Microsoft co-founder Bill Gates has thoughts on a lot of things, from the demise of lecture halls to the awesomeness of the patent system. He took the stage at the company’s annual Faculty Summit — a two-day affair where researchers get together and talk about computer science — to answer questions from former Microsoft Research head (now part of the operating sytem team) Rick Rashid and the audience about all of them. Here are the highlights.
On the disruption of American education
Gates talked a lot about the issue throughout the Q&A session, and his hypothesis is simple: Education in the United States is broken — it has the highest higher-education dropout rate among rich countries — and MOOCs can help fix it. In fact, the Bill and Melinda Gates Foundation has invested a lot of money into the education field (to the chagrin of some experts), including strong support of Massive Open Online Course, or MOOC, startups such as the Khan Academy.
Gates acknlowedged during the session that some of his work might have unintended, negative consequences, but not this one. “In the education space,” he responded to a question from the audience, “I frankly don’t see that much of a downside.”
Online courses can give students access to new areas of study that can align their skills with high-paying jobs. They can even help physical institutions personalize student learning through gathering data about attendance, engagement, real-time understanding of the subject matter and other things. They can give those few elite minds responsible for such great inventions even easier access to new knowledge.
But, Gates acknowledged, we’re also a way out from online education achieving its full potential. We need to develop better understanding of what makes a good online course (“just sticking a camera in front of someone … who has a captive audience [won't cut it]“) and how to replicate non-lecture experiences like lab time and study groups. We also need to figure out how to supplement the cognitive and social development that comes along with attending school in person (although, he noted, MOOCs might also be able to help teachers focus on these things).
“We’re at the beginning of something really quite profound,” Gates said, “even though the temptation to oversimplify it is really quite great.”
On the greatness of patent law
“Thank god for commercial software,” Gates told an audience member who asked about the disconnect between Microsoft’s historically proprietary nature and all the charitable work Gates now does.
Intellecual property in developed countries pays salaries and lets software companies, pharmaceutical companies and others actually be able to invest in the innovation that helps improve our world, he explained. Then, when organizations like the Gates Foundation are doing work in undeveloped countries, pharmaceutical, IT and agri-business companies can afford to give away their work for free. (A skeptic might say that’s like robbing from the not-so-rich to give to the poor.)
“Anybody who thinks getting rid of [patent law] would be better … I can tell you, that’s crazy,” Gates said. “My view is it’s working very well.”
On the golden age of computer science and the stagnant state of computer programming
Gates told the audience we’re in a “golden age of computer science” thanks to the nearly limitless amounts of power and storage at our fingertips. Everything he talked about, from MOOCs to climate models to personal assistants, is benefiting from this and will continue to do so.
Consumers will also benefit, Gates told an audience member who asked about the gap between how computer scientists use computers to automate their lives and how most people don’t really know how to use them effectively. Yes, many people still don’t know how to use Office to its fullest potential (apparently, a little dog popping up with suggestions didn’t help, Gates joked) or perform efficient searches, but more data and more-powerful devices will let computers bridge the gap between what software can do and what its users can do.
“As everyone gets, essentially, what we’ll call the personal agent … I think we will be more connected,” Gates said.
However, he cautioned, the world of programming probably has to evolve if we’re going to accomplish some grander goals such as large, complex systems spanning entire industries. There are more programmers and they’re better than they were 10 or 20 years ago, but there is no objective metric by which someone could say the state of the art has significantly improved. Things have changed, Gates said, but there’s still serious work to do on knowledge representation and logic representation, among other things.
On the flexible definition of “robot”
An audience member asked Gates about a past prediction on the personal robot revolution and how that’s coming along. “The word ‘robot’ can be interpreted very broadly,” he replied.
In some ways, we’re already seeing really advanced robots in the form of farming implements that can detect the difference between weeds and crops and apply herbicide accordingly to each individual plant. In 20 or 30 years, Gates predicted, maybe robots in remote areas without a lot of doctors will be able to perform C-sections.
“We’re always on the verge of ‘When is the breakthrough going to happen?'” he said, and current advances around machine learning, image recognition and other things make it seem that will be sooner rather than later.
On the difficulty of modeling climate change
Machine learning, big data and general advances in computer science have been great in certain areas — his foundation is doing much work with stochastic models around the spread and evolution of disease — but climate change is a tougher nut to crack, Gates explained. Climate models tend to be closely fitted to historical data, and it’s difficult to predict the effect of minor variations in systems as complex and multifaceted as ecosystems, oceans and forests.
“It’s still not nearly sophisticated enough compared to the phenomena you’re trying to model,” Gates said. “… They haven’t been able to resolve some of the basic uncertainties.”
On getting rich people to part with their money
Overall, Gates said, “It’s a work in progress.”
Still, there certainly are shining example of wealthy individuals willing to donate their money. Gates pointed to Warren Buffett, who contributed half of the Gates Foundation’s annual $4 billion budget, and the other billionaire members of the Giving Pledge that have promised to give the majority of their wealth to philanthropy. Even mega-yacht builder and Oracle CEO Larry Ellison is on that list.
And interestingly, Gates said, rich individuals in China tend to be more generous with their money than those elsewhere because so much of that wealth is first-generation wealth. There aren’t ruling-class families who consider themselves dynasties, but rather people who recognize the ridiculousness of one person accumulating so much money so fast.
On the world’s most-pressing problem
Asked about the most-pressing problems in the world and the ones that scientists can best help solve, Gates pointed to nuclear and bioterrorism as the thing we most want to avoid — but not the world’ biggest problem.
That, he said, is probably the “ongoing disaster” that is 7 million children a year dying. He took some credit for getting it down to this number from 20 million when he was a kid and 12 million when the Gates Foundation began, citing new vaccines as a major cause for the improvement. In several years, he predicted, the number of children dying each year should be down to 3 million.
Yeah, political disfunction, unemployment and war are all important concerns. So is the fact that malnourishment and other environmental factors have reduced the average IQ in sub-Saharan Africa to 82. But, Gates said, “Childhood death gets pretty high up for me.”