Remarks by Steve Ballmer, Chief Executive Officer
May 26, 2011
STEVE BALLMER: Yeah, baby, it’s fun to be here! (Cheers, applause.) I don’t know what I expected, but it’s fuller than that.
It’s a real honor and privilege — if I don’t get too loud — it’s a real honor and privilege for me to have a chance to be here today. I understand we’ve got folks from actually not quite all over India but from a little broader territory than just Delhi, and I know we have folks here not only from the IIT but from a number of other schools. And to all of you I say thanks, it’s my honor to have a chance to talk to you, and hopefully have a chance to take some of your questions, and certainly the energy in the room here is great.
I understand this is the 50th year for IIT Delhi, Golden Jubilee. Congratulations to — (Cheers, applause.)
I have to tell you I’ve been hearing about IIT for many, many years. We’ve got a lot of IIT graduates at Microsoft, and some guys who I’ve had a chance to work with a lot over the years. There’s a fellow named Anoop Gupta who works for us, who was Bill Gates’ technical assistant and really pushed us in the whole area of video and voice communication, a good friend. We have some folks in Microsoft Research. Madhu Sudan is at our research lab in Boston, and was an MIT computer science professor, and a Gödel Prize winner, but a number of folks from IIT Delhi.
But I actually had my first interaction with IIT the day I arrived at Microsoft in 1980. I wasn’t sure what I was getting into when I went up to Microsoft. I was actually an undergraduate student — or sorry, a graduate student in business at Stanford. My parents thought it was great, I was getting a master’s degree. My father and mother, neither one of them went to college, so this was really important to them. And I had gotten a call a few months earlier from Bill Gates asking me if I wanted to drop out of Stanford to come to Microsoft.
I’m not recommending anything to anybody — (laughter, applause) — but in my own case I said, blah, blah, blah, talked to my parents, eventually, OK, I dropped out.
My first day at work I get there, I had no desk, I had no chair. Bill told me I could push some papers on his couch down and sit on his couch.
And I said, “OK, I’m here. What’s my first project?” (Laughter.) This is an IIT story.
And he pointed to a stack of letters. This was in the old days, you know, when people actually wrote letters on paper and the like. There was a stack of letters this high. And I said, “What is that?” He said, “Well, there’s some crazy guy in India who’s been — from IIT — who’s been writing me letters, wants to come work for Microsoft.” I said, “Well, when did he first start writing you?” He said, “About a year and a half ago.” “How many letters has he written?” Bill looked at the stack, and he said, “I think he’s written me every week for the last year and a half,” 18 months, every week, so there were like 70 letters. And I said, “Well, do I need to read them all?” And he said, “No, they all say the same thing: ‘Mr. Gates, I’d like to come work at Microsoft.'” I think he was working at the time maybe for I think it was called Hindustan Computer, what’s now HCL, and been an IIT guy, telling Bill IIT is the greatest, you’ve got to hire me, please help me, please, I’d love to work at Microsoft.
That was passion, that was drive. I was so impressed, my first activity when I joined Microsoft, I decided I’m going to hire this guy.
Now, at the time, getting visas to come into the United States was even sort of crazier than it is today, but literally the first person I ever hired at Microsoft was the most industrious, hardworking, aggressive, persistent IIT guy I’ve ever met. He’s now long retired and still lives in Seattle, but I thank IIT from my first day at Microsoft to today for giving me something to do. (Cheers, applause.)
What I want to do mostly today is have a chance to share a little bit with you about some of the exciting things we see going on in technology, what I think some of the opportunities are that that presents, and we’re going to show you just a little bit of a couple things that we’re working on.
How many of you are graduates or undergraduate students in computer science? Show of hands. Electrical engineering? Business? Something else? (Laughter.) All right, that’s fair, that’s fair.
I was going to say I thought we had everybody here was in computer science and electrical engineering, but I will tell you, if you’ve made a choice in your life to be in the technology field, to study computer science, to study electrical engineering, to do research, you couldn’t possibly have made a better choice.
We’re sitting here — I was talking to some of the faculty about this beforehand. Sixty year ago, approximately, the first computer came out. And yet the level of innovation in our industry, the change that will be driven, the value and importance of people who understand how to write software and design hardware is just going to continue to grow.
So, if you made a career choice to pick this career, congratulations. There’s plenty of work and plenty of opportunity to really make a difference in life.
As we look at it today, there’s a lot changing in the technology landscape. There are sort of big tectonic shifts in the core technologies in our business. And every one of these is a chance for an invention, for innovation, it’s the chance for somebody in this room or elsewhere around the world to conceive of a new product, a new technology or a new idea, and to make it into something big.
If you take a look at it today, one of the themes I think is most important is natural user interface, the notion that says we’re going to teach machines how to interact with us the same way humans would, they can learn to see us, to hear us, to let us touch them and interact with them. Just think about the impact touch user interface has made in the context of the phone over the last few years.
We launched a product last November, which we’re going to show you in a little bit, called the Kinect. It’s a sensor with microphones and cameras on it that works in conjunction with the Xbox and soon the PC. But it recognizes you, it can track your actions, it can reflect what you’re doing and see you, and respond to you.
Think about the innovations that we’ll see with voice. Over time, you want to be able to in many settings talk to your — I’ll say your computer. When I say computer, you can think phone, PC, TV, smart, intelligent, electronic device, you can interact with it.
When I look out in the audience, I see some people with phones to take a note or two. I see some people with computers. I see some people who have competitive devices and Windows devices. But mostly when I look out in audiences like this, and today is no exception, I see people who also bring pencils and paper. That’s all an opportunity for natural user interface invention. We need to have devices in which we can interact the way we still do with a lot of analog technologies.
The second one I’ll highlight for you is natural language. If the first job is to teach the computer to recognize us as human beings, natural language to me refers to the notion of teaching these systems to understand us, not just to recognize us but to understand us as human beings.
When I was getting ready for my trip to India, I said to my assistant, “Prepare for my trip to Delhi.” She knew what that meant. She knew that she should go check the schedule, see who I was visiting, pick up their bios, any customer records at Microsoft, any e-mails that I’ve had with the Microsoft India staff. It turns out what she knew to do was to go do a bunch of work on the computer. She does the same thing before every trip I make.
Why can’t we turn — why can’t I just somehow type or speak to the computer, “Get me ready for my trip to India,” and it automates and understands that whole activity?
Now, some of you will say, oh, that sounds crazy, that’s a big idea, that’s not going to happen anytime soon, but if you rewind, you could say search has some of these characteristics today; they just don’t mostly work.
When you type into a search engine, you’ll get — whether it’s Bing or Google — you’ll get back links to websites, and then you can click and go do something.
But it turns out you can actually say anything you want to a search engine. You can type in anything, and at least a search engine will try to do something on your behalf. But it’s not good enough. Today’s search engines really only understand nouns, they don’t understand how to do something. They don’t have enough semantic understanding of all of the world’s information. They index it all, but they don’t understand it well enough. They don’t understand you well enough to translate their understanding of you and their understanding of the world into a set of verbs or actions that they can take on your behalf.
That innovation is going to happen. It’s going to happen in a variety of different quarters and a variety of different ways over the next several years.
And so one of the real issues is how do we bring the full power of what people want to do in program development to the languages and technologies that everybody is learning? How can you write the richest applications in the world, building with the tools that are most comfortable and familiar?
Chips and form factors. When Bill Gates convinced me to drop out of college, this was the pitch, because, look, I had written a few programs but I wasn’t a technical guy, I was a business guy. I still am at the end of the day. And he said, “Steve, let me tell you why this is going to be so cool. The microprocessor is essentially free intelligence, and all we have to do is write the right software to take advantage of it.”
Well, we’re sitting here 30 years later, and the microprocessor still — and storage systems and network systems, they’re still really getting cheaper and more flexible and more adaptable than ever before. You know, Moore’s Law, the notion that you get more, more, more continues to work.
Literally, the notion, if you’d said even 10 years ago, essentially you’ll have a huge index of all — there’s literally today, if you want to have a good search engine, there’s 100 billion documents that you wind up indexing on the Web. You’re going to store all of that. If you’d said 10 years ago that 25 percent of the world’s pictures can be stored for free on a thing called Facebook, you might have surprised some people.
If you take a look at what’s going on with the ARM microprocessor today, where we’re seeing designs of systems that are essentially lower power and not necessarily higher performance, we get other adaptations of the technology.
So, the magic of what the hardware industry can do has not run out of gas. There’s still more to do with the great software taking advantage of essentially processing power that can be put in your pocket, in your television set, in your PC, in the cloud, on your behalf.
Last but not least is this phenomenon that we like to call in our industry cloud computing. Nobody knows what cloud computing really means. I mean, honestly, everybody in the industry describes it a little bit differently, depending upon what they’re trying to sell you. But somehow the notion that we can really use — we can really write programs in a new way where the speed with which you can develop an application and get it deployed globally, managed trivially, and let it have access to all of the world’s people and all of the world’s information, that’s a powerful notion.
I’ll give you just a couple things to think about. We’ve done a project with the University of Washington around our Windows Azure cloud computing infrastructure. They’ve set up an application that collects oceanographic information for research institutions around the globe. You wouldn’t write that in the old days. Every one of those institutes would have had its own data. Now it’s easy to think about collecting petabytes of data, bringing them together, and letting scientists collaborate to understand what’s going on beneath the ocean.
And particularly when you think about things like the tsunami and natural disasters, this is really important stuff, but it’s bringing together data in new ways and applications that get written and deployed with fundamentally lower cost and lower complexity than ever before.
I’ll give you another example to think about. We launched a service last week in collaboration with Facebook where we’re integrating the best of Bing and the best of Facebook. What does that really mean? It means when you do a query, you ought to be able to understand what your friends think about something. Say Spanish restaurant, normally you get back a set of 10 blue links and you might get a set of ads for Spanish restaurant, blah, blah, blah. Probably what’s most important to you is what Spanish restaurants your friends have checked into recently or which they like or don’t like. And what we’re seeing is this ability to mix and match information about all the world’s people and all the world’s data to build a new kind of cloud-based application.
So, there are a number of things that I think are just going to enable huge new opportunities, and for everybody sitting in this audience, it’s an opportunity. It’s an opportunity to start your own company, it’s an opportunity to work at an existing company, it’s an opportunity to stay in India and create software for the world or travel abroad and work in companies hither and yon.
Certainly, at Microsoft we’re working hard to seize on these technologies, to create new products, new products that help make it easier to automate and operate business processes, new products and services that make it easier to learn and consume information, to analyze huge volumes of information and take action.
I talked about the oceanographic work, but there’s a whole general theme of modeling the physical world in the virtual world, and speeding up the pace of scientific exploration in the environmental area, the energy area.
I noticed there’s, what is it, the Max Planck — it’s the Max Planck Institute I think here at the IIT Delhi. I didn’t have a chance to ask our people what they do, but if you’d asked me to guess, it’s probably some kind of computational physics or something where scientific data can be used in new ways to model the physical world in the virtual world.
Creating and collaborating on information, that’s kind of, you could say, where the PC grew up, and yet the ways and which that’s changing in the cloud is enormous.
Enjoying and socializing with others, this is an area we really push in Xbox and we’ll show you a little of.
And last but not least is the notion of communication. Since I’m here a couple of weeks after we announced that we’re acquiring Skype, suffice it to say that I think that the opportunities to innovate in the ways people communicate are quite high.
The way the audience responded when I came onstage, I think you’re glad I’m here. I’m only sort of glad I’m here. I’m glad, of course, to be at the IIT in Delhi, and I’m sad that we don’t have virtual technologies where you would say it would be just as good if I showed up virtually as showing up physically, right? It’s just not as good still. You want real people to show up. Today, the scenario we have to invent is where it feels as good to participate in kind of virtual reality as it does to participate kind of in the real thing.
I thought what we might do is show you a little bit of a product that we launched last year, but it brings together some of the notions of natural user interface, some of the notions of social interaction in the cloud, and that’s our Xbox and Kinect gaming environment.
Some people particularly will say, well, gaming, not serious. I actually think what we’re going to see is a lot of pioneering of new technology will start out in the entertainment space. High-end graphics has come from the entertainment space and now moves into the mainstream of computing. Next generation social interaction, communications, natural user interface, vision and voice recognition, people are a lot more tolerant of kind of new things that mostly work when they’re being entertained than when they’re trying to be productive.
So, we thought we’d show you just a little bit of some of the things that we’re doing with Xbox and Kinect. So, please welcome onstage two of the Microsoft employees here in India. Sonali Bora and Aditya Mohan will do a small demonstration. (Applause.)
STEVE BALLMER: Two or three or four things to notice. No. 1, the camera and sensor recognized and took action. One of the things we showed you in the little video is we’re getting better at recognizing faces, smiles, facial gestures.
Sonali and Aditya were both here. The fact of the matter is they could have been in different parts of the world and still played the same game together because of the social characteristics and sort of the cloud back-end, if you will, which is really kind of interesting and important.
The number of new applications you can bring to this is amazing. One of the ones I’m most excited about is actually the one that keeps track of the number of calories you burn. Clearly, I would say Sonali is doing a little bit better job right now in working herself into shape. We’re going to get Aditya working on that. She did a great job.
But it is interesting, when you think about even things like home health care and what this can mean outside the gaming environment in terms of giving people feedback and letting them interact, we’re really, really excited.
I think I said, we sold 10 million of them in the first 60 days. It was the fastest selling consumer electronics device in history was actually the Kinect, and it was based on a bunch of work that was originally done in the Microsoft Research labs. The core technology for vision and recognition is still being pioneered and pushed forward in some pretty amazing ways.
So, we thought we’d share it with you. Some of you have seen it before, but thinking about it through the computer science lens and where things are and where they’re going I think is quite remarkable.
There’s a broad set of things we’re working on at Microsoft. I talked about some of the scenarios, but whether it’s search or mobile devices, the PC, the desktop, the move to the cloud in terms of our core server products and interaction techniques, as a company we feel fortunate for all the great success we’ve been able to drive, and yet our industry is one where you have to constantly be moving forward. We’ve got great competitors. We have a lot of work to do to stay ahead, to push ahead, to catch up in some areas, but in all of what we see we see the opportunity to create and innovate, and that’s the opportunity that I think brings us all here today.
We see a lot of great things not only going on around the world by engineers who were educated in India or grew up in India, but we also see a lot of interesting and new applications here in the Indian market.
The one that our guys were most excited to tell me about when I landed this time is a cloud application built on our Azure platform from — they’re called SportingMindz. It’s used by the Indian Premier League. The Royal Challengers apparently pioneered this thing. I’m no cricket expert, but I was reminded again of just how enthusiastic Indian people are for cricket.
This is a huge application with literally terabytes of information, videos from all the cricket matches around India. It’s used by players and coaches to comment, to critique, to analyze, and literally all of this is kept online for sharing and storage through some fantastic work that SportingMindz did here.
So, you can see these applications, you can be part of creating them and inventing the future as you transition out to the workforce here in India and around the world.
I had a chance about two weeks ago to give a graduation speech at the University of Southern California, and I don’t give graduation speeches, I give technology speeches. But what I wound up doing was thinking about one of the most important lessons that we’ve learned in our business and what does that mean for graduates, and at the same time, what are the most important things for graduates who want to come in our industry and make a difference.
And I wound up kind of distilling it down to three things: No. 1, great ideas matter. A lot of what you read and hear about in the popular press will have to do with working hard, driving, speed. The most important thing you can do in business, the most important thing you can do in technology is pick the right thing to do, dream the right dream, invent the right idea. Working hard, working fast is all essential, but when Bill Gates and Paul Allen started Microsoft, they had the right first idea, that the personal computer, the microprocessor would unleash a new generation of computing. And whether you look at us or the guys we partner with or the guys we compete with, there’s some underlying great idea that underpins Google, Apple, Microsoft, Oracle, VMware; no matter who you pick, and no matter what you do, great ideas can matter and do matter in our business. And there’s no monopoly, there’s no restriction on who’s going to come up with the next great idea.
No. 2, and I say this particularly to all the students in the room, find something in life to be passionate about. It’s not easy to find your passion, but you have — at least in our industry, if you don’t come to work every day fired up and excited, and really wanting to make a difference, it’s impossible to do great work.
In my own case I had like four things I thought I was going to do with my life before I came to Microsoft. I was going to be a Ph.D. mathematician and researcher, and I had decided I basically didn’t have a good enough attention span. I thought I was going to be a business guy selling brownie and cake mix for Proctor & Gamble. Well, that fizzled out. I thought I’d go into the movie business. (Laughter.) That was fast-paced and moving quickly. Bad idea. Then I thought maybe I’d go to work for Ford Motor Company like my father. Eh. And then I just got lucky. I dropped out of business school because I liked Bill and I thought he was the smartest guy I’d ever met, and I found a real passion for the kind of work we do at Microsoft. So, try a few things, and find your own passion.
And then No. 3, and this is the one I want to emphasize perhaps the most is you have to be tenacious. It is almost certain that over the course of your career, your lives, there will be something you find that you fail at the first time or you don’t get right or you don’t figure out.
Almost every great business in our industry didn’t succeed immediately. There’s some myth in the technology industry all good things happen with two people in 30 days. Bill Gates and Paul Allen started Microsoft in 1974, and yet in some ways I think you would probably say it wasn’t until about 1995, when Windows 95 got released, that the PC really took off.
The Google guys were at it for about 10 years before the search engine really took off.
Apple almost went bankrupt before it came back.
Tenacity: Stay at it, stay at it, stay at it. And tenacity is different than passion. Passion is about being excited, tenacity is about patience and long term optimism.
And so to all of the students who are in the room today, technology will be a great choice. Find your passion, be tenacious, but think and dream and have the right ideas.
I’ll look forward to some questions and discussion, but I’ve really enjoyed the chance to speak with you today. Thanks. (Cheers, applause.)
MODERATOR: Steve, what we did was — in the interest of time, we have asked the students to submit the questions online, and they have submitted several questions over the past few weeks. Thank you all for those of you who have submitted the questions online.
We have seen some consistent themes emerge out of the questions that were submitted online. So, what I’m going to do is ask some of those questions, to the extent we have time, to represent their interests.
The first question is, what’s next for Microsoft, advances in search, robotics, communication, or what it is? What’s next?
STEVE BALLMER: Yes, all of those things and more. No, I hope you got something of a sense out of my remarks, but as we sit here — and I think this came up in the discussion with faculty — there really are sort of two different things we have to do as a company. There are areas where we’ve had great success and we really had to remake ourselves in order to sustain and grow that level of success.
PCs are going to have to be reinvented, productivity and communications has to be reinvented, the back-end of computing, where about 75 percent of Windows Server, has to be reinvented for the future.
At the same time, there are a lot of areas where we have not yet had the success we want to have. So, what is the future of mobile devices? We’re working on it. How will you interact with these devices? How will we use the technologies that we talked about in these new and varied ways?
The third thing we get to do is to try things that are just completely different. I noticed whoever posed the question asked about robotics. We had a whole team of people who are really excited about robotics. And yet the timeframe on when we could really dream up a robot that does something that’s useful, that comes for the right price and does the right things, there’s still a lot of questions. And then there’s a whole host of things that I know people are trying to invent, to try, to guess for the future.
This whole Facebook thing is a good example of this. I was an undergraduate, as Prasad said, at Harvard. And when you get to Harvard as a freshman, even as old as I am, the first thing you get handed on the first day of school was something called the Facebook. And the Facebook just had a picture of all — I still have my freshman Facebook. It’s got a picture of me with hair — (laughter) — Bill Gates still looking about six years old, as he did until he didn’t, but it had a picture of all the kids in your class with their names and schools. You know, somebody had to then conceive that you could not only automate that, which is not hard, but to turn it into something very different.
So, the number of new things that will be dreamed up by people at Microsoft and others that are unrelated to anything we’re working on today, I think that, too, will be an amazing opportunity for us and others.
MODERATOR: Great, thanks. And looking at the talent that has come here, Steve, I do believe there’s a lot of innovation that can come out of this group. And the question that strikes home is, could there be some interesting areas, research areas that you would recommend that students should pursue to drive such innovation, the compelling innovation from India?
STEVE BALLMER: Yeah. No, look, I tried to highlight the new technologies that I think are most important, but the thing I would suggest is you sort of mesh that with your passion. If you have a new idea, try it out. If you have a passion for one of those technology areas, try it out.
We were looking through the list of research topics that researchers here and professors here at IIT Delhi are working on: computer vision, graphics, the applications of computing to help with some of the challenges that we see in developing countries.
I was talking to a government official today about just sort of water and quality of water and how we use science and technology to speed that.
Bill Gates was on a panel, we had a conference in Seattle last week with CEOs from around the world, and he and Vinod Khosla, who I think also might have been an IIT Delhi graduate, they were all talking about using information technology to speed the pace of understanding of energy and the environment, because the truth of the matter is the world is not going to use less energy in the next 50 years. We want the world to use more energy. That’s how more people will ascend out of poverty. But we need clean and good energy. And the best way to experiment on the physical world may well be first in the virtual world.
So, there’s a lot of areas where I think research can be done, but you’ve got to find one that’s passion — that you have a passion for.
MODERATOR: I think I’m running out of time, but one last question, so that I represent them. Would there be a mindreading version of Kinect in the future that we should expect?
STEVE BALLMER: Yes. (Cheers, applause.)
The funny thing about this question — (music).
Wow, that was a mind-reading music thing or something, I don’t know. (Laughter.)
The funny thing about this actually is one of our marketing guys the other day came in and showed me the idea, they said, we should call Bing a -reader. I said, “What do you mean?” He says, isn’t that what we dream of? We dream — and I went through some of the scenarios — we dream that these devices will anticipate mind where we are, what we’re doing, and help give us the information that’s important to us.
And I think that’s going to be possible. I think that’s possible over the next — what am I — what do I want to watch tonight on television? Who do I want to play with, who do I want to interact with?
Yeah, I do think over time we’re going to look back and say, in some senses technology does help me read my mind. And I know that’s a little farfetched and crazy, but whoever asked the question, you’re as farfetched and crazy as I am, or as the marketing person who pitched the idea last week, or as a lot of other people will be in a lot of other ways. And I think it takes a little bit of craziness and a little bit of dreaming in order to come up with the kinds of ideas that are going to continue to transform society.
And I certainly wish all students here the best. I encourage you to contact us when you’re looking for jobs. We need the best and the brightest at Microsoft. SteveB@microsoft.com is my address, if you don’t have any other way to reach us. Please ping me. And in the meantime, I wish you all the best and all the great success in finding your own way, your own ideas, and your own passion.
Thanks a lot. (Applause.)