On October 6, U of T’s Department of Computer Science hosted a special lecture by Craig Mundie, the Chief Research and Strategy Officer for Microsoft. His lecture, entitled “Converging Worlds: A New Era in Computing,” addressed how technology is poised to solve the world’s most complex problems. The Varsity sat down with Mundie to chat before the talk.
The Varsity: Your event poster mentioned that new technologies that are defining the era of computing will help solve some of the world’s most complex challenges. Are there any specific challenges?
Craig Mundie: One that I worry personally about, and where I think these things may afford us some direction, is providing health care or education for the rest of the world to some extent. The planet today has six and half billion people. Fifty to 100 years from now it will have somewhere between nine and ten billion people… And so, how are we going to scale the current models of education and health care when we can barely do them for the rich people? You know, to a planet that has another six or seven billion people who don’t get any?
I think the answer to that is to realize that the current models won’t do it, but that we will mass produce computerized tutors and doctors that will essentially lower the cost and increase the quality of the care that can be given to those people. So that’s just an example of where these things will come together.
TV: Interesting. There has been talk of the possible threat of cyber wars. How practical is the possibility of policing the Internet?
CM: Well, there’s many levels to the Internet. The Internet rides on a physical infrastructure of wires and airwaves. Those things are already regulated by at least each country they operate in. There’s a bigger challenge of how we create some global governance model for these things, that I don’t think anybody has a good idea about yet, but that may have to emerge.
On top of that, we build all these different kinds of applications. I think, as is the case in any other technology that emerges and that society becomes dependent on, eventually the legislative bodies have to decide where they think the balance of interest lies between something that appears to be a threat and something that appears to be a good.
But whether it’s nuclear power versus nuclear weapons, almost all these technologies have some kind of dual use when you take them to the limit. I just think we’re creating a whole new genre of technologies that you don’t know where to draw those boundaries yet, but we’ll probably try — as we have with everything else.
TV:But what about having a licence to use the internet? I just mention this because there was a quote from you suggesting that people should have some sort of a licence.
CM: I was actually misquoted. I made a comment about this where I was talking about [how] we expect, in other parts of our lives, for people to have credentials. And the point I was trying to make is, in the US or Canada, driving [a vehicle] is a privilege, it’s not a right. You’re given that privilege by being given a driver’s license. And actually, you have some obligations that accrue to you if you’re the owner of the vehicle. You probably have to have it inspected and maintained and have a license plate and all these other things. Why do we do that? It’s because, if you don’t do that, your car and your ability to drive it are a threat to all the other drivers.
The problem we have in the Internet today is we haven’t [reached a decision about licensing]. It’s now becoming clear that machines that aren’t maintained and the people doing the bad things or who steal your identity — they’re a threat to all the other people. That’s how it’s becoming clear, but we haven’t figured out what the equivalent of a licence plate for your computer is or a driver’s licence for you might be, in order to begin to get a handle on how we want to control these kind of threats — where an individual’s action can be a threat to the global Internet society, if you will. So, I don’t know how that will work out, I don’t think that it will be exactly like the car analogy. But I had offered that as a way to try to get people to understand there is an emerging set of problems.
We’ve seen these kinds of things before. When the world had the horse and buggy, and people invented the first cars, legislators hadn’t already decided you needed to have a licence plate for your car and brake inspections and everything else. That only happened when everybody got them and started bumping into each other. I think … in some ways, people are starting to bump into each other on the highways of the Internet and we don’t really know what we want to do about that yet. We don’t know how bad the problem is. Some of them we know are getting very bad.
The level of criminal activity on the net is very high. The fact that there are no identities on the Internet: this was actually the comment I made in Davos [Switzerland]. If you go down here to the National Bank of Canada and you just walk in the door and say, “I’m an anonymous person, I’m giving you no credentials, but I want to go in the vault please,” they wouldn’t let you in, right? But on the Internet we haven’t created a way you can reliably say who you are. And so, I think there’s a lot of things that are going to have to come. Society is becoming completely operationally dependent on this infrastructure, and we’re still trying to figure that out.
TV: There seems to be a patent war between Google, Microsoft, and Apple. Do you see this alleged war hindering innovation in the technology market?
CM: Well, one, I’m not sure I would describe it as a war. This is a normal business process; it goes on between companies, especially big ones, but even big and small companies on a global basis. It becomes particularly evident when you see a lot of new activity in a new space, where there’s a lot of new patents or a lot of new players. So this is not a new thing, it’s just a repeat of things that have happened countless times in other fields, and even in the computer field in the years gone by. So that part’s not anything new.
I think that we’re big believers in intellectual property. It was created with the idea that if you’re willing to exchange knowledge with the society about how you do these things, the society grants you a property right [to that knowledge] for a temporary period of time. It was that trade-off in the large, we think, [that] creates a more innovative environment because everything isn’t necessarily hidden. That’s the bargain that society decided to cut with inventors. This is just the ongoing process of that. These things stabilize fairly quickly, either because companies will go out and invent new things or they’ll buy other companies or merge patent portfolios, and you see all that present in the case that you described now. So, for business people, we think that this is all part of the normal activity, and [in] no way is really stifling innovation.
TV: Technological trends significantly alter how education is delivered; will expectations from technology re-route the structure of post-secondary education?
CM: I think that colleges and universities are already adjusted by the use of technology. I would bet that most of the students here have or use a computer, and probably they all have a smart phone today or will have one relatively quickly. And once that kind of infrastructure diffuses broadly within a community, inevitably, you start to use it for things that you just didn’t do before. So I think that there’s some operational aspects of the classical education systems that have been automated, like any other business automates things. The posting of grades, the giving of tests, the distribution of homework assignments, whatever they might be: most of these today are all funnelled electronically.
I think that the more fundamental transformations haven’t really started yet, and in fact, probably will require some of the stuff I’ll talk about today, where you just can have a higher level of semantic interaction with the machines and the network, and they become less of a tool and more of a helper. When I talk about, for example, health care or education in the rural poor environment, where there is no teacher and there is no doctor, putting a computer there that presents itself as an avatar or even a robot and that knows how to do diagnoses and give you medical information or becomes a personal tutor — those things, at any level of education, would be transformational.
Look at some of the things that are happening online, like the Kahn Academy work, that just grows virally in terms of its adoption, and many students are using it themselves to supplement the curriculum that is being presented to them in a traditional classroom environment. The people who are studying [this] are finding that, actually, it may be appropriate to completely invert the model; the classical model where the professor gives a lecture and the students all watch it (and then there’s some [requirement that] they all move at some common pace of learning), may be actually the wrong way. Maybe the right way to do it is that all of the basic training is done through personalized consumption of these lectures, with real-time monitoring of your understanding and a dynamic path through the curriculum in order to maximize your understanding. The faculty is there to provide personal consultation when there is some element of this you just don’t comprehend. That would be a complete inversion of the current model. But certainly people are starting to see interesting results along those lines.