Steve Ballmer: Web 2.0
Oct. 22, 2007
A transcript of remarks by Microsoft CEO Steve Ballmer at Web 2.0.

Remarks by Steve Ballmer, CEO, Microsoft Corporation
Web 2.0
San Francisco, Calif.
Oct. 18, 2007

JOHN BATELLE: All right, really? Okay. Well, I'm just flummoxed. This is like the second time I've been flummoxed on stage. The first time was when Zuckerberg told me the financing was going well. (Laughter.) All right, so we're not running a -- can I get a -- do I have Steve's -- there he is. Come on up, Steve, we'll just start it live.
(Applause.)

JOHN BATELLE: What happened there?

STEVE BALLMER: The guys de-cued it.

(Break for direction.)

JOHN BATELLE: So, Steve, now that we've loosened up the crowd with that non-video -- (laughter) -- I want to ask you the same question I asked Mark Zuckerberg. How's the financing going? (Laughter.)

STEVE BALLMER: As Mark said, it's going pretty well. So, I guess that's the answer.

JOHN BATELLE: Are you involved?

STEVE BALLMER: We've got a great partnership with Facebook on the advertising side; very, very important to us. We loved the partnership and, you know, we'll see where it all takes us.

JOHN BATELLE: So, you're not involved? (Laughter.)

STEVE BALLMER: We have a great partnership with Facebook. We'll see where it takes us.

JOHN BATELLE: All right then.

Okay, I want to sort of ask a pull-back question, a question sort of -- Microsoft is a vast company. How many people work at Microsoft?

STEVE BALLMER: Eighty thousand.

JOHN BATELLE: Eighty thousand, and every one of them is your personal friend, right?

STEVE BALLMER: Absolutely.

JOHN BATELLE: Absolutely. When you look at all the holdings of Microsoft, what are the -- what of the company are you happy about, what do you feel really good about, and what are you unhappy about and feel like it could use some improvement?

STEVE BALLMER: Well, at the end of the day, you know, I'd probably tell you I'm happy with everything, and I'd probably tell you everything needs a little bit of improvement. You know, it's sort of like asking somebody, hey, you know, what do you think of your kids? How are your kids? And you get up in front of 1,200 people and say, well, I really like Sammy but I really am upset with Aaron these days. It's just, you know, duets. I mean, everything we've got going on, in our desktop business, with Windows and Office, you know, we've got -- we had a phenomenal launch last year of Vista and Office 2007, things are kind of roaring ahead, great opportunities, and yet we have some things we're going to work on in Vista. We've got opportunities for improvement, you know? The question is, you know, people ask, where do we go next? And, yes, we're going down the path. There will be more operating system releases and Office releases and blah, blah, blah, blah, blah.

In our enterprise business we're going gangbusters. Really, we're going gangbusters, and yet at the same time there's so much of the enterprise world we've barely scratched. In the advertising world and our online business, you could say in some senses we're more nascent but then we've got some well established properties like Hotmail and Instant Messenger, et cetera.

So -- but we've got our work cut out for us, and I'm happy with the kind of work. I love the new Live Search. We're in the game. We've come from nowhere quickly. And yet, hey, we're just, you know, we're kind of a relatively small player compared to the market leader.

JOHN BATELLE: Right.

STEVE BALLMER: And then in our kind of devices and entertainment business, what's not to like about Halo? And at the same time, you know while we'll sell maybe 20 million or so Windows Mobile devices this year, that's still just such a small fraction of the billion-plus phones that get sold.

So, in every area there's a lot of good and a lot of areas of improvement.

JOHN BATELLE: All right, all right. Let me ask you -- let me repeat to you. You mentioned that leader in search. So, let me ask you this: Is that leader in search still a one-trick pony?

STEVE BALLMER: When I use the word one trick pony, I mean it in a very specific way, and by that definition most companies are, and certainly most technology companies. I think one of the things that's true about most technology companies is they start in an area, they get really good in the area, and what they tend to do is then fill out around their core. IBM started in the enterprise and filled out around it. Cisco is filling out everything they can around networking and making bits move around the Internet. People fill out around a technology model and a business model.

The thing that's actually probably most unique about Microsoft is we've already got two of these things. We started out as what everybody loved to call a desktop company, and now we've got a huge enterprise business, and we're trying to be a three and four trick pony because we're trying to do devices and entertainment, which is yet another business and technology model, and, of course, advertising and the Web, another model.

So, yeah, I would say we're fairly unique ourselves, and I would say if you take a look at almost any other technology company -- doesn't mean they don't have huge growth potential, doesn't mean they're not hugely successful -- but, you know, it generally confines you if you're sort of wed to one basic technology approach and one basic business model approach.

JOHN BATELLE: And yet the rewards, at least in terms of Wall Street, seem to go to those who have the focus, right? Has that been a frustration of yours in terms of the range in which your stock price has traded in the past six years?

STEVE BALLMER: No, I actually disagree that the rewards on Wall Street go to the people with the focus. Wall Street just likes success, frankly speaking. Now, if -- you know, if we produce a business, an advertising business or any other, that generates another US$5 or $10 billion of profit, we'll get rewarded. Wall Street won't miss that fact, they'll focus in on it; we've just got to go do it. And when you have, as any technology has, we have competencies, and then people want to pigeonhole you and say you're only about this, you can only do that; oh, now you do this and that, but you'll never do a third thing.

JOHN BATELLE: Right.

STEVE BALLMER: And your own people want to pigeonhole you. You know, people say, well, that's not the way we did it in the other business. So, you've got to build new muscle and new skill.

And yet if you really want to drive growth, operating income growth, share price growth, you've got to surprise people by developing an ability to not only do the old things better and better, but to go do new things, and I think we've proven that we have some of that capability and we're going to prove it again in both the devices area and the online area.

JOHN BATELLE: Let's talk a little bit -- I want to -- I know that you and some colleagues have some stuff to show us, so we're going to get to that in minute, but since you brought it up, I wanted to ask you sort of a two piece question. You're famous for declaring: developers, developers, developers, developers, developers, developers, developers.

STEVE BALLMER: I do love developers.

JOHN BATELLE: Yeah, you love developers. And yesterday we had a conversation with Mark and others where we discussed a new platform started by a Harvard school dropout that has attracted a ton of developers very, very quickly. What do you make of that? Mark's claimed yesterday that there were 100,000 developers on Facebook.

STEVE BALLMER: See, actually I think more than that. It's 100,000 different companies. If you look at the number of people, I'll bet it's even bigger than the 100,000.

JOHN BATELLE: Yeah, so what do you make of that phenomenon? I mean, you have a partnership, so you've got poll position into understanding that company in a certain way, but do you look at that and say, that's a threat to my operating system business, that's a threat to my advertising business?

STEVE BALLMER: No, I don't look at it and say it's a threat. They've done a really nice job, a really nice job on the developer platform, and I think everybody should recognize that.

There are basically -- any really exciting application, in my opinion, will have a developer story, and yet it doesn't necessarily replace an operating system. I mean, there's a developer story around SAP. There's a developer story around Microsoft Office. There's a developer story around search and advertising platforms. They don't replace the core operating system, but they have their own unique and exciting kind of extensibility model, developer community, and certainly, you know, the Facebook guys have done a nice, nice job on that.

JOHN BATELLE: Well, the next question that I'm going to ask you might lead into what you were going to show me because there's another area where there are a ton of developers. And we had the CEO of Adobe here yesterday and --

STEVE BALLMER: A former Microsoft guy.

JOHN BATELLE: I'm sure he would acknowledge --

STEVE BALLMER: He learned all his tricks.

JOHN BATELLE: -- he learned everything he knows at Microsoft. (Laughter.)

But it seems you might be playing a little bit of catch-up with this particular alumnus of Microsoft.

STEVE BALLMER: Sure.

JOHN BATELLE: And, so, this might be a good time to ask you about Silverlight and where that's going, and maybe you have something to show me there, as well.

STEVE BALLMER: Yeah, we'll have Dan Fernandez come on out, and maybe I'll just make a couple comments. Dan's going to show us a little bit of Silverlight and this mash-up tool that we've been working on that's going into public beta called Popfly.

But, you know, at the end of the day you have to say, okay, what's going to happen in the future of the Web? And the truth is, Web applications will continue to get richer and richer, both in terms of the kind of media they present and the kind of logic they want to associate with them.

At Microsoft, we wouldn't predict that Windows and Mac apps go away, they do another job, but Web sites, Web experiences are going to continue to get richer, and want to be more programmable.

And we think there's a lot of opportunity for innovation. Obviously, Adobe's done a good job on some of the rich media stuff. We think there's a long way to go.

And so we brought out this tool called Expression which goes with our Visual Studio product, and a runtime called Silverlight, which is cross browser, cross operating system, and allows you to write rich applications, starting with media and eventually richer and richer in terms of computation and logic.

So, maybe I'll have Dan Fernandez come out and do a little demo.

JOHN BATELLE: So, Dan's going to show us a few things, and including, as I understand, this is a public unveiling of something, is that right?

DAN FERNANDEZ: Yeah, we're actually -- today Popfly is now in open beta. So, everyone here can now sign up for Popfly; just wait till I'm done with my demos. (Laughter.)

So, I'm going to show you two things. The first one is Popfly, and Popfly is a Silverlight application, and it's designed to let anyone create without writing code. So, what I'm going to do onstage is build a couple applications using services from Facebook.

So, this is Popfly. We have these blocks, and I'm just going to type Facebook here. What I want to do is create a couple applications using my friends from Facebook. So, if I zoom in here, you see these great effects with Silverlight. I'm just going to get my friends, and if I click this little light bulb, you'll see that the left pane shows me what other blocks can connect to Facebook. Now, we have data source blocks from Facebook, from Twitter, from Windows Live, as well as transformation blocks and display blocks using Silverlight.
So, I'm going to click page turner, and I'm going to use Facebook to build an actual face book, a photo book of my friends. So, to build this application I just connect those dots, click preview, it goes to Facebook services and in real time I have my first Silverlight application; didn't need to write a line of code.

Now, because this is Silverlight this does dynamic resizing, as you can see here, and, in fact, I could embed this into Facebook directly.

So, there's my first Facebook application. I'll show you one more. Just for a little fun, we'll build a game. We'll build Whack-A-Mole and send in my friends here and click preview. And there are the photos; I'll make sure to only hit Microsoft employees. Steve, you tell me if you want to me to hit John and -- (laughter) --

STEVE BALLMER: Give me another 25 minutes and I'll tell you. (Laughter.)

DAN FERNANDEZ: It's a minute to learn, a lifetime to master.

So, because it's Silverlight, we can embed this not only in Facebook but in Windows Live Spaces. So, here's the same application running in Windows Live Spaces dynamically resized.

Now, let's build another application, and this time we'll build a presence application where we can see what our friends are doing. I will do a quick search for our bubble updater, and just connect these two blocks together. And if I zoom in, what bubble updater is going to do is pull the name and the status, what my friends are doing, and their photos. So, I click preview and there's my status of my application.

Now, the great part is, using Windows Vista I can actually drag and drop this onto my desktop, so anytime, anywhere I can see what my friends are doing on Facebook, and all of these gadgets on Popfly can live in Windows Vista sidebar.

All right, so, let's build a little more complicated application, this time not just Facebook, we'll use Technorati. So, I'm going to add Technorati, which is a blog search engine, and what I'm going to do is connect Facebook to Technorati, and I'll zoom in and see what I can do with Technorati, and I'll say, get search summary, and I'll pass in my friend's name. What I'm going to is that'll go over to Technorati and give me how many times that person's name is mentioned in the blogosphere, and I'm going to use another block, Silverlight block, called photo rank. And I'm simply going to connect Facebook to photo rank and Technorati to photo rank, and what this will do is I'll use the search result, it'll dynamically size the image larger or smaller based on how popular they are on Technorati's search engine.

So, we go to Facebook, we get our friend, then we call Technorati, passing in the name, and then we combine it all together into Silverlight. I didn't need to write a single line of code. Let's click preview, and there's our image of my five friends, pulling the result and, John, you can make fun of Tim for having the biggest head.

So, that's Popfly, great tool. The last thing I want to show you is this is the Web site for Quiksilver. So, anyone that's ever worn a bathing suit is probably familiar with Quiksilver, but this is their Web site for the world championship, which happened recently in France. What's interesting is this is all running on Silverlight Streaming for Windows Live.

So, we have statistics here, and I'll just pull up some of their top 10 wave scores and show instant replay. Notice we have picture-in-picture. And they had over 418,000 videos, comprising over 16,000 hours, all on Silverlight Streaming. In fact, end users could actually publish their own videos.

So, there you see a perfect wave ride. I have the ability to share, blog, and rate this, as well.

Now, another interesting aspect that Quiksilver did was actually add Windows Live Messenger integration, so the 250 million users of Windows Live Messenger surfing fans can actually communicate with each other.

So, I'm going to start this activity here, and there's Brian. Now, Brian can be anywhere, and him and I can see videos in real time. So, I am just going to share the wipeouts video here, and give this a second to cue up. And now Brian and I -- hey, Brian -- and now we can see the agony of defeat as people have terrible wipe outs. Ouch, yes, painful. I'm sure there's an Internet wave joke in there.

Okay, so, we've seen how you can use Silverlight, how you can use Windows Live and Silverlight Streaming to host your business. Thank you. (Applause.)

JOHN BATELLE: Well, that was very Web 2.0.

STEVE BALLMER: Good. This is a Web 2.0 conference.

JOHN BATELLE: It felt like that was quite Web 2.0. It sounds -- and the kind of thing that perhaps will encourage a lot more developers, because a different class of -- the same almost kind of developers that might be drawn to the ease of the Facebook platform, for example.

STEVE BALLMER: Right. I mean, in almost anything you're going to get a range of developers, from hardcore folks who want to roll up their hands and use, you know, C# and .NET or C++, all the way to the old macro programmer crowd who like -- and this is sort of designed for kind of that end user, quote, "programmer," somebody who doesn't have to necessarily be a code head but wants to do sort of fairly sophisticated rich presentation Web mash-ups.

JOHN BATELLE: Cool, cool. Well, congratulations. And I know there's other announcements that you've made today, as well, but that release -- the release is out there. We're not going to cover everything there because --

STEVE BALLMER: No, it's fine.

JOHN BATELLE: -- I've got a lot of other stuff to get to here.

STEVE BALLMER: Yeah, let's move on.

JOHN BATELLE: Let's move over to Office and particularly what is to me still a bit of a confusing brand, Office Live. What do you make of the Zimbras the world, the Google suite, basically, almost every platform company really taking a -- looking at the market of Microsoft Office and saying, I know how we're going to kind of get a piece of that market; we're going to do it online and we're going to do an end around your packaged goods shrink wrap business? Not that it isn't integrated already; I still get those updates all the time from Microsoft Office via the Web, but what do you make of that? What's your strategy? Are you concerned about it? Do you wake up at 2:00 AM worrying about Zimbra?

STEVE BALLMER: Well, I like to sleep well and I do. That doesn't mean we don't have a lot of good competition. I think the way I would recommend everybody think about the issue is not that my job is to do apps of a given sort of technology computation model, but we think of our job as to deliver productivity.

And, you know, just take our Outlook application. What is Outlook? Can you define it by its technology model? Well, it's a rich client app if you want it to be. You can access Outlook over the phone if you want to. We have mobile versions of Outlook for mobile devices. It runs in a browser through our Outlook Web Access technology, which is where we kind of pioneered some of the concepts in AJAX many, many years ago.

So, what is Outlook? Outlook is whatever best suits a given user at a given time in a given direction. So, I don't think -- I think what people, at least what we will do, is focus in on delivering an experience and taking advantage of all kind of technology models and models of computation: things running in the cloud, things running locally, things using, you know, kind of a browser-based delivery approach, rich client approach, a Silverlight-type approach. All of those things will play into the best experience for somebody who wants the kind of productivity that Microsoft Office provides.

JOHN BATELLE: Do you think Google Docs & Spreadsheet is a good product?

STEVE BALLMER: If you want to actually do what most of our customers do with Microsoft Office today the answer is, no, I don't.

On the other hand, you know, if you're trying to just get a few people to collaborate on a fairly simple thing, there are some nice collaboration benefits across the Web. And if you take a look at this Office Live Workspaces thing that we announced last week and we're moving into beta, essentially it brings very rich, even richer collaboration capabilities for every Microsoft Office user to that experience.

So, yeah, sure, we pay attention to what the other guy's doing. If there's good ideas, we're moving aggressively, but if what you really want to do is something that you would do in Word and Excel today, Word and Excel is the best place to do it.

JOHN BATELLE: Got it.

Let's talk about search. Is it one of those kids that you might privately smack on the back of the head and say, "Do a better job in school, Jimmy"? (Laughter.)

STEVE BALLMER: No, I actually think -- no, absolutely not. Absolutely not. I would say, hey, you know, you're just three-years old and we've got you in there playing basketball with the 12-year olds. And you know what, you're growing up quick, you're getting better every day, you've got all the potential in the world, and it may take until you're six or you're seven or you're eight or you're nine or you're 10, but you're going to dunk, and you're going to dunk on the other guy some day, Johnny. That's what I would say. (Laughter, applause.)

Now, what were we talking about? (Laughter.)

JOHN BATELLE: I think we were talking about the Zune. (Laughter.)

STEVE BALLMER: You need challenges in life, don't you?

JOHN BATELLE: Well, good. You know, actually I will acknowledge, as someone who's thought about search a little bit, that it's getting better. It really is. Obviously, it's important strategically, you would agree, and it's important for building an audience and particularly an engaged and committed audience, which is a bridge to my next question.

Microsoft, 25 percent of revenues -- you recently stated that your goal would be in a few years to move -- not to necessarily move -- to build a business at Microsoft that accounts for 25 percent of your revenues which is an advertising business. Now, you already have a very large advertising business but it's nowhere near 25 percent. Twenty five percent at current run rate, I'm saying, you know, is about a 14 or $15 billion business. That's the size of Mr. Murdoch's advertising business, who we spoke to last night. That's bigger than the current size of Google's advertising business. How do you get there?

STEVE BALLMER: Well, they -- you know, at the end of the day, we have to do three things, three things -- four things very well. Number one, it turns out you have to do search well. You have to do search well. The most valuable advertising real estate in a sense comes out of search, because about 20 percent of searches have some sort of commercial intent, and it's sort of -- you're never going to grow a big advertising base, you're never going to attract the critical mass of advertisers if you can't do -- if you can't be credible in search, so search is very important.

Number two, you've got to be sort of good at community and communications. That's why we think it's important the success we've had in mail and IM and other things we will do there. The partnership with Facebook is very important to us around that because that's another big sort of community phenomenon. That's very important.

Number three -- and these are kind of on the audience side but they're important.

Number three, you've got to have a very strong advertising platform that delivers kind of what I like to call all payloads on all media with all business models. So, you want to be able to deliver to TVs, to phones, to the PC screen. You want to be able to deliver text, pictures, videos. You want to be able to run auction rate card. You want to be able to sell demographics, behavior. You need a very strong advertising platform that brings together buyers and sellers of advertising.

And number four, you've got to actually have a lot of ads you sell on behalf of other people, not just ads you sell on behalf of your own audience, because I believe that these ad platforms will be real platforms --

JOHN BATELLE: So, the syndication business.

STEVE BALLMER: Yeah, syndication business, the syndication business, which I think will get redefined over the next five years, because it's got a fairly narrow definition today, and yet I think that will expand as we go forward.

JOHN BATELLE: You've made a couple moves in what might be seen as syndication business -- in the syndication business, some might say the boxing-out business, Facebook being one of them, we discussed briefly that deal, Digg being another recent one. How's the Facebook deal going? Are you making money?

STEVE BALLMER: Are we making money? Rumors have it that we're not, but in a sense the most important thing for us is every day we're smarter, every day we're learning more. There's more inventory that we have and more options we have to take to our advertisers, primarily right now what I would call the head of the curve not the tail of the curve advertisers, but more and more we want to be able to take not only our own inventory but our syndicated partners' inventory out to the tail of the curve.

So, we're delighted about kind of where we are and what we're learning, and, you know, there's a lot of things we invest in, and this partnership and this relationship and this ad platform is a place we're investing.

JOHN BATELLE: Now, speaking of investing, you made the largest acquisition in the history of Microsoft by quite a bit, as I recall.

STEVE BALLMER: By about a factor of five, yeah.

JOHN BATELLE: By a factor of five, in purchasing aQuantive, and you've since put the head of aQuantive as the head of your platform basically and your advertising strategy, and Brian will be out here later this afternoon. I'm looking forward to that conversation. We'll go a little deeper into the advertising piece.

But what's the -- you know, what was the -- what was it about aQuantive? What did you -- what do you think you got that made it worth $6 billion, and what's the timeframe for making that work for Microsoft?

STEVE BALLMER: As I said, one of the four key attributes is the ad platform itself. The ad platform requires, in fact, three or four things: Great tools for advertisers, great tools for publishers, great kind of matching algorithms and business logic, and customer data and intelligence that sits between.

And if you look at aQuantive, aQuantive had a fantastic tool set and presence for advertisers, an incredible amount of important customer intelligence in data, and a really nicely emerging business in tools for publishers, not just the technology tools but also including the ad network, Drive PM, that was embedded. All of that is fundamental.

Now, there's also a large online ad agency, of course, as aQuantive.

JOHN BATELLE: A very large online -- yeah.

STEVE BALLMER: And it, too, has a significant value.

I would say the focus in our acquisition was on the tools for the advertisers, customer data and the tools for the publishers.

JOHN BATELLE: Right. Might you spin that agency back out?

STEVE BALLMER: No plan to do so, no.

JOHN BATELLE: We have microphones up here. I want to make sure there's time for questions, so please queue up, and then I'll grab you guys as you do. Please feel free, ask lots of questions.

While those guys are coming to the microphones I want to ask you this. What's the next $6 billion acquisition or is it 15 billion? (Laughter.)

STEVE BALLMER: Unfortunately, unlike Mark yesterday when he told you that financing is going well, I will tell you, no comment.

JOHN BATELLE: All right, let me rephrase that. (Laughter.) What are you interested in that Microsoft currently isn't doing?

STEVE BALLMER: Well, there are a lot of things. We buy a lot of companies. We'll buy probably 20 companies a year consistently for the next five years, and most of them you'll either sort of footnote, you'll say, oh, you know, they bought MotionBridge, that's interesting; oh, they bought ScreenTonic, oh, oh, oh. Because we're talking about acquisitions that are 50 to a hundred, to a couple hundred million. When you talk about --

JOHN BATELLE: Which is really no big deal; it's just a couple hundred million.

STEVE BALLMER: Well, not on our base.

JOHN BATELLE: Remember, we're in the Valley here. That sounds all right then. (Laughter.)

STEVE BALLMER: I don't know, there's some pretty big dogs running around the Valley, too.

But, you know, those are good acquisitions. They're important to us and that's most of what we'll buy and it is of most strategic importance, so it's important to us, too.

But when you say, what are you going to buy that's 6 billion, 15 billion, there aren't many things up in that kind of price range. At that price range you're talking about, you know, a handful, a dozen, a couple dozen things, and, of course, on any of those all I could basically say is no comment.

In the meantime, we will probably buy 20 companies for anywhere between 50 million to a billion bucks. Last year we bought a whole bunch of stuff in the Web space. We bought TellMe down here in the Valley. I was out visiting there. So, we'll continue to buy a lot of stuff, a lot of stuff.

SteveB@Microsoft.com if you have something you want to sell. (Laughter.)

JOHN BATELLE: Oh, my God, are you about to get a thousand e-mails. (Laughter.)

STEVE BALLMER: But maybe one of them will be worth buying.

JOHN BATELLE: Yeah, there you go. One in a thousand ain't bad odds.

All right, the mics are open. If you guys aren't asking Steve questions, I'm going to keep asking. Hold on one sec because I've got one more.

Everybody seems to want to put you together with Yahoo! What do you say?

STEVE BALLMER: Everybody should -- everybody can have their opinion. It's -- you know, Yahoo! is a great company. We've had a good relationship with Yahoo! We compete with them quite clearly.

JOHN BATELLE: Sure.

STEVE BALLMER: On the technology side we've had a good sort of constructive relationship, in the datacenter, with the browser, et cetera. I have a lot of respect for those guys. And I think it's kind of a thing people do when, you know, one guy is so much bigger, number one really is a lot bigger than number two or number three, and the logical kind of psychological thing for people to talk about, and that may or may not make sense to us, to Yahoo!, to investors --

JOHN BATELLE: Well, if I want to get to a $14 billion ad business I can get halfway there by buying Yahoo! right now, but that's just me.

STEVE BALLMER: Yeah, well, you should do it then. (Laughter.)

JOHN BATELLE: Let's get to questions.

QUESTION: John, you actually sort of took my question where I was going with it. I was surprised we hadn't talked about Yahoo! yet. As you think about trying to compete with that behemoth, and Michael Moritz talked about this yesterday, about how big Google is, how much power they show.

Is it one of those things where it's almost a necessity that there has to be some sort of either partnership or merger with Yahoo!? And because of the cultural difference and almost sort of how the Valley's been predicated somewhat to a degree against up north, you know, is there something that has to take place, first of all? Does there need to be a spin of MSN? Is a spin of MSN even on the table where then you merge MSN with a balance sheet with Yahoo!?

STEVE BALLMER: There's a whole lot of complexity in that question and I'd better give a whole lot of simplicity in my answer. (Laughter.)

If we didn't believe we could do well, we shouldn't be doing what we're doing, and therefore we believe, we believe in our independent path. We believe, based upon the investment we're making in R&D and cap-ex and aQuantive and everything else, we like what we're doing. And I'm sure if you were to talk to Jerry and the guys at Yahoo! they'll tell you they like what they're doing.

And, you know, if at some point it made sense, maybe it would make sense, but that's not where we're going. We're driving an independent direction that we got a lot of enthusiasm about, and everything you talked about sounded awfully complicated to me.

JOHN BATELLE: Over here.

QUESTION: You said that you believe that the ad syndication business will get redefined over the next five years. Could you elaborate on your thoughts there?

STEVE BALLMER: I'm sorry. I missed a couple key words. Just repeat, please.

QUESTION: You said that you believe the ad syndication business would kind of get redefined over the next five years. Could you elaborate?

STEVE BALLMER: Yeah, I think -- you know, if you look at today, most of what goes on today in syndication is done essentially with simple text ads that are largely sold via auction and with non-guaranteed delivery. And you have to say over -- in the future, if the bulk of advertising will be sold with at least an assist from some ad platform -- that means video advertising and pictures, in addition to text, it means you've got to evolve so that the ad engine can do -- the ad platform does guaranteed delivery, it does delivery via rate card not just on an auction-based approach and a performance base.

So, I think there's going to be a lot of changes to syndication over the next four or five years and a lot of technology that goes around that. I mean, today the only thing really anybody sells is context: I can put you someplace where I know what's there. And the truth of the matter is, particularly as advertisers more and more want to buy behavior -- I really want to be in front of anybody who shows the following behaviors no matter what they're doing -- not only does it define -- redefine advertising, it certainly redefines syndication in a fairly powerful way.

JOHN BATELLE: Right here.

QUESTION: I'm curious, you mentioned earlier on that Microsoft's now around 80,000 people. So, how do you keep an organization that size nimble and moving forward? That's part one. And I guess the second part is, there's a lot of smaller companies here that probably are interested in partnering with Microsoft. Do you have any advice on how to sort of rise above the noise or to partner with an organization that big?

STEVE BALLMER: Sure. You know, is Microsoft a company of 80,000? Sure, because we all happen to share the same balance sheet and income statement. There are some things that we try to do synergistically, and a lot of things that are done very autonomously and locally.

And so, in a sense, do you work for a company of 80,000 people? Some days it feels that way, and some days -- you know, I was down at TellMe yesterday. The TellMe group's just fired up focused on what they're focused in on. I was with the guys from our SQL Server team. I won't tell you they can't spell advertising, but they don't sit there and worry a lot about the changing dynamics of the syndication market. They've got a different set of issues on their plate. And part of what we need to do if we're going to be a multi-trick pony is we've got to let people specialize, and that means in some senses we're not 80,000 one company; we're 20 or 30 percent synergy, and 70 percent or whatever kind of independent action in the divisions, and I think we've driven that kind of hard.

You know, at the end of the day most of the key decisions will get made for companies that want to partner with us in our operating businesses, and the best way -- you know, people say, hey, is it hard to break through because you're a big company? Yeah, probably a little bit harder, but I also find that as soon as our guys actually get interested in something that's going on outside, things move awfully, awfully quickly. And so it may be hard to generate interest for something for which there is none, but it is fairly easy I think to convert when we have a little bit of interest in an area.

JOHN BATELLE: Over here.

QUESTION: A lot of the larger Web companies are based on Open Source software, and when you buy a smaller company it's really easy to move them over to your platform. How would you be able to handle taking in a larger company like Yahoo! or Facebook where you have a huge site built on PHP or Linux?

STEVE BALLMER: I can't say I spend a lot of time -- let's put Yahoo! and Facebook aside, because I think we can agree those are very different animals.

We will buy smaller companies. We will buy smaller companies, some of whom make some use of Open Source software. There are some IP issues that we have to think through that are associated with that. Sometimes we will migrate those properties. Sometimes we will migrate those properties slowly.

I don't discourage people to talk to us about partnership and acquisition just because they happen to have done some work on Open Source. That would take us out of the acquisition market, you know, quite dramatically, and we will do some buying of companies that are built around Open Source products.

JOHN BATELLE: Over here.

QUESTION: Good morning. I'm from G.ho.st. G.ho.st stands for the Global Hosted Operating System, and we're actually starting to compete with Windows but coming from a different angle, by giving a computing environment which is on the Web instead of a computing environment which is installed on one machine. So, I just wanted to hear what your thoughts are on this new competitor? (Laughter.)

STEVE BALLMER: Welcome to our world. (Laughter.)

No, I do think that there will be essentially evolution, and you almost have to think about platforms for mobile devices, platforms for servers, platforms for PCs, platforms for TV and platforms for the cloud. I actually agree with that. What level of affinity will people want to cross those platforms or not want, affinity in the way things get sort of managed, controlled, distributed; will things migrate? Many people think the world migrates to a single compute model where everything gets done in the cloud and then just painted onto screens. I actually think the world will have a much more balanced approach with computation occurring in multiple places but, hey, it's a -- the water's warm. We've got an entry across the spectrum, and we're glad to have good competitors pushing technology innovation at the platform level in all of those areas.

JOHN BATELLE: Thank you for the question. I'm sorry guys, but we've already run five minutes over so I've got to cut if off. But, Steve Ballmer, thank you so much for a great conversation, appreciate it.

STEVE BALLMER: Thank you.

END

Read More: