Remarks by Robert Youngjohns, President, North America Sales & Marketing, and Corporate Vice President for Microsoft, About Open Source Software in the Business Environment
InfoWorld OSBC Conference
San Francisco, Ca.
March 25, 2009
CONFERENCE CHAIR: Thank you, Jonathan. And for the Open Source entrepreneurs in the audience who are wondering what their business strategy should be, I think you’ve just given a blueprint of how to make money on free stuff. I thought it was fantastic. Our next keynote, at least initially, historically we would have thought polar opposite. But like with global warming, the polar caps are melting and Microsoft has warmed up to Open Source. Robert Youngjohns, President of Microsoft North America, will be addressing us next. I think is fantastic. We actually talked for a while with Microsoft about who would be the right person to speak. We’ve had in the past Brad Smith keynote; we’ve had others from Microsoft speak. Robert is different in that he actually has to make money. And so if it fails, I think he gets fired. So it’s good to have him up here to talk to you about Open Source and what it means for Microsoft.
MR. YOUNGJOHNS: Great. Thank you. I’m really privileged to be here. This is the third year that we’ve participated in this event. I’m delighted to be given the opportunity to speak. I feel a bit of a personality disorder at the moment. You should know that I’m actually a relative newcomer to Microsoft. I’ve been in the company just over a year. My former boss, Jonathan Schwartz, has just been talking to you. And then prior to that, after I finish speaking, my former boss from IBM will be speaking. So if I lapse into the wrong company, I have to ask you to just accept that.
What I’m going to talk about really is some of the approaches and ideas that Microsoft has about Open Source and about software development. I’m not going to talk a lot about products. I’m tempted to, having seen the size of the numbers I just saw. But I’m actually going to talk a little more about what we see going on in the business, what we see driving innovation, and how we’re responding to that innovation. I’m going to start by making some comments of what we’re actually seeing in the economy.
Every year Microsoft brings together a group of some of the most senior CEOs in the world to what we call the CEO Summit in Redmond. And I was privileged to be there last year. It was an interesting audience. This has the CEOs of Citibank, of JP Morgan, of Goldman Sachs, major retailers, major people in the luxury goods business, and so on and so forth. And there was one session where they were all asked, “So, what do you think is going to happen in the next 12 months?” And this was asked by Warren Buffett and Jack Walsh. They went around the room, and everyone said, with one exception, the worst that’s going to happen is a very, very mild recession. The only exception, interestingly, was the CEO of Costco, who said we’re in for a very nasty downturn. And everyone looked at him and said, “Well, you would say that, wouldn’t you, because you’re the company that stands to benefit from downturn rather than the rest of us.”
Now, we have that conference coming up again in May this year, and I fully expect when the same question is asked, and incidentally, some of the CEOs who were there last year won’t be there this year, I think the answer is going to be, “It’s terrible. This is worse than the great recession of 1642 or 1783 or whatever,” because everyone’s become historians now about the pattern of previous depressions. And I guess my conclusion is it’s probably never as good as it seems or as bad as it seems.
Certainly what we’re seeing in the market is, while clearly the desktop market is under significant pressure, our customers are really, really interested in how they optimize their infrastructure, how they get better value out of the computer systems they have, and how they have a pragmatic strategy of going forward. Now, I’m lucky enough to be in the industry a very long time. I get announced inside Microsoft as a veteran, and I’m not sure they’re flattering me when they do that. And certainly when I look at my own personal retirement portfolio and I stick that date into my retirement date calculator, I will be very much the oldest employee that Microsoft has ever had as I retire at the age of 92. But I have seen a lot of friends in this industry. And I’d like to reflect, before I start to talk about Microsoft’s view about where this vision has come from and where this journey has come from.
I started playing with Open Source in the mid-80s. I became fascinated by the mathematics of fractals. Now my wife will explain that I’m — in her description, she’s very good at bringing me down to earth, a million miles wide and a quarter of an inch deep on one subject. But I became fascinated by this idea of what happens at the boundary between order and chaos, and the incredible richness of that boundary. And I started to write programs in Basic, of all languages, in 1986-87, to try to actually resolve the map set on the IBM PC running at eight megahertz. This was not a fun experience. It actually took about 12 to 14 hours to resolve the basic map that you see up there, and that was just in monochrome, not in color. And then I discovered this interesting group called The Stone Soup Group. The name comes from a fairy tale from the Grimm Brothers, and it’s all about how a community in East Europe makes this great soup when this magician comes into the village and finds everyone starving and puts his magic stone into the cauldron. And as he does that, all the villagers bring the carrot they got stored away, or the potatoes they got stored away, and they create this great nutritional soup. And this group was formed to write programs to exploit and generate fractals. I actually started to write codes for it. I was at IBM at the time, and just before I submitted my first piece to this, after I had to check with my manager because being a law-abiding sort of soul, I did remember writing some or signing some document that said I will never move any of IBM’s intellectual property or intellectual property I derived by working in, et cetera, et cetera, et cetera, up to any third party, at least absolutely not.
Very much this is how this whole movement started, at the same time that the GPL license was being evolved. It’s a fascinating exercise and a great example of how a group of people come together to create code that is far more sophisticated that any individual can produce. And then we look at where we are now. This happens to be a map from my own personal desktop at home. For those of you who are interested, I’m running Windows 7 vapor edition here. On top of it, I’m running a combination of Microsoft’s Virtual PC and Sun’s Virtual Box software. I’m running Ubuntu Linux here. I’m running Open Solaris. I’m running NorPix. And then just for the hell of it I’m running Windows XP in a Virtual PC space with an emulator for a 1982 Sinclair spectrum. And actually it’s a really good platform gauge. 48K. It’s amazing what you can achieve within 48K. And I did this because we were having a discussion actually at a management meeting about interoperability between the Linux world and the world of Windows and how we needed to make a big push on that. And I said I could prove I could do this on my own desktop. Actually, there’s a copy of a Linux distribution called Frugalware sitting behind this. Frugalware is a very, very recent — if you go to distrowatch.org, you’ll find it. Someone has actually reduced the Linux kernel down to about 48 megabytes, and the only problem is it doesn’t have the GUI, which means it isn’t very good pictures here.
But this industry has changed. We recognize that change. There isn’t one constant model here driving this forward. It’s pragmatic. It’s about interoperability, and it’s about making sure you energize all elements of the energy there is out there to develop software. And Jonathan’s right. The developer is a really, really important part of this exercise. You can get best possible platforms on which those developers can create the next generation of products.
So what do I conclude? This is an evolution. It’s not a sudden change that’s been going on. Microsoft’s success really came from the opening up of the PC, in the 1990s opening up of the network. I was at IBM at the time working in the network systems division. And I remember the then-head of the network systems talking about TCP/IP as snake oil, with a hiss from her mouth as she said it. Now, the real world will get used to the fact that SDLC, SNA and all that sort of stuff is the only logical approach forward on networking. This TCP stuff is just too messy. It will never catch on. And it clearly did catch on. It created this concept at the end of the network. And I think today what Microsoft and all other companies are about is trying to create the most open, the most pervasive platform on which developers can create applications and which gives the customers the maximum amount of choice. That’s what we’re setting out to do.
But underlying all that, it is all about business. That developer of Frugalware — and I can’t remember now if it was Argentina or Poland where the developer is based — they’re not setting out to create a business from their distribution. They’re not doing it because it’s nice to do. Well, maybe they are. I don’t know. Sometimes my wife looks at the screens I put up at home and says, “Why do you do that? What’s the point?” And I go, “Because I can.”
What other reason do you need?
But this is all about business. It’s all about creating an environment where you can generate innovation for customers, where you can generate environments in which people can generate software, and you can make money from it. As I say, even that developer of Frugalware, it’s very clear they’re targeting a release, a distribution of Linux that will be suited to the mobile phone world and hope to build a business out of that. I respect that.
That’s what it’s all about, and every company up here talking to you is going to be from a different model to energize this whole community to create a business which they can then drive going forward which delivers innovation to their customers and delivers a new set of applications to users.
If you look a little bit more into that and look at how we see this. This is a constraint versus reliability issue. And actually I was reading Jonathan Schwartz’s blog a couple of weeks ago. It had a great comment in it. I think he made the same point with his three points just now. I think in his blog he said something to the effect — and I may not get the words exactly right — “We offer software which we charge for, for those customers who find “free” to be more expensive than a fully-supported software product.” Those are pretty well the words he used on his blog. I think what he’s saying there is this sort of matrix. There’s a matrix of, on the one hand, absolutely no vendor support. On the other hand, the world of the highly outsourced, I go to EDS, I go to an outsource and they run everything for me and I get no choices. And you make decisions along there. On one side it’s totally unconstrained; on the other side it’s completely constrained.
And as you go through this, you see that there is actually a map here which was what the next slide was about, about how risk maps onto that. At the one hand, the far left-hand side of this, where it is totally unconstrained, where you have a Linux release coming out every day, every week, you have this unconstrained environment, potentially very expensive environment. When I was actually at Sun three years ago, I remember making some calls on a bank that no longer exists today, and I was in the IT department, walking a long corridor past a lot of offices that said this VP, that VP and the other on it. And one of them was VP operating systems. And what this individual was charged to do was create a specific distribution of Linux for that bank. I remember pointing out at the time that didn’t seem to me to be the bank’s core business. The bank was digging down here right on the left, with having to create something from scratch and not having someone else whose job it is to be the supporter of taking out the risk from that equation.
Now, I wouldn’t like to suggest that was the reason why that bank subsequently isn’t with us, but it was a method of thinking at that time in that industry is that you could do this stuff and somehow it was cheaper and free. You know, Scott McNeeley has a wonderful saying that he used to use a lot when I was at Sun that says, “If anybody says to you something is self-evident, it’s very clearly not evident.” And this comment that somehow free is cheaper I think falls into exactly that category.
Out here is this equilibrium point. There’s an equilibrium point where you accept that there is a reduction of freedom in the sense that someone else is looking after your software environment, and as a result of that, you can get significantly lower cost at significantly lower risk. That is where we are setting up to position ourselves in this particular framework. Now, in my job when I was first placed in the introduction, the comment was made that I had to actually sell stuff and make money. If I don’t, I probably won’t be here a few months from now. I spend a lot of time talking and listening to customers. As we evolve our strategy towards Open Source and interoperability, it’s very much from this, listening to customers, that we derive the products and actions we’re taking.
So here’s some examples of things we’ve done recently. We’ve recently announced a web installer which is designed to help encourage the web ecosystem. It installs very quickly. I actually did it this morning before I came down here in my hotel room over a fairly slow Internet connection. I suspect a lot of people in this room were trying to share that same connection with me at the time. But this provides not just the Microsoft stack, but provides Open Source components such as PHP. The whole purpose is to get people going quickly.
We’ve recently announced that we’re teaming up with both Novell and Red Hat on how we create virtualized data centers and how we can support both Red Hat Linux, SUSE Linux and other variants within a data set, and give a common management framework for the managers, the executives who are trying to run those data centers and reduce costs because that’s what our customers are asking for. They’re asking us very specifically, “Don’t take a simplistic view of this. Don’t assume that we’re going to make a binary decision between Windows Server 2008 and a distribution of Linux. We have to deal with the reality of having both these present, and we want you, Microsoft, to step up to that and embody that in your virtualization technologies and your systems management technologies. And that’s what we’re setting out to do.
And if you look at some more specific examples of what we’re doing, we have a Web site, for example, called CodePlex. And CodePlex is a repository for the Open Source projects that are built around Microsoft code. It’s a bit sad to me that the most commonly used project is something about creating and comparing characters from War Craft. But that’s I guess more about the nature of the community that we all deal with. But AJAX’s control tool kit, for example, is being developed in there.
One point I’d make about CodePlex is that it relates back to some of the platform offerings that Microsoft has. SharePoint is a product, a collaboration software. If you like, it’s the Facebook for the enterprise. It’s the fastest-growing product in the history of Microsoft. It went from zero to a billion dollars, revenues faster than any other product that’s ever been done in our history. And right now, we have over 300 open-source projects building on top of SharePoint to deliver new functionality, new capabilities to the customer of that product. So this is one of the examples of how we’re trying to stand out and reach that open source community.
Another example is our commitment to interoperability. We made some significant announcements last year recognizing that where we come from, which was a world of proprietary document formats, which was a world where every application wrote to return code – including Open Office, incidentally – wrote to formats that then couldn’t be read by other office systems. We needed to move away from there. And we made some pretty significant announcements in our long-term commitment to interoperability, our long-term commitment that once we started these formats we would support those in the long term.
You see up there, the open specification promise is part of this effort, a commitment that we will create environments that will allow people to produce applications knowing that they can interoperate in the long term. This is the part of what we’re seeking to do as we move forward.
So my conclusion – and I’m probably finishing rather quicker than I expected to finish, and all of you I’m sure will rate it as good news – is that we think this is very much about the business. It’s about making sure that we take a pragmatic response to what our customers are asking for today; making sure we recognize they really are in heterogeneous environments; making sure that we provide the systems management tools that bridge across those heterogeneous environment departments; making sure we support interoperability; making sure we support open source code, where it’s appropriate and where it’s applicable.
Our support for PHP is a good example of that. Because we believe pretty fundamentally in the power of the software industry. One of our missions in the company is we develop software to enable people in companies to reach their full potential. And that’s something that drives us very strongly internally. But we also know that to continue that innovation, you need to extend to a wider community than just our own operating environment, and that’s what we’re doing through interoperability commitments and our commitments to allowing open source to extend. Our ultimate objective is to make Windows, our own platforms, the best possible platform for the deployment not only of open source applications, but applications that are coming from other development methodologies. That’s what we’ve committed to do. We believe it’s the right strategy. It’s a pragmatic strategy, and it recognizes where people are today.
As we talk to our customers about infrastructure optimization, the message they give us back today is, “We want you to help us take the total product, end to end, from this infrastructure down. We’re not looking for you to force us into binary decisions.” That pragmatic approach is very much, for example, how we approach cloud computing.
Just a comment here in terms of the strategy, because we see cloud computing not as a binary technology ultimatum between running software on premise or running it off premise. We believe the pragmatic in many companies need to be able to move between those two states. They need to develop software that runs locally and then deploy that very same software into the cloud. We don’t see the cloud as being a technology ultimatum that you deliver back to customers. We see it as much more pragmatic than that. We see a customer developing a piece of software, choosing to test and run that software on their own systems, and then at some point in the future be able to move that into the cloud. And by the way, move it back again later if there’s what their business needs demand. So as you talk about one specific example of what we do, which is our hosted mail solutions, it’s exactly the same software running on the cloud as is running on the premise. And that’s very, very attractive to our customers who often feel that they don’t want to make the full step of going into the cloud. They want part of their application running on premise and part running on the cloud, and they want the ability to make that change in the future.
So our strategy is about pragmatism. It’s about providing the best possible scalable platform on which application developers — where they open source, where they use conventional application development techniques or whatever to deploy their products. And also giving you the capability to deploy those products in the future in our Cloud computing environment in a very pragmatic way. So I thank you for your time. I hope you have a great conference.