Share on Facebook Tweet on Twitter Share on LinkedIn Share by email

Microsoft Research Faculty Summit 2006

Transcript: Meeting the Technical Challenges of the Future

External Research & Programs Overview
Microsoft Conference Center, Redmond, Washington
July 17, 2006


Harold Javid, Faculty Summit Chair, Microsoft Research

Sailesh Chutani, Director of External Research & Programs, Microsoft Research


Richard Newton, Dean of the College of Engineering, University of California, Berkeley


Dan Mote, President, University of Maryland

Craig Mundie, Chief Research and Strategy Officer, Advanced Strategies and Policy, Microsoft Corporation

Richard M. Russell, Associate Director, Office of Science and Technology Policy, Executive Office of the President

Lucy Sanders, CEO, National Center for Women & Information Technology (NCWIT)


ANNOUNCER: Ladies and gentlemen, please welcome Faculty Summit Chair Microsoft Research Harold Javid.

HAROLD JAVID: Well, we want to welcome all of you to the Microsoft Research Faculty Summit 2006. This is our seventh annual faculty summit, and we're so excited to have all of you here. We're also very excited about the program this year. We've chosen the theme of technology at the center of transformation, and we've written it in all of the various collateral, and e-mails, and things you've received in many different ways. But, fundamentally, we wanted to concentrate this year on the topic that many, many times technology such as the ones we're involved with in the computing area are key to transformation, whether it's socioeconomic or in the scientific realm, or many other realms.

As you will see as we go into this plenary session as well, we'll be talking about the future of that technology, and a lot about the concerns that people have, and try to look for what we can take away from that as we go into the future.

So, without any further ado, I would like to get the session underway. Sailesh Chutani is a director of external research and programs here in Microsoft Research. He would like to give you a brief update about some of the activities and the plans we have, things we've done last year, things we'll be doing going forward. He's been instrumental in helping us modify and come up with new visions. So, I'm sure he, as well as all of us, are looking for your feedback in this area. So, without any further ado, Sailesh Chutani.

SAILESH CHUTANI: Good morning. Good morning and welcome. It's a real honor and privilege to host you today, and I appreciate you taking the time to be here. What I thought I will do, very quickly, is talk to some of the high level goals for external research and programs, why do we exist, why do we do what we do. So, our goals are fairly straightforward. We want to partner with you, academia, and the governments across the world to foster innovative research, to invest in education, and promote science and engineering, because ultimately these are some of the core things that drive knowledge economies of the 21st Century.

And during the next two days, we will be talking about the details of how, exactly, we do that. Now, in these goals, we are driven by sound business considerations. If you think about what are the foundations of knowledge economies, you need a talented and educated workforce, you also need significant investment in fundamental research, and you in academia are at the forefront on both of those issues. Today, at the Summit, we have brought together about, I would say, 350 top leaders and influentials, not only from the United States, but also from other countries. We have six continents represented here. And our goal today is to stimulate a conversation on how we can all partner to shape the future of computing. As I mentioned, we believe that computing is fundamental to the knowledge economy, economies of the 21st Century, you are all the key players, so we want to stimulate the conversation around that. Again, during the next two days, you'll have a chance to see some of the results of the collaborations we have put in place in the last few years.

The last year's Academic Summit for us was a very pivotal year. We sounded the alarm on some of the key developments, and you're aware of that because you're on the ground. We're very concerned about the crisis in the pipeline, the drop in enrollment. We're also very concerned about the situation with the R&D funding, which has been going down. I'm sure you know the pain that comes from NSF's rejection rate being almost close to 90 percent, and the difficulty of getting any cutting edge, high risk research funded. We're very concerned about that. We think those trends have a long-term implication for the industry, and also for the competitiveness of the United States.

Bill Wulf, the president of the National Academies of Engineering, was here at the summit last year as one of the keynote speakers, and he made a very impassioned plea for reversing these trends, about investing more in research, and really getting more talented people attracted to science and technology as a discipline.

So of course since we sounded the alarm and we put forward our programs quite a few developments have come about. The National Academy of Engineering came out with the report about rising above the gathering storm, they talked about the issues of national competitiveness, and impact of reduced R&D funding, and makes very specific recommendations both to the government and to academia, how to wise up to those issues.

President Bush announced the American Competitiveness Initiative and the computing mobility mobilized, as well. I was at the CRA meeting a couple of weeks ago and it's very clear that computing research has mobilized. They see real issues if the trends continue, and they're beginning to lobby and speak up, and mobilize the community. Clearly, more needs to happen. We need action at the local level, we also need partnerships, and we need to build momentum. So it's a good thing to ask where do we go from there.

To that end we have put together a very distinguished panel. I will introduce the chair of the panel, Richard Newton, who will in turn introduce the rest of the panel. Professor Newton is a very distinguished academic and an entrepreneur. He is currently the Dean of Engineering at the University of California at Berkeley. And the list of interesting things about his career is very long, so let me try to state a few of them. He's the recipient of numerous awards for research, and he was named to the National Academy of Engineering in 2004, the American Academy of Arts and Sciences in 2006, and won the Phil Kaufman Award which is the highest recognition of the EDA Consortia.

He is originally from Australia, but he has been in the U.S. fairly long, but you can still tell the accent if you pay attention. We are really delighted to have him here, and I should also mention, because he is an entrepreneur and an educator, he has a very interesting perspective of the whole cycle. So please welcome Dr. Newton.

RICHARD NEWTON: Thank you. Can I invite our panelists to join us out here, if they would. While they're coming out, I'd just like to say that I'm absolutely pleased and honored to have the opportunity to chair this very important panel at the 2006 Microsoft Faculty Summit.

We're certainly living in very interesting times, and unusual times, and I borrowed a slide from Dan Reed at a presentation he gave recently at the Snowbird Conference. And he quoted from Charles Dickens, and I saw this on the Web and I thought, boy, he's exactly right with this. It was the best of times, it was the worst of times; it was the age of wisdom, it was the age of foolishness; it was the epoch of belief it was the epoch of incredulity; it was the season of light, the season of darkness, et cetera, et cetera.

That's the kind of confusion I feel, how many of you feel that way, about kind of all the dramatic changes that are going on around us, globally, locally. We hear about a crisis in the number of students applying to our computer science programs, yet we see tens of thousands of students going into fields that are using computer science more extensively than ever before. Is it really a crisis, is it not a crisis, how can we address it, et cetera.

These are the kinds of concerns that I think we're all talking about on a regular basis, and very, very important, not only to us as academics and industry leaders, but also as a nation, as represented by our federal government. So I would argue that nevertheless, due to the sustained efforts of many people on this panel, along with many of you out in this community, we are at a really important moment in the history of the field here in the United States in particular. It's a unique opportunity, it's a moment that probably won't last too long, and I think we have an obligation to think carefully about how we take advantage of this unique opportunity.

Many, many people have contributed to the development of a real imperative around technology, science, education and research in the United States. By the way, I include the press very definitely in that mix, because that's played a critical role in bringing these issues to the general public, and moving them all forward, as well.

So with apologies to the fact that we have many of our international friends here today, and hopefully they'll enjoy the discussion, as well, we have a very unique audience here today, as well. That is, we have an audience of academic leadership at all levels, from computer science across the entire nation. And the organizers of the summit and I felt that we really owe it to ourselves to take advantage of this unique moment, this unique gathering to see if we can think through some of the challenges that we're facing here in the United States, and how we can best organize ourselves to take advantage of this very important time in our history.

So one thing that is true is that we as an academic community have worked very hard, and some of the reports up there on the screen represent a lot of that work, to sort of make a point about technology, about education, K-12, higher education, advanced research, and in particular information and communications technologies. So has industry, independently, on our behalf, but there has got to be a lot more value in us coordinating our efforts effectively and coherently, identifying a small number of really important messages and projecting them onto the players that matter in the world, the federal government, state governments, and some of the consortia that we've developed around the nation.

At the same time I would argue that there's an opportunity for government here. We are looking at some initiatives that are going to deliver significant increases in funding, one would argue not enough, I think we may all argue not enough, but certainly significant increases in funding into science and technology education and research. How can the government best leverage those investments, working with us in public-private partnerships and other kinds of matching programs, that can really leverage and consolidate those kinds of funding activities.

So they're the kind of questions we'd really like to address on the panel today. What I'm looking for, though, is not just another discussion. The way we're going to organize this panel is we've sort of divided it up into roughly three tranches. The first tranche we're going to have a bit of a conversation about some of these topics, and give the panelists and opportunity to talk about some of these important issues, create a little context.

The second tranche is going to be open to you, the audience, and we're going to have a dialogue with you about your thoughts, and things that we could do collectively that you think are going to be important moving forward. Then I want to save a little time at the end for sort of action plans, specific recommendations of things we could do together, industry, government, and academia, to actually take advantage of this important moment in our history. So that's the plan for the panel.

Now, we have a very distinguished panel, and I'm going to introduce them. On my left, across, Dan Mote is the President of the University of Maryland College Park, has been in that position since 1998. I'm not going to go through the bios that are on the Web, I'm just going to give you some updates that I think are relevant to this discussion. Dan, of course, is one of our most distinguished Berkeley alums. He started his career at Berkeley in ‑‑ graduated in '63, joined the faculty in '67, and stayed there all the way through until his current position. He did lots of important things there. Perhaps one of the most important was, led the largest fund-raising campaign in the history of any public university, where the goal of $1.1 billion ended up at $1.44 billion by the time it finished, the New Century Campaign. I mention that only because he understands how to get money and what kind of messaging is important in that role. And I think that's going to be an important element of what we want to talk about here today, as well.

His research, mechanical engineering, his research in disk drives, all sorts of things that are related to our industry, but the most fascinating area was, of course, his work on the dynamics of ski bindings and improving the quality of ski bindings for all of us here in this room. So that was really important.

The other thing I should mention about Dan, though, that he was one of the co-authors of this report, Rising Above The Gathering Storm, and I'm going to ask him later to talk about that in the context of the ACI that Richard Russell will be talking with us about, as well. So that's Dan.

Next to Dan is Craig Mundie. In June 2006, just last month, Craig assumed a new responsibility here at Microsoft, the position of Chief Research and Strategy Officer. He's working very closely with Bill Gates as Bill transitions over the next two years, in terms of his role here at Microsoft. Craig has inherited all research at Microsoft. He had, historically, responsibility for the TCI initiative, for some of the incubators, for a lot of work that had to do with sort of international development, and working with foreign governments, among a number of other responsibilities, including work with some European research organizations, as well. Now he also has responsibility for all of Microsoft Research Labs, and one of the things we're going to ask him to speak to a little bit later is to give us maybe a quick introduction to how he sees research going at Microsoft in his new role.

He also partners with General Counsel Brad Smith to guide Microsoft's intellectual property and technology policy efforts, that obviously relate very deeply to Washington and the things that go on there. And in his previous position at Microsoft, as Chief Technical Officer of Advanced Strategies and Policies, he worked with Bill to develop comprehensive technical business and policy strategies for Microsoft on a global scope.

The last thing I'd like to mention is Craig personally I know profoundly believes in the important of a strong technology workforce here in the United States, but at the same time understands probably better than any this notion of a flat world that Tom Friedman has been promoting. If you read Tom's book, I did the count, Craig is quoted more than any other individual in that book, in terms of his input to that discussion. So he obviously has that context, as well.

Next to Craig is Richard Russell. Richard was confirmed by the U.S. Senate in August 2002 as Associate Director of the Office of Science and Technology Policy, or OSTP, in the Executive Office of the President. He's deputy director of technology, charged with the technology portfolio of that office, which includes departments in technology, telecommunications, and information technology. And he also has the space and aeronautics portfolio, as well.

In addition to his role at OSTP, Richard serves a senior director for technology and telecommunications for the National Economic Council, and prior to being chosen for this position, Richard served as OSTP's chief of staff, and on the presidential transition teams, perhaps notably for the Department of Commerce and the National Science Foundation, as well as OSTP.

Doing a little research on Richard on the Web you find that he was in an Ask the White House conversation a few years ago, and one of the questions that was asked of him was, "what three technologies do you believe hold the most promise over the next decade." Now, I'm going to put you on the spot here, it's three years old, he can maybe give us an update, but he answered, let me give you four, nanotechnology, hydrogen technology, biotechnology, and of course, information technology. So I think Richard is well positioned to respect the importance of the field we're talking about here today.

OSTP is authorized, "to lead an interagency effort to develop and implement sound science and technology policies and budgets, and to work with the private sector, state and local governments, science and higher-education communities, and other nations towards this end.

Last, but by no means least, we have Lucy Sanders. Lucy is CEO and co-founder of the National Center for Women and Information Technology, an extensive background in information technology, a software developer herself, worked extensively at places like AT&T, Bell Labs, Lucent, Avaya, where she specialized in systems-level software solutions. By the way, she's also the program chair for the 2006 Grace Hopper meeting, coming up in October, October 4th to the 6th in San Diego. It's a real important meeting. I hope to see many of you at that meeting, as well.

In the context of this panel, Lucy represents workforce development, so she represents what are we doing, how are we doing it to prepare the U.S. student body for science and technology moving forward, and careers in that direction. She will argue that the kind of work she does with women, and the development of interest in women are actually just as relevant to men, and other underrepresented groups, et cetera. So we're looking forward to hearing about that.

Now, so that's kind of our panel. A couple of quick data points, I'm sure all of you are aware of the American Competitiveness Initiative, developed by Richard's office, and moving rapidly through the political process. We were very pleased to hear on Friday that the Senate approved an appropriation for the NSF portion of this, with a slightly lower than requested increase, but really in the noise. Things seem to be going very well with that initiative, and it looks like it's a great vehicle for us to sort of work with as a community.

The major elements of the initiative, just to quickly recap, are commitments to double the NSF, DOE, and NIST budgets over a 10-year period, and to make the R&D tax credit permanent. On the education side 70,000 new teachers, alternative teacher certification, foster advanced placement, improve participation in math and science, and then there are a set of workforce and immigration issues, such as expanding worker training programs and adding flexible H(1)(b) caps and reforming some of the visa issues that as academics we run into regularly.

This is sort of a plot of where the funding will go. And one of the questions I think we have to ask is, sort of is this the right approach, is kind of doubling the budget of agency X over Y years the way we should formulate this. And I think there are some interesting questions that we have to address on that front.

Another really influential document that came out around the same time, early this year, and it had substantial impact, is this national academy's publication, Rising Above the Gathering Storm. Dan, as said, was ‑‑ I'll show you here, there he is ‑‑ an author of that report. And one of the things we want to understand is where the ACI stands, relative to some of the recommendations made in rising above the gathering storm.

Other agencies that work on our behalf in the federal government are the PTAC and PCAS, many of you are familiar with those. This is a slide from Dan Reed's presentation, again, that talks about some of the reports that they've produced ,and they're about to come out with a very important report on energy, as well, later this year.

As I said, one of the questions I think we need to think about is, what's the right way for us as a community to motivate our constituencies, the U.S. public in general, motivate our legislators to get the kind of long-term commitments we need to the funding of advanced research and workforce development, K-12 science and technology development.

Should we sort of look at this as kind of an input goal, or an output goal, in the sense, as I said, should we sort of formulate is as a, let's fund nanotechnology, it's important, which is sort of more of a lifestyle than it is sort of an output, in the sense of solving a particular problem. Should we formulate these challenges that we face in ways that people can get behind, like the U.S. needs to develop technologies to make it independent of foreign oil within the next 20 years, some kind of grand challenge problem like that. What's a better way of going, or is it some combination of the two?

One of the vehicles you as a community have available to you, and it was recommended for funding by the National Science Foundation last week, is this Computing Community Consortium proposal led by the CRA. This particular proposal will create the Computing Community Consortium, a proxy organization, so it doesn't have a research agenda, it has the role of leading a collection of efforts to develop some ideas around such grand challenges as the one we talked about in terms of the use of computing and its integration into a research agenda.

Hopefully that same integration can be used to sort of motivate and inspire young people and bring more of them into the field. The organizers of this activity would say, this is the best place. If we can all get behind this, industry, government, academia, develop a common message, use this as a vehicle, this would be a real contribution in terms of our ability to take advantage of this initiative. It's not establishing a research program itself, it's establishing a research agenda.

And just my last slide to sort of summarize a picture from the proposal that, as I said, was selected for funding $6 million as of last week. That doesn't mean the money is in yet, it means it might be in at some time in the future, as NSF does these things. The CRA will be leading and coordinating this effort on the community's behalf, and there are visioning task forces, and planning groups, and organizations that would develop some of these compelling visions for the research community.

So, that's a quick introduction to our panel, and where we're headed. And now what I would like to do is begin actually by asking Richard if he could give us kind of an update on the ACI and what's happening, and what you think we, as a community should be doing about it.

RICHARD M. RUSSELL: Thanks, Richard.

As Richard mentioned, we are having some significant success in the appropriations process. And I'll touch on, because there are obviously a number of components, it's not just R&D, but for today's group and for this discussion, obviously, most people are most interested in the R&D funding. We're having great success in actually getting the funding into the appropriation bill this year. And one of the reasons that's really important is that this effort is actually front-loaded. There's always a discussion on how you ensure, once you announce a long-term vision, in this case, the 10-year goal of doubling three critical agencies, National Institute of Standards and Technology, NSF, and DOE Office of Science.

Obviously, 10 years is a long after this administration will have ended, and one of the ways we try to set that up is, we actually front-loaded some of the funding. The doubling, if you think about it, occurs at 7 percent a year, if you do the math. This year, we're looking at a 9.3 percent increase for those three collective agencies, and it's not evenly distributed. Some of the agencies are going to get a lot more than a 9 percent increase, some a little less.

But, in the House of Representatives, which has passed through the entire House both its relevant appropriation bills, that's the Energy and Water Appropriations bill that funds the Department of Energy, and the Commerce and Science Appropriations bill that funds both NSF and the National Institute of Standards and Technology, they have funded the initiative at about 100 percent, so it's absolutely what we asked for, and it's fully funded. The Senate, which has acted on the committee level for both the Energy funding and NSF And NIST funded at virtually 100 percent. I think they're about $30 million short in NSF funding. So, we are very excited that we're actually really achieving what the president outlined. Obviously, it's very much of a cooperative effort. It's very nice for the president to announce something and ask for money but unless Congress actually appropriates it, it's sort of a hollow gesture. And so, we're very excited about that.

In addition, we're working very hard to get some of the other critical components through. I think most of you have been reading about immigration and the battles that are going on, and most of this discussion has not been about the high-skill side that is part of ACI, but, that being said, it's all wrapped up together. The Senate actually has moved a very good, comprehensive immigration bill that includes most of the high skill provisions that are in ACI, and so we're encouraged by that and very hopeful that ‑‑ and the president very much wants a comprehensive immigration bill to go forward.

In addition, on the education front, the House has acted actually on a number of critical components associated with ACI, including things like adjunct teacher corps, where we'd get highly skilled people from both academia and industry into the classroom to help teach math and science. The House of Representatives passed an authorization of that program, as well as what's called APIB, a program which is to train 70,000 additional teachers, so that we can have 700,000 additional students passing advanced placement or international baccalaureate tests. That obviously requires some funding, and it requires significant teacher training. One of the goals here is to make sure that don't leave anyone behind, as it were. That we actually have people trained to be able to teach AP courses, so that students in poorer neighborhoods aren't deprived of that opportunity. So, those provisions have passed the House, and actually have gotten funding in the APIB stuff, they've gotten funding in both the House and Senate, not fully 100 percent, but a good chunk. So, both those things are really helpful.

The next step is to keep the pressure on. It's really been a joint effort. Richard was mentioning a number of other reports, in particular the most prominent one, which was Rising Above the Gathering Storm. But there have been a whole host of reports which have come out with the same basic message, which is that we really need to work to stay competitive. We're in a good position today, but there's no guarantee for tomorrow, and if we want to be the world's leading economy into the future, we really have to work at it. And because there's been so much interest, and so many organizations, everything from the Council on Competitiveness, to companies like Microsoft, Intel, you name it, virtually all the various business organizations that are associated with what we would consider sort of high tech businesses have championed this notion, and have been pushing very hard in Congress to make sure that ACI actually happens.

And that's one of the reasons why we have actually done so well in the appropriations process. It's actually very unusual to get full funding in the first year of an announced initiative. When you look at other science initiatives, like the Hydrogen Fuel Initiative, despite that it's received funding, it has never gotten 100 percent of what we've asked for. So, we're very encouraged.

RICHARD NEWTON: So, what are your biggest challenges right now? What do you worry about the most?

RICHARD M. RUSSELL: The two things that I think are really key is, one, staying focused. You know, a lot happens in Washington. And every second-third day, there's a new crisis, sometimes it's every day it feels like. And it's very hard to sustain momentum over time. And so, there are two things that we want to make sure happens. One is that this year we actually get the funding, and we get that train in motion. The miracle of compounding interest, the first year really matters. And so, we are really going to make sure that in the first year we actually enact what the president has called for, and just staying focused for the remainder of the year to make it happen. The way it works in Washington in an election year, traditionally, the year is very short in terms of congressional schedules, and usually Congress has to come back after the election to finish all of the appropriations. So, not only do we have to make sure that all the money is included now, but we have to make sure that at the end of the year, when everything actually ends up being signed into law, there's still the same momentum. And we've got to start again next year. But, I think the first year, typically, is critical. If you don't succeed in the first year, your chances of success are very limited.

RICHARD NEWTON: Craig, I know that Microsoft has been a big supporter of this, and many, many other efforts, and I would love to get your read on how this is going, and is it enough? Where are we headed there? And, by the way, I would love, along the way, if you could sort of talk a bit about your vision for Microsoft in research and so on.

CRAIG MUNDIE: Let me answer those sort of in the reverse order, because it does expose the way I think about not only where we are and where we're going, but how it relates to this question. When you talk about Microsoft and you look at the scale that we operate in, the balance that the company has between its domestic operation as a historic U.S. headquartered company, and yet the fact that we make well more than half of our business outside the United States, we're always in these tradeoffs.

When you look at the questions you've posed, I find Microsoft as a microcosm of all of these issues. The question of is it a moonshot, or do you fund a lot of these just in a generic sense. We have the same challenge ourselves. And I guess my answer to that a little bit is going to be that you have to have a balance between the two. You can't get all the things you want done if you just do the moonshot, because you're not really prepared for the unexpected. Similarly, you're not creating the opportunities to support the prize. And, in fact, many of the great things that have happened, certainly in terms of business and things that get funded, they're the things that emerge from nowhere. And I don't think you get those only by the directed efforts towards engineering solutions to moon shot.

So, when I look at my new job, as we looked at how Bill would transition over two years from full time here to full time at the Foundation, we decided to break the job kind of into two parts, and gave Ray Ozzie half and me half. And the way that we thought about that is that, Ray's job is to replace what Bill has always done for the company in terms of coordinating our strategy across all the business groups in terms of technical architecture, and more or less the near-term engineering of our product lines. And so his time horizon you could think of as mostly in the zero to three to five year time horizon. I was given, in addition to all the policy and strategy stuff I had historically done, the mandate to manage research as a component of a much longer cycle of planning and research and development.

So, Microsoft Research to me is the way that Microsoft prepares itself for the future. And in doing that, we have both our moonshot programs, where there are certain things we think have to ‑‑ they're big problems, and we have to get them solved, and we attack a lot of that directly. On the other hand, we have a lot of areas where we think they may be important, and therefore we tend to fund them in the MSR world. My time horizon tends to be three to 15 years, and I think this country, and frankly I think all the countries in the world, now, in their own way, have got to come to grips with both of these problems. They have a set of near-term, more or less engineering challenges, and they have a longer-term set of problems.

The thing I think is going to be important, and the thing I probably worry a bit about when I think of this challenge we're discussing this morning, and even where the U.S. efforts are, as it relates to computer science itself is, I think we don't make an adequate distinction between the application of computing as we know it to this essentially infinite set of interesting problems to solve, and the challenge of making sure that computing itself continues to advance. And I would say for many years now, we've been tilting naturally in a direction of applying a relatively well-perfected concept of computing, and leaving largely behind the quest for some transformational model of computing. Even at that, we are now at a point where we know we're starting to hit some fundamental barriers in terms of how we manufacturer microprocessors, and they're leading us right down the path of saying, you know, within five years we'll face the single largest change in computer architecture and the related methods of writing programs for these things that we've ever faced, and yet if you look across the entire computer science community there's not that much being done to address that problem.

RICHARD NEWTON: Can you just expand on what you mean by that?

CRAIG MUNDIE: Well, if you look at the signals, I'll say, that come from Intel and AMD and the other microprocessor companies over the last couple of years, they pretty much have realized that due to heat limitations, they can no longer continue to increase the clock speed in CMOS circuitry, and yet that's still, for a lot of reasons, going to be the basis of our technology for a long time. And the implication of that is, we can't make the clock rate go up because that increases the heat. But Moore's Law, the real Moore's Law, which had nothing to do with clock speed, it had only to do with transistor density, it continues to function. So, we're going to get bigger dies with more transistors, but we can't actually clock them faster.

And so, if you want to solve bigger problems, you really are led down the path of parallelism. And yet, as we all know, the challenges of parallel machines have been largely reserved for the small community we call supercomputing, high performance computing, and, in fact, it's a hard problem to solve. And yet, everything we want to do in the future, where we want to get more performance, is going to come through the requirement of parallelism. And, we don't have great ways to do that. We know it's a much more error prone environment. We don't have languages that tend to naturally expose it. In a way, we've been down a path that dramatically favors serial programming, when the world around us and many of the challenges that we're funding with these competitive and science initiatives are intrinsically related to nature, bio, and energy, and other things, you know, in the real world they're all parallel processes. But we're trying to simulate them with systems that we don't naturally get to function that way.

And so, for those of us, I spent my middle career in supercomputing, and we know how hard these problems are, and yet we don't, as a collective industry, or even academically, have enough focus on that pending transition, and it's staring us right in the face. So, are the kind of things where I think we all have to step back and say, how are we going to not only solve the problems we can see at hand, but lay the ground work to be able to have either unexpected results, or create opportunities for people that are not readily anticipated.

RICHARD NEWTON: So, how do you see interacting with the government? Looking at this initiative, funding level increases, and so on and so forth, one of the challenges we have is that a significant amount of this funding is going to go into agencies like the National Science Foundation, and they do a fantastic job, but they're under incredible pressure in terms of the number of proposals they're receiving, the grants that they're giving, and the community in general is concerned about the size of those grants, and even if you double their budget it's not clear that that's a substantial increase enough. And then there's this other factor, right, which maybe will come to you later when Dan talks about Rising Above the Gathering Storm in terms of peer review and the tendency of a peer review process to lowest common multiple the opportunity. How do we get an investment vehicle, in your view, that is going to allow us to really address the risks associated with some of the challenges we face?

CRAIG MUNDIE: I think, in a way, we have to create a model that's a bit more like venture capital, at least for a component. The country has performed extremely well because it has this serendipitous formation of a cap allocation system that every country in the world is now seeking to replicate. It's actually hard to do. When people think that there's excitement, whether it's the kids going into fields of study, or people in business thinking they have a great opportunity, get venture capital. That's one way of funding a lot of this.

I do happen to believe that the entire process of funding some of this basic science and engineering has devolved into a risk averse environment. In fact, I put my money where my mouth is. A couple of years ago, one of my good friends is Lee Hartwell, he got the Nobel Prize in Physiology in Medicine, and runs the Hutch. My wife and I actually put up a prize for a fund-raising for the Hutch, which was called the Hartwell Innovation Fund. And it was matched. And it was several million dollars. What was interesting about that, and I didn't realize its importance when I did it until after it happened is, we gave this fund because, being an entrepreneur myself, I said, I believe in Hartwell, and I believe this guy can pick interesting things to do. And his frustration, as he often expressed to me privately was, because the really breakthrough ideas I can't get through the peer review process, and so they never get funded. And, therefore, we're always just sort of incrementing things along.

So, we put this thing together, and we gave him a fund that he has sole signature authority on. He can fund anything he wants. And it's remarkable, even with a modest sum of money, how many transformational things are now getting started just within the hutch and its related activities, because he has the ability to be a personal venture capitalist in biology. So I do think that if we could create an environment in the government where some part of this money was reserved, or where you were really picking the way business does ‑‑ business and venture capital people, they back people, they don't back ideas.

I remember when I was starting my company in 1982, when it was all done the venture capital people said to us, there are two things you have to remember, that first is, the business plan is only ‑‑ it doesn't matter really what it says, all that was important was having you guys go through the process of trying to write one, because you won't really operate according to that plan when it's done, the reason is things will change. And so he said, as a venture capitalist, you can think of this system as a binary system. It's two binary digits, people and ideas. There's only four cases, you've got good people and great people, and good ideas and great ideas.

If you look at as a venture capitalist, all the leverage is on the side where you have great people, because if you have great people and a good idea, they'll keep the thing alive, you'll get sustained businesses. If you have great people and a great idea, you get the Apples, the Microsofts, the Intels, whatever. The losing ones from a venture point of view are if you have okay people, it doesn't matter whether they had a great idea or a good idea, they won't make the adjustments to really get what you need. And I think to some extent, unfortunately, the peer review process and other things, tends to get you down more in the case of, you've got good people, but you don't have the brilliance the comes from a Bill Gates, or an Andy Grove, or the other well-known people who make these choices, they make them for themselves, they make them for their business, and they make them over a long period of time.

Somewhere we have to get a lot of the funding that's going into academia, in my mind, to at least have a shot at getting some of those decisions made, and I think the leverage that comes from that can be quite incredible. Right now business and venture capital tend to operate that way, but government and academia does not. I think if we're going to get above this rising storm, somewhere there's got to be more of a marriage there.

RICHARD NEWTON: That kind of leads to the point of the kinds of things that ultimately are going to inspire the next generation, and Lucy, you're working very actively and aggressively in this whole workforce development area, and how do we inspire young people, how do we get more people interested here in the field here in the United States. Tell us a little bit about what you've been up to, and what you think is going to be important in that regard?

LUCY SANDERS: Well, one of the things in thinking about rising above the gathering storm, and our own innovation and competitiveness certainly rests, to your point, in the people. And one of the things is, I don't know how many of you are aware of all of the statistics here, we don't have enough people. So the enrollments in computer science nationally are down 60 percent over the last four years. And that's alarming, and we think that's just a numbers issue, but think about there's also a related quality issue here.

So we just got some new data around the mean SAT math score for incoming computer science students, it's lower than English. So it's one of the lowest, only next to business, incoming business majors. I'm not disparaging business majors. I love business majors.

PARTICIPANT: The lawyers beat us?

LUCY SANDERS: I don't know about lawyers. It's about 535, and English is about 550 mean SAT. And this should cause us all alarm. So not only are we not seeing the numbers we need to do all this brilliant research and to keep our country competitive, we're not seeing the best and the brightest, and I will tell you from my work with women and underrepresented groups, we're certainly not seeing women and minorities. And that should cause us concern, again, not just for numbers, but for diversity of thought.

I think that everybody in this auditorium could agree that innovation is informed by diversity of thought, and any time you have like-thinkers in a room they're not going to come up with necessarily the best technology and the best solutions. So we see plenty of indicators. I could go on for hours on these indicators, but they cause us a great deal of concern, and I hope that they cause all of you concern, as well.

Now, after digging around in this area for the last couple of years, I was surprised to find, maybe I'm just arrogant, but that actually our profession, computer science, is a stealth profession. Nobody really knows what we do, and I'll start with Washington, just because he's sitting here next to me. You can go around Washington, D.C., and you cannot get agreement that computer science is a science. When Washington talks about science, technology, engineering and math they may or may not mean computer science. And I think we have to get the point across about what our profession is, what we do.

Yes, we're about applications. Yes, we're about core computing. Yes, we're about all of these things. It's a romantic, fun, energetic, very important discipline to the future of our nation and to the world. And we're not, we're very stealth, nobody knows what we do. I used to think it was just my mother who didn't know what we did. But, now I'm pretty convinced that the rest of the world is making important decisions about funding, and competitiveness, they don't know what we do, and that's a problem.

So one last thing to say here is, I think it's a problem, not because somebody did it to us, but because we're not taking good care of our discipline, and I mean myself in that, as well. I labored 25 years in Bell Labs, I had no idea what was going on with my discipline at all. And so we don't tend to be activist, and I think we should be. And that's not being activist about fixing somebody else, that's being activist about fixing ourselves.

There are plenty of things we can do. I think the bulk of the people in this room are from higher ed. We can start to look at intro computer science classes. We need to get away from equating computer science to just programming, and by there the logic chain moves to, if computer science is programming, and programming is moving off shore, what does that mean to my potential career in this space.

We can start looking at the image we convey, we can start to get more romance in this. I know, I think computer science is kind of romantic, at the least a lot of fun, and it's a great career. We're not getting out there, we're not talking to our K-12 kids. We're not putting it out there for people. So there's a lot we can do.

RICHARD NEWTON: Give me some examples, if you walked into a K-12 class, and you wanted to get across the notion of computer science as romantic, what kind of examples would you use, what kind of message would you deliver? How would you inspire the kids to say, wow, I want to be one of those people.

LUCY SANDERS: Well, the first thing I'll tell you about K-12 computer science education is there's very little of it, and half of the computer science classes in K-12 in this country are taught in vo-tech, and they are really about turning on computers and using computers, they're about computer literacy and they're not about computer fluency. And again I think vo-tech is terrific, but it's not computer science.

And I think the way you start to get into the romanticism of it is talk about what you do. It's a top down discussion, it's about solving problems, it's about critical thinking, it's about using computing to emulate what people need and what they want to solve business problems or medical problems or anything else. And so it's really appealing to that broader sense of what we use computing for. And I'd say instead what we opt out to is this is Java or C or something else, and I think that's just the wrong way to approach it.

RICHARD NEWTON: Yeah, there's a lot of evidence to suggest that that kind of motivation is particularly important for woman and underrepresented groups, that this kind of goal-oriented approach is important and relates back to our kind of moon shot versus discipline oriented discussion earlier. I mean, shouldn't we thinking as a community about creating some prizes perhaps or some grand challenge opportunities or a special science fair or something that kind of engages students with computing in the context of problems?

LUCY SANDERS: I think that would be an interesting thing to think about. One of the things that I believe is true is that when we start looking at it that way we're going to get our best and brightest men and women, and we will definitely attract more women I think and underrepresented groups, but we'll also start to see our brightest students, period, because that's what they care about. They need to understand the top down nature of it and also how foundational of a science it really is and how interesting.

CRAIG MUNDIE: Your point here that many people here probably have it, a few years ago, Microsoft Research in Beijing started this thing first in China and then Asia and now it's worldwide we called the Imagine Cup, which has been an unbelievable success in terms of getting kids to be excited about the use of computing to solve problems. You know, it's just been remarkable to see the energy that that process creates. And that's all just done by Microsoft and the various universities. But if that's an indication of if it was scaled up to a set of interesting problems and got kids involved, I think that there's just huge promise in that.

LUCY SANDERS: Well, and getting away from the notion again that computing equals programming, okay, because I think that that has been where the discipline has headed over the last couple of years, and I don't think it's done us good service.

RICHARD RUSSELL: Yeah, I mean, I remember Ed Penhoet, who was founder of a company called Chiron, founding CEO of that company, one of the most successful biotechnology companies to date, now running the Moore Foundation and very active in the stem cell initiative in California, at a meeting in 2002 he stood up and said very profoundly that the biotechnology revolution is not as much about biology as it is about information technology. And when I think those kinds of leaders stand up and make those kind of profound pronouncements, it really does affect people's thinking, and that's really important.

LUCY SANDERS: Well, and to your point, I mean, and everybody in this room, I'm sort of calling everybody to action here, and intend to continue doing that. When we were in DC a couple months ago, a very high up Department of Ed person said directly to a group of us, about 350 of us, that -- and this is a direct quote -- "You IT folks need to sharpen your elbows. We hear from all kinds of other disciplines; we don't hear from you."

RICHARD NEWTON: Well, if they do hear from us, they hear from kind of a random noise of different requests and --


RICHARD NEWTON: -- it's not coordinated, it's not coherent.

LUCY SANDERS: So I started working on sharpening my elbows, and I'd be happy to help anybody else sharpen theirs.

RICHARD NEWTON: So, Dan, you know, you get to kind of be the rounder of all of this conversation. As a university president, clearly you're interested in pipeline issues, you're interested in computer science issues, science issues broadly, you're involved in the report. You know, in terms of what you hear and where we're headed, what do you think we should be doing as a community?

DAN MOTE: Well, just speaking for a moment from this "Rising Above the Gathering Storm" report, it basically had two purposes, if you read it carefully. There are two things it was looking to do. One was provide high quality, high paying jobs for Americans. And the second was solve the energy problem, as it were, long term, renewable, available, environmentally safe energy source.

The quality job problem for Americans is really a serious problem. Lucy alluded to it very strongly. Most people don't see how the middle class in America is disappearing. The average wage in the last three years for Americans has dropped 3.6 percent. This is in a time of great economy, of low unemployment, circumstances that should be very, very good.

This report essentially identified that as the key problem going forward, especially in this globalized environment. In fact, it's pretty hard to see that, in fact, this won't continue, unless we take some pretty drastic steps.

The issue about of where we can be going, I would say with regard to the "Rising Above the Gathering Storm" report, there was one piece of it that didn't make itself to the ACI. By the way, I support the ACI very enthusiastically, as do all the other 19 members of the Rising Above panel.

In fact, just let me -- a little non sequitur on this. In July of last year, Senator Lamar Alexander came to the National Academies and said, "I would like to have a short list of actual items that we can enact now to make the United States more prosperous and safe in the 21st century, and I want it now, in ten weeks from now we want this report." This panel of 20 people was put together and worked on this problem essentially over the summer -- this was your summer vacation job -- and in the middle of October that report was presented, and now there's been hundreds of op-ed pieces and so forth went out on that.

Senator Alexander had in mind when he started that he would do his best to get the president to mention this in the state of the union address, so that was a strategy right from the front-end of that whole process.

So the report in that time scale is not tricky at all. It's just, as one person described it, basic blocking and tackling: K-12 education, basic research, higher education, incentives for industry, four pieces that came forward.

And not only did it get to the president's state of the union address, it became a feature, a theme of the state of the union address in the ACI. This is an unheard of trajectory to go from an idea in July to a theme of the state of the union. And even more so it's now maintained its momentum, as Richard so carefully described, so it's actually going through Congress, which is even more remarkable.

So this is not a normal circumstance, this is, in fact, in your lifetime you may never see this again, I've certainly never seen this before, and it shows the concern people have about the seriousness of this problem.

However, there were a couple of features naturally that don't quite get carried along, and one feature was a feature that Craig described basically was that 8 percent of all basic research money from the federal government should be delivered by program managers, professional program managers who can use their own judgment in terms of how to allocate 8 percent of the entire federal research budget is going to universities. So this would be in DOD, this would also be in NSF, and even in NASA and so forth, so areas that are outside of the current ACI.

That was a reflection of the Rising Above committee's agreement I would say that you did not get whiz bang, big high risk initiatives out of normal peer review. Peer review tends to give you good stuff, very valuable, from people who are well-known and so forth, but very incremental in general, not really risky and unproven ideas.

Well, that unfortunately, that 8 percent number or any number didn't make it to the ACI, so that's really I think a work in progress. It's something we need to do going forward.

With regard to decreasing enrollment in computer science but also in other parts of science and technology, by the way, this is a really serious problem, and we have to realize that people don't find value in our field. It's a value judgment people are making. And it's not just students from poorly educated inner city high schools; you go to any spiffy high school and you survey the 14-year olds in that high school, and you'll not find them selecting science and technology and computer science as their subjects, because they do not see opportunity. The biggest problem we have as a nation is the students do not see opportunity in our field. And we can complain about science and technology and math education in high school and all those sorts of things, but actually that's not the most important problem -- that is a problem, of course, we all agree -- but it's lack of perceived opportunity is the problem.

So if we want to take on a big challenge, we ought to take on the challenge of helping them see opportunity, working with industry, working with government, foreign government, foreign industry. We don't have a national industry anyway, we have a global industry.

And so I think we need to see this global picture, and see the need to create opportunity. I think Craig's discussion about the Microsoft initiative in Japan is just great, it's basically a challenge, it's interesting, it's exciting; we should do more of those kinds of those kinds of things in this country as well.

RICHARD NEWTON: Any other ideas along those lines in terms of creating that sense of opportunity? I mean, one of the things I've been absolutely blown away by at Berkeley is the fact that we started an initiative seven years ago called CITRIS, Center for Information Technology Research in the Interest of Society, and we said let's pick some grand challenge societal problems and see if they're good at motivating students and integrating the research.

DAN MOTE: And it completely blew me away, guaranteed, no problem, you'll get great results, very exciting stuff. It's just like Craig described, you won't be able to tell where it's going to go, but it will be interesting.

I think in terms of your own departments, I'd speak to my own guys, as a matter of fact, you might just talk to your students and you'll find they're not excited about it, and they see programming, as Lucy suggested, as what they do. We might start at home a little bit and talk about what kind of environment are we creating for our own students inside our own departments, how excited about -- you know, call together 15 of them some time and just have lunch with them and ask them, you know, you may get stunned about what you hear. They should see opportunity, they should see interconnections to other departments, they should see connections, of course, in Maryland's case, we have all kinds of labs and stuff around us, all kinds of opportunity.

So I think it's not just so much what do you tell the kids in some high school when you go about why they should be computer scientists or something or mechanical engineers for that matter or whatever, it really is how you create a culture that has a lot more excitement in it, and I think this is a national problem that we need faculty members and we need the leadership and political leadership and university leadership and everybody to work on.


CRAIG MUNDIE: I'd add something about this question of excitement. You know, when you talk about excitement, I think there's two kinds. There's one kind that I'll call a celebrity and the other I'll call economic. And historically we've had people chase a particular dream for one or the other reason. You have a lot of people who devote themselves to things where they know they're not going to get rich, but they believe that what they're doing is important or it's high visibility, high profile. And then, of course, I mentioned earlier the venture capital thing.

To me, spending so much time outside the United States, and then actually looking at this precipitous decline in enrollment, I think there's an interesting nexus of some of these questions. I contend, and you get guys like Dean Kamen, you know, who's a preeminent entrepreneur, engineer kind of guy, who's running the robotics for kids, you know, he points out that in the United States we don't just celebrate engineers, that there's very little notoriety, the National Medal of Honor gets handed out by the Secretary of Commerce I think it is or something or an undersecretary, but ask how many kids have ever seen it as something like really, really important. It doesn't have much notoriety.

I contend the United States, and to some extent Europe and other places, are increasingly a culture that is derived from media, mass media, and the mass media doesn't actually celebrate this. If you ask most kids when they're really young when they grow up what do you want to be, they're more likely to tell you they want to be Tiger Woods or Brittney Spears or some either athletic or entertainment celebrity than a science --


CRAIG MUNDIE: -- teacher or an engineer, or Bill Gates. Well, it was interesting, as Tom Freedman put in his book, when you go to China and ask that question, the kids, they actually answer Bill Gates.

And it is interesting in that the media that we have here, and have had for a long time, does not exist in those countries, and they don't have a shortage of people going into math and science. In fact, but more than 50 percent of kids in India and in China, given a choice, the most prestigious thing to study is math, science and engineering. And so that's creating this disparity.

Now, one of the reasons I think, and maybe Lucy should comment, that we've seen such a big decline in the last few years in enrollment and average SATs, is we've also seen a precipitous decline in the number of foreign students that are getting into the country. And partly that's the practical screw-up of the visa problem, but when you go into those countries, even the kids who can get the visas don't feel welcome anymore. And so we've seen a side effect that I think will be more persistent than even just fixing the visa problem in that people are saying, hey, they don't want us there anymore.

And that is the really, really big crisis, in my view, because given that we don't have in absolute terms the number of people, and given that IQ is nominally uniformly distributed in the global gene pool, if we can't figure out how we're going to get those people to want to come to the United States and think that there is something here that's not only welcoming, but an opportunity that is more available to them or has a higher probability outcome than it would be if they want back home or stayed home or went to another country, if we don't fix that problem, all this stuff is going to be for naught.

And I think that that is a huge challenge. And if we can't create excitement around what these things do, if people don't understand that there's nothing anymore that's cool, that doesn't come from computing as an integral part of it, if you want to make something cool, I don't care if you want to make a new energy or bio or nano or whatever it is, if you can't compute it, you aren't going to do it. And yet that message is not out there.

RICHARD NEWTON: Well, just quickly on the graduate students, the foreign students, too, I mean, not only do we do that, the ones that do get in, as soon as they finish their degree and we've invested all this money in them, we tell them to go home. But, I mean, I don't understand why we don't just staple a green card to every graduation certificate. It makes a hell of a lot of sense to me.

DAN MOTE: I've made that point; it's basically the argument I've made is that very argument, staple a green card to every advanced degree.

CRAIG MUNDIE: It hasn't happened.

DAN MOTE: No, it hasn't. I'm still working on it.

CRAIG MUNDIE: Well, Richard is going to make it happen.


LUCY SANDERS: I want to make a comment about the image issue, because I think there's something that we may be asking all of you to tap into here. A group of people in February from CRA and ACM and also Microsoft, HP, Intel and others met in Washington and talked about this issue of image. Rick Rashid has been involved with that.

And we've decided to fund a full time person to really start to look at this. We've all talked about it long enough. And I don't think it has to be a big Madison Avenue PR campaign or something, I think we can start to do some really interesting things around just going to talk to screenwriters in Hollywood. I had a meeting with the president of Industrial Light and Magic, who they do all the special effects for lots and lots of movies, and I mentioned to her the only thing I've ever seen in a movie of a little girl actually programming something, that there was a normal little girl was in Jurassic Park probably about 20 years ago at the Sun workstation -- remember this, where the tyrannosaurs were coming in. And she was all excited about what she could do to help to fix this problem.

So we need a person, it sounds like not very much, one person, but I think we can start to take this on as a group, and I just want to put a plug in for that activity.

RICHARD NEWTON: So on that vein, what I'd like to do now is open it up. We could go on forever as a panel, but I'd love to get some input from the floor. So we have three microphones that are wondering around here in the audience I believe, I don't see there but there's one in the back. And so if you raise your hand, the microphone will come to you.

And what I'd like to do is first obviously introduce yourself and your affiliation, but then sort of let's try to finish off this session with some calls to action, some specific things we can do as a community to really affect change. Here's one down here. Why don't you come around and grab this one here in the front? Go ahead, Dan.

QUESTION: The latest count I think is that of 635 or whatever congress people, there are three with engineering degrees. If you look at every other country, it's sort of the opposite kind of statistic. I don't know how far we're going to get if our national leaders don't come from this background. So what steps can we take to make that happen?

RICHARD RUSSELL: You want me to take that? A couple things.

LUCY SANDERS: Run for Congress. (Laughter.)

RICHARD RUSSELL: You know, this sounds obvious but we live in a democracy, so actually you the people choose who represents you. And so to the extent that that's important to the people of the individual district, it will be reflected in Congress.

That being said, it doesn't take -- I mean, if you look at the champions of R&D funding or trying to get the elements of Gathering Storm enacted into law, you will notice that they aren't necessarily engineers. And so I don't think understanding the issues that we face in terms of long term international competitiveness and having an engineering degree are sort of synonymous.

And, I mean, there are a couple things; I don't know how many engineers run for Congress. I mean, you have to sort of break that apart. I mean, how exciting is being a congressman to an engineer.

CRAIG MUNDIE: They're too logical. (Laughter.)

LUCY SANDERS: No, I think they should all run for office.

RICHARD RUSSELL: So anyway, but obviously the first thing you need is you need to have actually engineers running for office in the first place.

CRAIG MUNDIE: But I would say one thing, that whether you ignore the Congress or not, the people who are I'll say the permanent staff that make the government run, they have engineers, arguably they aren't the best engineers. And I think one thing you could ask the Congress, and to me a stark contrast is like Singapore, you know, Lee-Kuan Yew basically had made a decision, he said, look, if I'm going to have a government in this tiny place and we're going to stay ahead of the power curve, we need the best people in the country to be in the government. And so they made a policy choice that says they're going to pay salaries for government jobs in these areas for minister level positions in the permanent staff that are absolutely 100 percent competitive with the best you can do in industry, and they do that. And they get some of the elite people in the country into all these jobs on a long term basis and they rotate them around. The United States has not done that. And one thing you could ask is how do we solve the salary disparity problem. We were talking before the meeting that it's just a crisis.

So if you think that you want government or you want program managers in government to be able to make decisions, you know, we're not providing a compensation opportunity that even for people who want public service roles are going to be able to elect that if they have any sense of responsibility to their families, for example.

RICHARD RUSSELL: I mean, that's a real issue. There are special rules that you're allowed to apply to have salary bump-ups in the federal government, but it's still not competitive. We at OSTP like bringing in people from the outside. About half our staff is on loan in one way or another. A lot more of them used to come from private sector and quite frankly academia, and we would have a reimbursable type of arrangement. Quite frankly we can't afford them. Even academics get paid a heck of a lot more than we can afford. So it is a real issue, a long term issue.

RICHARD NEWTON: Okay, a question here.

QUESTION: I was wondering if you could comment on the feasibility of targeting applications, and like 6-12 education or higher ed, so a couple examples would be gaming or data mining in something like CS1, and then if Lucy could comment on gender issues relative to that.

LUCY SANDERS: So the question is around can we take specific applications and then do we understand if they appeal to women more than men, do I have it right? She's nodding, so okay.

I have seen some classes -- I'm not an expert on this, we're just starting to understand some of the novel introductory classes that are out there. Mark Guzdial teaches a really good one at Georgia Tech around computational media, and using the manipulation of media, video, sound, and other things, to teach the basic computing concept. And that seems to be working very well with both men and women, and it's a very popular class. It's been used in about 12 other institutions now, and we're starting to try to roll that kind of a concept out with computational media.

I have certainly read some things around gaming. I'm not as familiar with that. But I do think that taking an approach that kids already know like this is my life, I understand games, or I understand media, and then working the concepts of computing into it has a lot of merit.

QUESTION: I was more interested in your assessment of the sustainability of that, given the fact that the target applications do change over time.

LUCY SANDERS: That they keep changing. Yeah, I think that, well, it's probably something we have to keep up to date, you know what I mean? I mean, it's one of these things we can't just say this is the introductory class and then that's it, because we're in a fast moving field. I also think there's some complication around like the computer science AP test and what's that going to end up looking like. And they're due to redesign that test in the next year or so, and it mirrors what the introductory courses in computing are sort of nationally. And so taking an application specific area and trying to make that into something is going to be hard I think for the computer science test as well.

I realize I'm not answering your question exactly, because I don't think I know the answer, but I think it's worth exploring those application areas as a way to introduce kids.

RICHARD NEWTON: Is there a way of coupling more closely with industry and using industry as kind of an educational vehicle for injecting that state of the artness into the curriculum?

LUCY SANDERS: Possibly, yeah, possibly.

RICHARD NEWTON: Okay, we have another. Let's see, where are the mikes? I can't see the paddles. Does someone have a mike? There's one over there. And then could you give her the mike next up here? So you have the floor, sir.

QUESTION: Okay. I'm Todd Shum from Howard University and the University of Southern California.

And I want to make you aware of a project and program that Microsoft is sponsoring that has been very successful called the Windows Marketplace Skin Player Challenge. I don't know if you're familiar with it, but for the last five years Microsoft has sponsored it in conjunction with Howard University and University of Southern California. And what it does is it brings people into the space by having them create Windows Media Player skins. You can actually go and visit the site, it's And what you'll see there are examples of students being innovative.

But the real grind to the competition is that it brings in students in graphic arts, in communications. And one of the plus sides has been that we've gotten students from non or traditional non-CS departments to come into computer science from that program.

The other thing that we're looking to do going into the fall is we want to expand the program, but also align it so that it provides students with confidence to participate in the Imagine Cup.

The last thing that I want to say is I take a little different perspective on this image problem that engineers have, and it's because one of the things that we find is that students elect other disciplines because engineering is considered hard, you have to take the prerequisites, math, calculus, physics and chemistry, and also they have this perception that it's not important to get paid. So if you have to take chemistry, physics and calculus versus creating a hip-hop record, and you make the same money either way, then what are you going to do?

So when we look at the enrollment, when we look at the enrollment we find that there is a disproportional number of students that are enrolling in radio, TV, and film, because they think that it's easier and the compensation will be the same.

Thank you.

RICHARD NEWTON: So to that point, maybe what we need to do is talk about all the failures in those disciplines.

QUESTION: They don't see the failures, they don't see the failures.

RICHARD NEWTON: No, I know. Maybe that's our challenge.

CRAIG MUNDIE: Maybe we should publish the statistics.

You know, it is interesting, kids have genetically a better chance of growing up and being Bill Gates than growing up and being Tiger Woods, I mean, just in terms of the genetics of abilities. It's quite amazing, we should probably publish these statistics.

RICHARD NEWTON: Okay, number one.

Question: Thank you very much. I'm Jeanette Wing from Carnegie Mellon University.

I have three comments to make, and I wanted to focus these comments around action items. So the first is a bit of an immodest plug of mine. I wrote an article that appeared in CACM March 2006 called "Computational Thinking". And I tried to inspire not just the young but the field into rallying behind my vision for the 21st Century, which is that computational thinking will be a skill that everyone in the world will need to use and be using in their life, in their daily life.

But the end of the article is the action item for all of us who are in academia, and that is speaking directly to what Lucy was suggesting. We need to change or at least amend what we offer in our introduction to computer science courses. And right now most of us offer courses like introduction to Java programming, right? Well, that's fine because there's a group of students who will need that, but I think we should also be offering courses like ways to think like a computer scientist or fundamentals of computing or something like that, because when we change what we offer our freshmen, then the K-12, the AP, I think everything else will, if you will, with great optimism, fall into place.

But it is partly our responsibility and perhaps partly our fault that we are in this position. But we can do something now, we have this opportunity, if the AP people are really going to be changing, or the ACM, then we have an opportunity now. So that's action item number one for us to go home and think about.

RICHARD NEWTON: Could you be a little bit --

QUESTION: Okay, okay, okay.

RICHARD NEWTON: Because we have other people who'd like to speak.

QUESTION: Yeah, yeah, yeah, okay.

The second one -- okay, I'll skip the second one.


QUESTION: The third one has more to do a question to Richard Russell perhaps, but that is what can we who are in positions of administration and management at our universities, what can we be doing with respect to having our congressmen and VC people support the ACI? I mean, we can write to our -- what would you suggest that we actually do, or our presidents of our universities and provosts, what do you suggest we do?

RICHARD NEWTON: And how important is that?

RICHARD RUSSELL: It's extremely important, and I would say on a sort of continuum of level of importance, congressman and senators respond most quickly and most significantly to their actual constituents. There's a popular misnomer that what matters in Washington is just people who run around and hammer them while they're there; what really matters is what they hear when they go home, because it validates whatever they've heard in Washington, and they believe it's true, because in Washington people are constantly selling themselves. When they're home and they're meeting with a group and someone stands up and says, "Why aren't you supporting ACI, or what are we doing about competitiveness, or how are we supporting basic research," that really hits home. And every time I meet with a senator or congressman, the first thing they tell me is, "Gee, when I was on congressional recess last month, and I was home, four people approached me about this. Those four people are more important than the 15,000 lobbyists that were knocking on their door.

QUESTION: (Off mike).

RICHARD RUSSELL: No, letters are important as well, but --

QUESTION: Letters from the presidents of our universities?

RICHARD RUSSELL: Letters from the presidents of your universities, but I would say go ahead and target them to the congressman and senators that represent them, and go to those town hall meetings. If you can actually spend your own time, and it doesn't have to be the department, it can be students, it can be professors, actually engage in a dialogue, that leaves a real lasting impression.

And let me say one quick thing in addition. I think the other thing you might want to think about arguing very strenuously for is the Adjunct Teacher Corps. I mean, what I'm hearing here is that we need to inspire our students, and I absolutely agree with that. One of the ways they're doing that is actually getting real live people who work, who get paid and do interesting things into the classroom. That's what the Adjunct Teacher Corps is all about. And we haven't been able to get that through Congress. We've gotten it through the House, but they had to vote on it, on the House floor, it was a big contentious battle. We can't get it through the Senate.

So call that out, say, listen, we want people with experience in the classroom, not replacing the teachers but helping them convey to the students that this is a good career, this is an interesting area to be working in, you actually have a future if you go this direction.

RICHARD NEWTON: By the way, you know, a resource we see that is entirely underutilized is our retiring workforce, and so this Adjunct Teacher idea, obviously they need to be credentialed and brought up to speed in terms of teaching, but there's a real opportunity there for people who want to do that.

You know, just really to this point though in terms of academia and what we can do and industry, we talk to, there are a lot of lobbying groups that work for industry, and when we talk to them as academics we sort of say, you know, boy, don't you think science and technology education is critically important, they say, absolutely it is, and it's up there with the treatment of stock options. But unfortunately the treatment, with the exception of Microsoft, of course, but the treatment of stock options kind of comes first, and you only get sort of one short when you lobby.

Now, I would argue that there's a real advantage if we could sort of get more academics, both from the legitimacy of the lobbying that industrialists do. If they come along with a university president, I think that does add credibility to a case. And then vice versa, I would say that if we can get more university presidents bringing in, as we've seen in the successes here, industry, so is kind of the combination a really valuable tool?

RICHARD RUSSELL: Absolutely. But it is happening in this instance. I mean, that's one of the reasons why we've seen the success that we've seen to date is that you are getting a real marriage between academia and industry.

There also are a number of venues where that happens. We have our own for the president and for the advisory body called PCAST, the President's Council on Science and Technology, which is industry leaders, largely CEOs or chief technology officers and university presidents, but there's also the Council on Competitiveness, there's obviously the group that was assembled for the Gathering Storm, which actually there's a lot of -- interestingly enough, there's a lot of crossover between these people, there are a lot of the same people.

But I wouldn't discount what you can do locally with your Washington representatives. That's very important. I think nationally there's a very good effort on. I think locally that's the next step.

CRAIG MUNDIE: I just want to add one thing that I think this is something people could do. Richard mentioned that like even the Adjunct Teacher Corps, which to anybody at this level seems like a no-brainer, right. Why is it so contentious in Congress? Why didn't it pass the Senate? Answer: The teachers union. That's I think 100 percent the answer.

And so one thing you could say, whether it's the university provost or yourselves, if you go at a local level, if you want this to change, you need to convince the teachers and their unions that this is actually going to be good for them, not going to displace them out of a job, and if you do not fix the union problem, I don't care how much discussion we have in this room of this stuff, these changes will not be made in K-12 education, period.


QUESTION: Joe Rosen from Dartmouth.

So this week before we came here it was the 50th anniversary of AI, which we had at Dartmouth, and everyone came up for that and had some criticisms that seem more to what you're talking about. And next year is the 50th anniversary of DARPA, which has some of the same problems.

And this sort of reminds me of a real senior player in Washington. I was at a meeting, we talked about whitespace where you hear this talk for an hour and you ask what isn't being discussed that we're missing, which is actually the solution, so within this framework we're never going to get there, and I think the answer is Washington we used to call is disruptive technologies. So there needs to be an activity to support heavily disruptive technologies, and what I found in Washington the last few years, especially at OST and other groups, is we're veering away from that, and it's disruptive technologies, which I think really excites the kids at getting involved, not just incrementally adding another line of code to 50,000 lines of code, but looking at something that truly is going to change things. And unless we do that here, and the question is DARPA is a group, the AI conference, other groups, OST can support that activity, but I really have not seen that. And mostly maybe in the intel community there is an activity level for disruption, but this notion can now, if they're not going to be doing it, could come out of industry academia, but that doesn't have to be big projects, but they're exciting projects that suddenly attract people to join this activity, and I think that somehow is not being encouraged.

CRAIG MUNDIE: It's the missing 8 percent.

RICHARD NEWTON: So let's speak to that. Richard, you know, I mean, we've heard this both from Craig and from Dan as well relative to this sort of entrepreneurial program leader opportunity. And I think many of us who have lived through the DARPA area, as the various generations of DARPA would agree, that back when the program managers A) had a job for a long time where they could actually start a program and end it, and sort of see kind of success or failure as a result of their investment and get the satisfaction from that, and also when they had more independence and sort of rope, if you like, it was a different era, and a lot of the big breakthroughs, a lot of the people that were educated during that time, the Bill Joys, the whole Alto development, things like that that really came out of a lot of those investments in many ways, but really where we did change the field in some significant ways. How are you going to -- I mean, is there anybody in Washington thinking about how to do the 8 percent?

RICHARD RUSSELL: Actually, I mean, it's very much a topic of discussion that's built into a number of the senate bills, authorization bills that are circulating right now.

Here's the challenge, because I don't think that there is a disagreement with the goal. There are two things that typically happen with money in DC. One is that if you don't create a framework by which you can argue that the money is better spent by a group of in theory academics or industrialists with the best interests of the nation in mind rendering judgment, the money instead gets earmarked. And there's absolutely no question in our mind that peer review is intimately better than earmarking.

DAN MOTE: Absolutely. We all agree with that.

RICHARD RUSSELL: And I think no one would disagree with that. So that's the first thing.

The second thing is in terms of why it is we're becoming more conservative, and certainly we've heard that a lot, I guess one of the questions is, is there something we can do to the peer review process to change that, and I think that would be a worthwhile -- something that's worth our time spending and thinking about.

Lastly though, having a set-aside program is very difficult within individual agency budgets. So let's take NASA. So we set aside 8 percent of NASA's budget. Well, the problem is that if you look at NASA's budget, a huge percentage is going to the shuttle. You can't just take 8 percent of that and use it for high risk stuff, because it's being used for the shuttle. And you'll find that replicated in virtually every agency, with NSF being really the only agency where that's not as big an issue, because obviously they're giving out individual grants.

CRAIG MUNDIE: That's the difference between business and government is this sense of entitlement. I mean, all you're saying is that every existing program is an entitlement and therefore you can't take the entitlement programs away to fund the new stuff. And that's what business does differently is they decide what's basically end of life and shoot it and then they go on to the new thing. And I guess the question is how are we going to do that with some of these folks.

RICHARD RUSSELL: I think that's something that we need to do more work on. There's no question that one of the biggest problems, everyone -- that's why everyone looks at the marginal annual increase. We're spending $137 billion annually on R&D. This is the federal government, U.S. federal government alone, $137 billion on R&D. The problem is the only thing people care about is whether it's 138 or 139. That's the only thing that matters. The 137 is all baked, it's all used up, it doesn't matter. Well, that's ridiculous. And that code we haven't cracked yet.

RICHARD NEWTON: The last question over here.

QUESTION: Wow, the last question. Beth Mynatt from Georgia Tech.

Lucy mentioned Mark Guzdial's work and how we've changed computing at Georgia Tech, and I wanted to provide some commentary and return back to this question of what challenges do we directly face, and what comes for free when we do that.

What Mark was able to do and what we're able to see now in terms of increasing numbers, increasing diversity, and increasing quality was changes throughout the college in computing how we approach our education. That came from innovative and socially relevant research that predated those efforts five years or more by people like Mark Guzdial, by people like Tucker (Balsh ?).

The challenge is, and I think what we need to do explicitly, and then the rest of this comes for free, is how do get that innovative and socially relevant research funded. Mark -- he's not here, so I can say this -- Mark couldn't get funded for years on what he was doing, and now he's practically Saint Mark at Georgia Tech at this point. How we get that innovative research funded.

And then the second thing is, and this was an incredible challenge for us, was breaking down the political walls between the research side of the house and the academic side of the house. Actually letting your research efforts filter through and connect all the way through to the experience that every freshman at Georgia Tech has about computer science, every freshmen, not just the CS majors, every single freshmen on computer science, that now connects deeply to the research that we're doing. And that's actually not a trivial lever.

RICHARD NEWTON: That's a great model, and congratulations on everything you've done.

But again I think the point you make, which it comes back to, is this point of that funding of that innovative research. I think we've covered a lot of ground here today, but if there's one thing that stands out in my mind it's the mythical 8 percent in terms of the real input we've had.

What I'd like to do, we've got a couple of minutes left, I'd like to just sort of go along the panel and ask if anybody has anything they'd like to sort of add to finish up with.

DAN MOTE: I'll say one thing, and this thing about the Gathering Storm report was a report written to Congress, but, in fact, if you look at the areas that it addressees, K-12 education, basic research, higher education, and incentives for industry, a lot of the territory is, in fact, not in the federal government, a lot is in states and local regions, state economic development folks, qualification for teaching and so forth.

Who is going to transform that, going to take those threads now of this momentum we talked about down to the grassroots and the dirt in the local community who is actually going to do that?

The National Academy is having a program in September the 28th where it's looking to get all 50 states to show up to try to get the states to essentially take up that mantle. Maryland has already done it, by the way, in April, but I think no other state has done it yet.

So if you want to do something, and you're interested in actually trying to carry this through and see how far this momentum will actually take us, this September 28th meeting in the National Academies is one that you might bring home to your universities to see if you can participate in that at the state level, from whatever state you come from.

RICHARD NEWTON: This is a point Dan has made a couple of times, too, and I think he's right on. It's going to end up being these kind of partnerships between industry, state and federal government that are ultimately going to deliver on a lot of this promise of getting the state governments actively involved in this process through tax incentives and other things is really going to be important.


CRAIG MUNDIE: I think the other point I'll make, other than this 8 percent, is the value of creating excitement. And it would be interesting to say instead of advanced placement, can you take a job, a programming course while you're in high school, it would be more interesting to say, you know, you want to get advanced placement, you have to participate in high school in one of these contests, and you could have ten of them, and pick your favorite contest, participate in it, if you're on a winning team you're advanced placed. And you have an element of competition, the kids are doing something.

The thing that's been remarkable to watch is once you create that element of competition, the kids come out of the woodwork. I mean, you've got hundreds of kids in a university competing to be on teams of four, and then they only get one team of four that goes to the regions, and then they can beat the nationals and the global. I mean, it's remarkable. And if you couple that with the interest they have in gaming and the whole community process, think of applying MySpaces and things like that to the idea of monitoring. That's where you can get celebrities today. You don't have to be on television. Use the Internet and all of the stuff that these kids are doing, the digital lifestyle they live, to create and reinforce this competition. But you've got to have something that they want to go and participate in that's exciting, and I think that these competitions are the way to get the kids engaged.

RICHARD NEWTON: Great. Richard?

RICHARD RUSSELL: Just echoing on something that Dan just said, there are two areas that we really didn't spend any time talking about. One is the state R&D funding, and just because I happen to know it off the top of my head. The states are spending half again as much as the federal government is on nanotechnology right now, and the federal government is spending a billion dollars a year. That's almost unheard of. That's because there's interest in the area, and there are a lot of states that are really ponying up. So that's a great area, and I think we need to make sure that there's a nexus and synergy there with the federal government and the private sector.

Secondly, the private sector spends twice as much annually obviously on much closer, more applied research than the federal government, so twice as much money is being supplied by the private sector. And so that's something we didn't discuss but something I think is really critical and we can't lose sight of that.


LUCY SANDERS: Maybe just to reemphasize again the importance of the CS introductory class, and I think if there's one thing we can all do that's very high leverage it's really take a look at that and make it more interesting and compelling and relevant to students. And then once we do that, the redesign of the AP computer science test, which is they are committed to over the next couple of years, or hiring a person. I intend to have some input as to who that is to actually look at this. That starts to then help the K-12 level, because, of course, the K-12, they teach to the test. And so maybe it is a contest or something else, we need a lot of creative ideas, we're going to need your help, and so please do really seriously consider looking at your courses.

RICHARD NEWTON: I'd like to thank our panel. We've covered a lot of ground in this discussion. I think there's a lot we can do as a community, and I'd just like to reiterate that I truly believe, as I think the panelists do, that we're at a very special moment right now in terms of the field and our activities here in the United States. I hope we can really get some focus on this. And I thank you and I thank the audience for your participation. (Applause.)


Related Links
 > Events > Microsoft Research Faculty Summit 2006