure and trying to tighten IT budgets. So, I do talk to you from within that context of doing more, if you will, with less.
I think in that frame the first thing I want to touch on is my long term outlook for technology.
If you asked the question, will the next 10 years -- what will they be like, a lot of innovation, a lot of change, quick pace, or will things in a sense slow down, it's amazing how quickly technology has progressed over the last 10 years. Ten years ago, just to give you a couple things to think about, statistically most people didn't have a mobile phone 10 years ago. That's unbelievable if you think about it today. Ten years ago, most people didn't have a PC. Most people didn't know what the Internet was, and broadband versus narrowband was a future question that only the high priests of technology were asking.
And yet I am very optimistic that the pace of innovation will proceed just as it has the last 10 years. We'll continue to get more powerful computers, more storage, better bandwidth, more disconnected wireless operation.
Moore's Law isn't going to work the next 10 years exactly the way it worked the last 10. What Moore actually said was, every 18 months we can put twice as many transistors on a chip at the same price. That stays true, but for all this time -- all this time, that additional power went into making computers go faster and faster and faster. We're not going to get faster processors anymore. Why not? Because at the current speeds, there is no physical material known that can cool them down.
So, what is Intel doing now? Giving us more microprocessors: one core, two cores, four cores, eight cores.
Unfortunately, most of the software we've written and you've written isn't designed to work that way. So, there's going to be a lot of innovation just to keep up with the advances in hardware.
Screens: This today is a physical screen. Ten years from now this will be a touch screen. This will be connected to the Internet. We'll be able to draw any of the world's content.
I look out in the audience today. Most of you have either a pencil and a piece of paper or nothing. Just to show that I'm in the same boat, that's a piece of paper. Ten years from now, we will actually have digital screens that are this thin, this light, this flexible. Screen technology will continue to advance.
All of the world's content, transactions, and information and communication will move to travel over IP networks. That's a big deal. Most of what we read today we still read in paper. Most interactions with television and advertising are still, if you will, offline, it's not online. That all changes over the course of the next 10 years.
Just one other of my favorite in business. If I asked CEOs what's their number one frustration with it, they always give you the same answer. I spend a lot of money, but when I have a question that I want answered, I often can't just go to the computer and find it.
The ability in businesses not just to find a number, to find a customer record, to do some analysis, there's still much innovation that needs to happen in a lot of different dimensions.
One of the big areas of innovation will be the way the software itself gets transformed in an environment where we can assume high-speed Internet. We call that the software plus services revolution. Some people like to use the word "software as a service." We say software plus service. Some people say cloud computing, some people say grid computing. Some people say on-demand, blah, blah, blah, blah, blah, blah.
What everybody is talking about is fundamentally the same thing. It's a rewriting of the way information technology works so that instead of having software that is handed to a user and the user takes care of it, the software instead is managed and controlled and lives in the Internet to some degree, in corporate servers to some degree, and it lives intelligently in phones and PCs.
It's not about thin client computing; it's about distributed computing where all devices are used intelligently, all devices run code, but all devices are managed centrally; the software manages itself so the users and the IT managers of the world are not doing nearly as much work as they do today in the management and administration of software.
Microsoft as a company is trying to play in that world by building seamless experiences that leverage that power: the magic of software, the power of the Internet, a world of devices: PCs, TVs, phones, servers in the enterprise, services in the Internet cloud, and software that distributes itself and manages itself.
We're already seeing some of that in the consumer world, but those same redirections, the same fundamental innovation will also be applied to the enterprise market.
Our company is pretty unique. We're the only company that's trying to work across all of these spectrums. That will either be a great advantage for us or not. At the end of the day I think it's a great advantage, because much of what's going to happen in the enterprise is going to start in the consumer market, and really being able to participate and help shape and drive and add value I think gets us smarter and better.
In our quest as the David to Google's Goliath in Internet search, we got very interested in enterprise search, and we've built out phenomenal capabilities in that area, including the acquisition earlier this year of a company called Fast that's got some of the most amazing enterprise search capability on the planet.
When I talk about this transformation -- and I'd say everybody agrees on it. You go to IBM, they'll talk about on-demand computing. They basically believe in software plus services. You go to Salesforce.com, they talk about SaaS. You go to Amazon, even Amazon has provided these services to the market at large.
The only person who says they don't believe in this phenomenon right now seems to be Larry Ellison, which means he must really believe, too, because everybody knows he likes to be contrary publicly. And I mean that very nicely.
When we talk about software plus services at Microsoft, we think it means not just how do we write things for the Internet, it really means a remaking of a number of things.
We'll need a new operating system. Just as we have an operating system for the PC, for the phone, and for the server, we need a new operating system that runs in the Internet. I bet we'll call it Windows something. We're going to announce it in four weeks. We might even have a trademark by then. So, for today I'll call it Windows Cloud. And Windows Cloud will be a place where you can run arbitrary applications up in the Internet that runs .NET.
We need to remake our development platform and our management and deployment tools so that the software really can move around to the PC, across servers magically. That involves taking .NET, our programming surface, to the next level.
Part of that means putting .NET in the browser, which we've done with our Silverlight technology. And yet I don't think the whole world lives in a browser. PC applications have better user interface, and you can integrate them more. Browser applications run on non-Windows machines, and they're easier to manage. We need to bring the benefits of both of those things together on Windows, and through our Silverlight technology permit the targeting of other systems.
One of the things that happens when you start thinking of the Internet as a platform is you really need to say what are the new capabilities that people expect in modern applications.
The young people you hire today, they grow up on MySpace, Facebook, and instant messaging. They grow up with a fundamental notion that applications have knowledge of other people. In order for business applications to go that direction, we need to provide fundamental platform operating system services that really provide what I might call the social web or the social graph.
You could say we have all those technologies today implicitly in things like Active Directory, but really turning them into the applications and platforms so that instead of you having corporate Facebook -- well, instead of having Facebook, you have corporate Facebook: my colleagues, my colleagues of colleagues, we form groups of colleagues who are interested in a given topic. These concepts that are being pioneered in the consumer side also have great purpose in business.
And, of course, with the move to software plus services, there will be new business models: advertising, subscriptions, and others.
Some of you may say, oh, does that mean all the software that we buy for our enterprises are now going to be advertising funded; that means you don't pay anything. I don't know that most of you will prefer that. At the end of the day if software is going to be advertising funded, it means whoever provides it to you needs to know what your users are doing, what they're typing, what they're reading, the Web sites they're visiting, and I think most corporations will say I'd prefer to pay a little money and retain corporate privacy.
And I'm not being silly here. Even in the consumer market I think we'll have some consumers who will opt to pay as opposed to be well targeted, if you will, by ads.
We are going to build on the present as we get there. This is the direction; the world is going to move to this world. And yet I don't think anytime soon you're going to shut down your datacenters and move everything to the cloud.
We will tell you about it, we'll encourage you, but at the same time our strategy is to build a symmetric stack: Windows Server, Windows Cloud; Active Directory, Active Directory with Live IDs as an identity system in the Internet; SQL Server and an implementation of SQL Server for the cloud. We now have CRM; CRM in the cloud. We have SharePoint; SharePoint in the cloud. We've got Exchange; Exchange in the cloud.
And we have big customers already moving this direction. People like Nokia, Coca-Cola enterprises and others have signed up to move to Exchange online, SharePoint online. It's not going to be for every one of you. We are glad to work with you in either dimension. But our goal is to provide you with a symmetric platform. No matter how you choose to deploy, you can do that very, very seamlessly.
I'm going to turn now from technology trends to more short term things that I know are on your mind. Windows Vista is on my mind. I hope actually it's on your minds.
Windows Vista is a product where we made some very conscious choices for some very good reasons that have been very painful. We decided to change things in a way that broke applications and device drivers, but enhanced security.
We're here now about 18 months after Windows Vista shipped. We've shipped about 180 million units. We've had a chance to get market feedback and issue the first service pack. We've had a chance to work with the people who write the device drivers and applications. Those have mostly now been made compatible. You have your own corporate applications. We've had a chance to work with you on that for a while. Deployments in large corporations are now ramping up quite nicely across the world, but in the enterprise I would say we are still earlier.
The other thing I would tell you is that Windows 7, the next version of Windows that we're working on, will be compatible with Vista -- no more breaks. So, any work we're doing together with you or you're doing on your own to test your applications for Vista compatibility will also apply to Windows 7. We hope you choose to deploy with Vista, but all of that work is good, important work for the long term.
And, as we targeted, Vista has prove statistically to be the most secure version of the operating system that we've ever put in the market.
This is a big year for us in terms of new product releases. We have new technologies, exciting new technologies I think for our enterprise customers coming in a number of areas. Our unified communication products, particularly our Office Communications Server, will be dramatically enhanced: videoconferencing, audio conferencing. It's really exciting stuff I'm sure our people will talk to you about.
Enterprise search: Most people will never really get acquainted in a business with all of your applications. You want to be able to search from within a browser all the way back in to your SAP system, your CRM system. Our enterprise search technologies give you those capabilities.
The so-called cousin to enterprise search is business intelligence, how after you find data do you manipulate it, do you present it, do you view it, in order to get insight to make decisions.
We've extended the business intelligence capabilities in Excel, in SQL Server. We recently acquired a new company with very high-end analytic services called DATAllegro, and we'll be productizing their technology, and we have a whole new line of business intelligence products for scorecarding, budgeting and reporting that we call PerformancePoint.
Virtualization: We shipped our Hyper-V and System Center products. VMware, they're out there, they're the market leader, and yet only 4 percent of servers are virtualized. Our design point has been to provide great virtualization technology, but with an integration, low cost and an ease of use that lets you get to the next 30, 40, 50 percent of servers and virtualize them. And we recognize the leadership of VMware, so we also provide tools that will help you manage VMware virtual machines.
I talked about the cloud and social computing, and of course there's always more coming in mobile, our Windows Mobile 6 operating system, next year 6.5, and as a platform for business applications and business use it's unparalleled really.
So, I'm excited, do more with less. Things like cloud computing and virtualization hopefully let you both do more and with less. Things like unified communications and business intelligence enables you to equip your users. I think our company has a breadth of innovation that we bring you with better and better ROI in terms of value compared to cost. And, of course, we're trying to provide the choice and flexibility that software plus services as well as enterprise computing solutions permit.
So, with that, I want to end my prepared remarks and I guess Vincent, Didier and I will have a chance to do the Q&A session together, but it's been my great pleasure and privilege to talk to you today. Merci beaucoups. (Applause.)
MODERATOR: Thank you. It's open, the discussion, Steve. So, you clarified our strategy and the software plus services, and you defined the directions where we are going. But just to help our customers set their expectation, could you please precise the list of services and functionalities, capabilities which are already available and which are currently run by these large customers you mentioned?
STEVE BALLMER: Yeah. In the market today we provide our SharePoint Online service. I won't say broadly available. We're trying to take customers at the right pace. If you're interested, of course our team is happy to talk to you, but it's not sort of self-serve signup for the average midsized company yet. But for people in this room I encourage you to talk to Vincent.
Exchange Online, SharePoint Online are both available. Office Communications Server Online should be available in the next several months.
The Windows cloud is not yet available, but we'll be announcing its beta availability soon.
Some of the social computing technologies are available for trial today under the Live Mesh name on our Web sites, but they're in beta form initially for the consumer market.
The SQL Server data services are currently available, but will be quite a bit enhanced, and Microsoft CRM Online is available.
MODERATOR: Okay, thank you, Steve. First, I want to express participation too - I've seen your vision for the future, okay? You know that we at Essilor we are always interested with visions. That's something, okay, we do not do very well, sharing this.
Now I am going to ask you some questions coming from the CIGREF members. The first one is maybe not very interesting, but crucial for us. It's about your licensing policy. Because, okay, if I share your vision, okay, there will be more and more -- more and more, okay, Microsoft products, Microsoft software in our companies. It happens that licensing seems to be more and more complicated.
So, do you have plans to I would say get easy to do business with you in the future?
STEVE BALLMER: Well, there are -- it's like a test: There's only one right answer. Yes, we want to make things simpler and easier to do business with.
What it means to be simpler is actually complex. It turns out that what we have learned over time is every time we change our licensing, we remove options, we do anything of that nature that's to simplify, what it generally winds up doing is making things more complicated for the customer who used that option or more expensive for somebody who had figured out how to use that option to save money.
So, I am -- what shall I say? I'm very hesitant to withdraw options, even though that may make things simpler.
What we are trying to do, unfortunately perhaps, is to create new options, simpler options for people to take advantage of.
I think in the datacenter there will be new options from us to make it easy to license our line of business application platform. It's a big story in Microsoft. When I was last in Paris, Didier explained to me one of his vendors, they actually charge, Essilor, by the contact lens, if I remember correctly. I tell this story all the time at Microsoft. It doesn't work for somebody whose business is not contact lenses, but it reminds us we need to come up with sensible programs that can then be adapted to individual customers.
The thing that I think is most important is we're trying to get what I would call the desktop infrastructure to be more simply licensed somehow by the number of employees or the number of PCs. Certainly in a software plus services world where we're running the infrastructure we must do that.
So, it's very much on my mind, and yet we'll be somewhat conservative, because if we're not conservative we actually may do more harm than good through simplification.
MODERATOR: Thank you, Steve. I remind you that if you want some customer tests or customer advisory boards, CIGREF is here to just test some ideas from you.
The second question is a quite very personal one, because there are a lot of CIOs in the room who budget time, and a lot of us we have to choose between going to Vista next year, waiting for Windows 7. So, could you, in your view, okay, what could be the pros and cons to go to Vista first or to wait for Windows 7? Do you have this kind of experience?
STEVE BALLMER: I'll give a very frank answer. I'm not sure whether it will encourage or -- some of you may move faster and some may move slower, but let me give you an honest answer.
Everybody has kind of a cycle of what you're trying to do in your own organization. Some people will re-buy their computers, a third or 25 percent a year, so they replace all their computers in four years. If you're on that strategy, my recommendation would be get all your applications tested with Vista and then move to Vista as soon as possible. Don't do a full replacement, but since you're buying new computers, why buy new computers with XP instead of Vista on them.
Some people plan big projects and they don't touch their computers, they don't buy new computers, except in big what I would call one-shot purchases. The upgrade of the branch banking network at -- I was going to say at Fortis, but that's probably not a good one today, but the upgrade of the branch banking network at some company, Credit Lyonnaise or someplace, if you're going to do a project like that, then I think you have to say to yourself, when am I really going to start the project? If I'm really going to be halfway through the project when Windows 7 comes out, you're going to have to decide, maybe it makes sense to wait, put Windows 7 only on that project. That's a different model. So, I think it will depend.
The one thing I am not saying to people they should do is run out and upgrade all of their old PCs today, because I think there's more of a natural flow. If you're prepared to do it, you should do it -- I would say you should get ready to do it with XP -- or sorry, with Vista, but then depending on the timing you might wait for Windows 7.
So, my point isn't to encourage you to do it immediately. Of course, we'd love you to do it immediately. I don't want to get in trouble with Vincent. My real advice is to do it in the natural rhythm of your PC upgrade cycle, and our team would be glad to give advice in that context.
Most of you will not upgrade the software on existing hardware. Some will. Most of you will actually choose to buy new machines when you move forward, and so we should work with you in that context.
MODERATOR: Okay. So, the next question is, today are you ready to put your head on the table about the date for Windows 7 or is it too early?
STEVE BALLMER: Too early. (Laughter.) I've had my head on too many tables too many times. I'm holding it up right now.
MODERATOR: You just raise your head, your hair, not your head. (Laughter.)
Okay, now a subject, okay, I don't want to be rude, but we are maybe going to spend five minutes about your competitors. So, let's phrase the question that way. We understand that for years as far as your fee systems are concerned you had very, very fierce competitors. We are for the free market. We don't like monopolies. So, today it appears that maybe you will have a competitor named Google, and it reminds me of something. It reminds me between Netscape and Microsoft, okay, as your leaders in '95, '97, because Microsoft has a different business model for Web browsers, and it was the end of Netscape. Now you have a big company beginning by a G, and it happens to be a new business model, competing different from yours, okay. Do you see it as a competitor and in which applications or which functions?
STEVE BALLMER: Of course Google is a competitor to us and we are a competitor to Google. There are businesses like Office productivity where we are very strong; they're going to try to compete. There are businesses like search where they're very strong, and we're going to compete.
Competition is a good thing, you hear me say the same thing, and the truth is we've had a lot of good competition. We've had StarOffice, we've had Open Office, we've had various -- well, if you ask me today, I'm not recommending Open Office, but Open Office is a stronger product than the Google word processor and the Google spreadsheet.
The truth of the matter is Google charges. It's not a different business model. Both Microsoft and Google charge business customers for our software.
I think we offer much, much, much, much stronger functionality, we offer much stronger control to the CIO, and we offer an equivalently good economic value.
I think they have a long way to go. Frankly, I don't even think they're today our strongest competitor. That doesn't mean they won't get there. I'm talking about now in Office productivity.
At the end of the day they only offer a hosted model. Their stuff has never been tested in terms of directory, privacy, security.
So, if you're asking me to be competitive, I could go for an hour. I think it's very different, but they're attacking. And we're not standing still, we're moving forward, and we're attacking, and that's good business. That's good for you, whether you choose over time to stay with us, which I hope you will, or not. Everybody is moving, moving, moving faster for the experience.
But I do have to remind you we have had competition. We've had Open Office as a competitor. Open Office is free. Open Office has not been -- it's not as powerful as Microsoft Office, but it's much more powerful than the Google products today.
I encourage you to test that if you don't believe me, and at the end I encourage you to return to Microsoft Office.
MODERATOR: It was my next question, Steve, because we all know that a big French company decided to sign a contract with Google for office, and we all know -- maybe not, but some know that you were not very happy with that. So, could you comment? Do you see it as an error for the customer or as a threat to Microsoft?
STEVE BALLMER: I don't like losing customers ever. I don't like losing customers in the U.S., I don't like losing customers in France, I don't like losing customers in Angola, I don't like losing customers anywhere. So, when we lose a customer, I like to understand the why and the how and make sure it never happens again.
I don't always agree that the customer made a good decision for the customer. Sometimes I do. But I don't always agree. I'll visit the customer, I'll try to understand the issue. And frankly if we feel the customer is fully deployed, I will never let it -- I don't let myself believe the customer has been lost.
Even today the city of Munich was a famous customer for Linux on the desktop. I still don't think it's lost since most people in Munich still actually use Windows instead of Linux.
So, we have to keep in there with the customer over time, and Google has very, very, very few customers for Google Docs and Spreadsheets. And I will say, yes, I have paid attention to all two of them, and I have visited them both.
MODERATOR: So, now always --
STEVE BALLMER: I don't know, I'm trying to not be too competitive, this is the new Microsoft, but you're really getting me going today. (Laughter.)
MODERATOR: Okay. You will come back next year, you will.
STEVE BALLMER: Absolutely. For the CIGREF anytime.
MODERATOR: So, maybe a personalized question. If you look at promises for Google, say, okay, get rid of all the hassle with workstation, Office, everything is hosted by Google, this is a fixed price each and every year, it seems to be very appealing for CIOs because we have such many issues to drive. So, saying, okay, this is not an issue for CIOs anymore, this is outsource on the cloud in the Google.
So, the question for me is, are we going to try this or is Microsoft in a position to say, I will offer you exactly the same kind of service business with all the Office suite at, I don't know, okay, generally the first of each year? Are you ready to --
STEVE BALLMER: We're offering that. The customers I mentioned are buying that service from us today. If you're Nokia or Coca-Cola enterprises, we'll offer you Exchange, we'll offer you SharePoint, we'll offer you distribution and management of the Office software --
MODERATOR: But not the same?
STEVE BALLMER: Sure.
MODERATOR: Excel, Word, everything?
STEVE BALLMER: But instead of offering you a toy word processor and a toy spreadsheet that doesn't read all of the existing documents, we'll actually offer you Word and Excel and PowerPoint and Outlook, the stuff most of your users really want.
MODERATOR: So, it's ready today? We can go today or tomorrow?
STEVE BALLMER: Today, tomorrow, yes. Vincent is ready to help you. (Laughter.)
MODERATOR: Thank you.
STEVE BALLMER: The only thing we don't really have in the portfolio today that I'd like us to have over time is the ability for us to provide more complete management of the non-Office applications on your desktop. That is not an offering that we make today. But we do make an offering that includes the management and distribution of Word and Excel. In fact, our lead customer on that is this company Energizer. They make the batteries. They have the ad with the rabbit running around. I bet it runs around in France, too. It's a global symbol. But we take all desktop deployment of Office in their company, and they're always on the most recent versions. They're moving forward. They're very happy.
MODERATOR: Good. I will ask the last question, because I know your agenda is very tight.
STEVE BALLMER: Sorry.
MODERATOR: From a personal viewpoint, what is your view about the economic crisis, finance crisis? Are you still optimistic? Are you going to retry something in Alaska or Nebraska?
STEVE BALLMER: Well, I'm as optimistic as ever about the technology. I do think we could be -- as a global economic society we could be in for a tough time. It could be a very tough time or a tough time. Hopefully we'll see good wisdom out of government worldwide. I'm optimistic about that, because I'm an optimistic person.
But at the end of the day I don't think anybody will be immune. The world's banking system is too interconnected, the world's economy is too interconnected. I think these issues will be worse in the United States than they are in most other countries. But at the end of the day we will get through it, the world will get through it, the economy will get through it, the technology industry will get through it, and I view it, no matter what happens, our company is going to continue to innovate in the important technologies of the future.
MODERATOR: Good. We are on time, so Steve once more thank you. It's always a pleasure to have you with us in Paris. I was sure that you speak and understand a kind of French. So, could you say something for the final one?
STEVE BALLMER: Oh, I hate speaking. I like listening and having nobody know I understand. (Laughter.) [In French].
MODERATOR: Thank you. (Applause.)