CRN Interview: Microsoft CEO Steve Ballmer

Microsoft CEO Steve Ballmer spoke about the Vista challenge, how Microsoft has fared against Linux, how partners will play in the emerging software-as-a-service arena and how Microsoft can become what he calls the software industry's first "N-trick pony" in an interview with CRN Industry Editor Barbara Darrow and Senior Writer Paula Rooney.

CRN: First of all, there was news today about Adobe and Microsoft, and a story that Microsoft is dropping some PDF support from Office 2007 on Adobe's insistence. Can you comment?

BALLMER: I think we're talking about it pretty publicly The only thing I'd say is that PDF is a published open spec, and the second most-requested feature we got for Office is 'save to PDF'.

CRN: There's a software-as-a-service push now that Microsoft started talking about publicly last year. What is your take on where bread-and-butter Microsoft Gold partners will fit into that over time? There's always a fear of disintermediation.

BALLMER: Let's not start with the partners first. Let's start with the customers and then back into the partners. Let's talk about big enterprises, not worrying precisely how we define them. Then medium-sized customers, I'd say someone with at least three or four IT people. Then small businesses with an IT person and small businesses without an IT person.

Let's not worry about software-as-a-service yet. Let's talk about the partners who serve them. Large enterprises tend to be served by large or semi-large integrators and tend to be served by application development shops--big ones--or vertical ISVs.

Medium-sized companies with a decent-sized IT staff, that's the bread and butter [arena] of the Gold [partners]. Sometimes they'll be up in the enterprise. Then you get smaller application development partners in the fray. I'm just talking about service partners, not reseller partners.

The small business with an IT person or two is not basically served by our Gold-type partners. Some Golds go down there, but [that segment is typically served by] our certified channel and some of our registered members. For the small guy with no IT, we have a small-business specialist channel, but mostly it's Charlie the brother-in-law who takes care of the computer or maybe a small partner or maybe DIYers, the do-it-yourselfers.

Why did I go through that litany of things that you know perfectly well? Software-as-a-service has a different time frame and a different impact on each of the segments.

In big enterprises, I don't expect software-as-a-service to be a raging phenomenon [yet], except on a departmental basis, for some focused apps like Salesforce.com has been. I don't expect it to be raging phenomenon except in specified ways for a number of years yet.

NEXT: Ballmer's take on the Vista delay and why it will never happen again.

For the medium-sized businesses, I think that's where software-as-a-service will have its biggest impact. Is there a service to clean out spam or do A/V [antivirus], or does every medium-sized business have to install and operate its own spam and A/V software? You saw us buy Frontbridge, for example, in an effort to be part of that phenomenon. We'll have a reseller channel involved, and [there will still be] plenty of work for the Gold channel. Again, you'll get specific apps, spam and A/V, and maybe some other special-purpose apps where software-as-a-service will hit.

Then you get down to the bottom two [segments]. It's not that software-as-a-service won't affect the top two. It's just going to be a few years, and we're going to have plenty of time to work on the model. I think software-as-a-service will be more interesting in the bottom two [segments], probably mostly in the bottom one [where there is ] no IT person or maybe a small company with one IT person.

Take a look at Office Live. It'll have to be customized, custom SharePoint templates. Even in our beta, we've been working through how to do that. We've got four partners now.

When we get the model down, one of our big attractions should be we'll have a lot of partners doing both vertical apps and customized apps that work with Office Live. I see a big opportunity in there for our partner community. We don't have anything to talk about at this stage. I'm sure we'll want partners to sign up new customers, and there will have to be a business relationship that facilitates that.

In those two parts of the market, typically a lot of Gold guys aren't playing, although registered members and small-business specialists do play. I think we'll have good, complementary offerings between Small Business Server, an active channel product in Office Live. What we're trying to do is have some things on Office Live and some things hosted on their own premises. [So at] every level--integrator, developer, reseller--there'll be opportunities.

CRN: Is the growing relationship between Microsoft and Best Buy going to hurt your more traditional partners?

BALLMER: No. Customers don't pick people because of who partners with who. Customers pick the service partners who do a good job for them. That's my fundamental thesis. Today, one of the largest tech service forces going is the Best Buy Geek Squad, an organization that required people to get certified. They don't do the same things, frankly, as the bulk of our solution provider channel. They tend to focus on the consumer channel, some on business. I don't think we want to deny the world of us helping get their people certified and ready. I think it's good and exciting. I think it's good for the customers. They are another channel partner. Our job is to welcome all channel partners and offer opportunities and training for all partners. This is another new entrant we're excited about.

CRN: Let me switch gears to technology. Microsoft is no longer the upstart challenger. People now see you more as an IBM. You have legacy systems and customers. When you talk to people about where innovation is coming from, most people don't point to Microsoft but to Linux, Google and Apple. You guys talk about innovation, but these products are stuck in the production line.

BALLMER: Look, we have one product--it just happens to be our most famous product--that has a bigger gap than it will ever, ever, ever have again in its release cycle. That will never happen again. I know how we got there. I'm not going to go through all that. I know what we're going to do differently. I'm not going to go through that. It'll never happen again.

Other than that one product, there's Xbox. A lot of people give us a lot of credit for innovation in Xbox. There's Windows Mobile, people give us a lot of credit in that. We've shown Office 2007 at [Wall Street Journal's Walt] Mossberg's D conference; it's the only thing that's ever gotten applause at the D conference.

We have sort of three issues swirling all at once. One issue is this big gap on Windows. Even look at [Windows] Media Center; people think it's pretty interesting. The gap is a big issue on our flagship product, that's No. 1. No. 2, because we are 'big' and we've had the PR around the regulatory issues we've had, people want to analogize. You can't stay the cute little darling your whole life. You want to be the leader your whole life, you want to be the innovator your whole life. But any company that thinks it's going to stay in its cute phase it's whole life is probably naive.

And the third issue is there are one or two things where we've got to get going, namely music and search. On that I hear you. Take a look at all the good stuff we're doing--Xbox, Windows Mobile, Office 2007, the Atlas stuff we're doing for AJAX development, the stuff we're doing to ease database development. People are super-excited.

Amongst the developer community, [there have been] a lot of improvements in how you can write apps. I can go on and on. There's more to Office 2007 than the end-user demos we do. We get more and more capabilities for business. I absolutely defend our record.

Now we've got to get Vista out--quality, but out. And we've got to get back into our pattern. We've put the master in there now: Steven Sinofsky. And believe me, we're going to get that thing into a regular pattern and drive hard. Our search has gone form nowhere to being pretty good. We're going to put pedal to the metal, get past the competition in that area. I would say there's some opportunities here to get out with new innovations, get Vista in the market

I'm not unaware of what you're saying, and I feel very good about our pipeline of innovation.

NEXT: Ballmer's view on partner opportunity.

CRN: In terms of the timeline for Vista, you were quoted by the foreign press as allowing that it might slip into February. What are the factors that could lead to that possibility?

BALLMER: All I said anywhere is quality, quality, quality, quality, quality. The betas are just out: Quality, quality. I get an e-mail from a customer who says, 'I'm worried about the following problem with the beta.' That's what betas are about. I say, 'Don't worry. Quality, quality. We're just working on quality.'

We will ship quality, security, quality. The features set is all there. Now it's all about performance, quality, quality. If I get an e-mail, 'Should I worry about what you're going to ship if you're forced to ship on blah blah blah?', I say, 'Quality.' That's essentially all I said in Asia.

CRN: Is it possible that you'd give existing customers that have Software Assurance a pass to mollify them?

BALLMER: We're going to ship quality. We're not going to ship poor quality and a patch. We're going to ship quality.

CRN: Microsoft generates a ton of cash, but the opportunity for software partners is less clear. When you look at where venture capital is going into new things, even Paul Maritz, a former senior Microsoft executive, is with a startup that is Linux-focused. They may not even do a Windows product. If even former execs don't see an opportunity doing things on Windows, what does that say?

BALLMER: I don't know what Paul is up to, and I don't really care. He's been gone for many, many years at this point. I don't know what that question is about.

CRN: What the question is about is that venture capital investors see more opportunity in the non-Microsoft stack than they do in the Microsoft stack.

BALLMER: Oh, the question is about Linux. Let's talk about it. Four years ago, people wrote that we'd be wiped out by Linux. Four years later, how are they doing?

CRN: They're doing well on the server side.

BALLMER: No they're not. They're not gaining share. Four years ago, we were supposed to be wiped out. Maybe we've both gained share. We've not lost share to them--maybe we're down a point--and that can almost be all accounted for by the number of Linux servers Google's put in, for gosh sake. It was supposed to get hot and it hasn't. I'm not saying it can't. I'm not trying to sound over the top or arrogant. What I am saying is there's no new news here, except that what people predicted hasn't happened. This is the year where we're well-equipped to come back into the Linux strongholds and take some share.

We have our Windows Compute Cluster Edition. We'll get back into high-performance computing, which, at the end of the day, is something like 30 percent Linux servers. We've got the new Windows Server with fantastic improvements in Web serving, along with the Atlas technology for AJAX development. I think we're in a place where we can grow share not relative to Linux but absolute.

We're at a point in time where we have technologies that will let us take some of their users. If the question is about Linux, we're in there every day innovating, every day competing. We've got great stuff. There can always be another startup that wants to mess around with Linux. But Linux has traction only in certain workloads today. And if we do our job right it won't get traction anywhere else, and we will take traction in those areas.

CRN: What kind of things are you seeing from Vista beta 2?

BALLMER: I'll tell you what I told analysts this week. It is early. You'll always get stories about people who have problems installing and stories about people who love the UI. I've got them both. We're in tens and tens of thousands of sites. One story here, one story there is insignificant.

CRN: Microsoft is in diverse markets ranging from the consumer arena all the way up to the enterprise. How do you prioritize that? I'm not sure any other company plays in so many markets and competes on so many fronts.

BALLMER: That's why we have a number of businesses and technology leaders, and each focuses on different areas of innovation and different competitors.

NEXT: Microsoft looks to be an "N-trick pony."

CRN: But broadly, who do you worry about most?

BALLMER: You're not asking that and I don't want to answer that. I want to answer this question. The question is, 'How do we do it all?' The answer is we have capable leaders that can actually run--I won't say independent businesses--but run them in a way that is sufficiently focused and protected from a resource and economy perspective to get something done.

Robbie Bach and J Allard are not coming to work every day worried about the same problems that Bob Muglia is. Steven Sinofsky and Kevin Johnson have different concerns, [as does] Jeff Raikes. And thank goodness. These are some strong leaders. [Editor's note: Bach is president of Microsoft's Entertainment and Devices Division. Allard is corporate vice president and chief XNA architect. Muglia is senior vice president of servers and tools. Sinofsky is senior vice president of Windows and Windows Live Engineering. Johnson is co-president of the platforms and services division. Raikes is president of Microsoft's Business Division.]

We do get a chance to get some technology synergy, brand synergy, sales force synergy, which is all good. When we go build something, I expect there to be a team of people, very capable, as capable as anybody in the world, at building great stuff and competing. If we can't do that, we have an issue.

What's key to me, the most important thing that keeps me up at night, is having the talent and technology approach and management approach and innovation approach that lets us be what I might best refer to as the only N-trick pony in the technology business.

You might say, 'What is an N-trick pony?' Most software companies do one thing. We've proven we can do at least two, maybe three. We do the desktop, we do the server. I think we've almost proven--we haven't proven we can make a lot of money [on it] yet--that we can really do gaming. Sony is helping us in that with some of their announcements. I want us to prove we can do multiple things really, really well. The innovation around that, how does it flow? How does talent flow? The leadership, that's kind of job one. If I think of where I apply most of my time and energy, it's probably there.

At the top of the business, what we think a lot about is new business models. How do we embrace them or compete with them? Open source is more of a business model than anything else. Over the last several years, we can't embrace it being a for-profit institution, but we've spent a lot of time thinking about how we can compete and focus on innovation, value and total cost of ownership. We've done a very good job competing.

If you look at other business model ships, in the consumer market, advertising revenue can be an important source of funding. The notion of a technology ad platform, we probably could have grokked earlier in our lives. We totally grok it now. How do we embrace it where appropriate, make sure we get critical mass, which we're doing, and succeed with an advertising platform, including the applications like search and others that would be built around it?

What keeps me up at night? It's embracing or competing with new business models, the talent and approach and innovation that lets us be a multi-trick pony.

CRN: Getting back to the partner proposition, the Adobe thing illustrates this. Software companies partner with Microsoft with a certain amount of trepidation. Does that hurt you in public opinion and in your ability to deliver products?

BALLMER: No. We're coming further up the stack in the Office 2007 time frame. I hope that's also clear. We have document management and workflow, which we've never had before. Enterprise search we're taking to the next level. Business intelligence we're pushing forward in a fairly significant way. Unified communications, including voice, will be very significant for us in the Office 2007 time frame, but we are pushing further up the stack. We are a platform company in the sense that all our stuff should be extensible in that third parties can add value. We're not a platform company in the sense that we don't think about how to complete the loop with some sort of end-to-end solution. We've always been an applications company. In the early days of Windows, we weren't a platform company. We were a platform company with key applications that work with our platform.

CRN: Even Microsoft Business Solutions is pushing its apps as a platform.

BALLMER: It is. It is a platform and applications. Most good pieces of software are both.

CRN: VMware is getting ready to launch its next-gen platform, and they're competing against Microsoft and Linux rivals. All are integrating Hypervisor with the operating system. VMware said there's no advantage because it's integrated with the metal itself, and there's no technical benefit to the customer. What's you're take on that, and how do you see virtualization shaping up as a system service in the next five years?

BALLMER: Virtualization will be a very important system service, no doubt about that. In terms of the detailed technology, I don't particularly have a point of view on it.

CRN: What would you tell a small software company to convince them to work with Microsoft vs. a non-Microsoft stack?

BALLMER: It probably depends on what they're involved with. At the end of the day, we've consistently proven in most areas that we actually get more customers. If you're looking to add value, it is probably as important as anything else as to who creates the most opportunity for you.

Let me talk about a partner I talked to today that didn't exist nine years ago and focuses on just Microsoft technology and is now [taking in] over $2 billion nine years later. HTC, in Taiwan, builds Windows Mobile devices. Has it been good for them? You bet it has. Fantastic company. Doing some of the most exciting work going on in the mobile industry. Is it good to bet with us to get customers and make money? Most people can. The technology is good, and innovation is strong even though Vista is late. We don't like this gap. It won't happen again. Let's get on with it, man.