Don’t Cry for Me, Argentina

While it is well hidden, wrapped in a very careful press release, Phil’s sharp eye has caught the details in SJSU’s press release about the next phase in the Udacity pilot that suggest the partnership between the school and the company is winding down. When Carl Straumsheim of Inside Higher Ed asked an SJSU spokesperson point-blank whether Udacity would continue to be involved with the courses, the reply he got was “Good question for Udacity.”

Ouch.

The Schadenfreude surrounding Sebastian Thrun’s fall from grace has been intense ever since the Fast Company article quoted the man who the author labeled the “godfather of free online education” as saying that he realized he had a bad product, and noted that his company is “changing course” to focus on corporate training. Mike Caulfield captured the tone of the reaction in ed tech circles rather nicely when he wrote:

Thrun can’t build a bucket that doesn’t leak, so he’s going to sell sieves….Udacity dithered for a bit on whether it would be accountable for student outcomes. Failures at San José State put an end to that. The move now is to return to the original idea: high failure rates and dropouts are features, not bugs, because they represent a way to thin pools of applicants for potential employers. Thrun is moving to an area where he is unaccountable, because accountability is hard.

I imagine that it would be easy for somebody running or funding an ed tech startup to draw the wrong lessons from this sad story. Consider this blog post to be an open letter to my friends at ed tech startups with some advice about how to avoid the kind of disdain and ridicule that Thrun is receiving now.

Pride Goeth Before the Fall

In some ways, it’s hard to separate Thrun’s current problems from his biography. He’s the guy who invented the self-driving car. He’s been a research scientist at Google and a professor at Stanford. He’s a competitive cyclist. And now he is building a startup that will, he hopes, transform education. He is Silicon Valley’s own Buckaroo Banzai. None of which is something that Thrun is to be blamed for. The facts of his life are the facts of his life. But it all plays rather nicely with the story guys like Tom Friedman love to tell about how some technology genius is going to build the new gadget that will blow away all those stodgy old institutions that are holding back human potential and save the world. After all, Thrun invented the self-driving car! How hard could education be?

Thrun didn’t create these narratives, but neither did he discourage them. To the contrary, he did things like making himself the face of the California SB 520 bill by showing up as a featured speaker—and sometimes as the only featured speaker—whenever a legislator or the Governor staged an event about the bill. The unavoidable implication was that Udacity was expected somehow to “save” higher education  in the state of California. Thrun was apparently either oblivious to or comfortable with that message.

And then there were the things that he has said.

Make no mistake: Sebastian Thrun is not being mocked because he said that his company had “a lousy product.” He is being mocked for the things he said before he said that. Like telling a reporter from Information Week last August that Udacity “has found the magic formula.” (Was that before or after he realized that his product was lousy?) Or like telling a Wired reporter that he thinks that in 50 years, “there will be only 10 institutions in the world delivering higher education and Udacity has a shot at being one of them.”

It is this last comment that I want to talk about in particular, because it is particularly instructive.

A Lousy Product

It may sound so far like I am suggesting that Thrun’s mistakes were about marketing, but I am not. The problem I’m concerned with is product design in the deepest sense possible. I’m talking about how you conceive of the problem that your product is designed to solve.

Suppose somebody came to you and said, “I’ve solved it! I’ve solved the problem of data.” Or how about this: “In fifty years, there are only going to be ten apps in the iOS app store, and we have a shot at being one of them.” You would think that person is an idiot. If you want to tell yourself a story of the Silicon Valley hero riding in to save education from the hands of selfish and incompetent bureaucrats and union interests (to the cheering of the huddled masses yearning to be free), then it’s probably easy to convince yourself that the reason many educators scoffed at Thrun’s “ten universities” claim is that it threatened their livelihoods. And honestly, there was probably some of that. But mostly it was because the statement was nonsensical on its face.

Silicon Valley can’t disrupt education because, for the most part, education is not a product category. “Education” is the term we apply to a loosely defined and poorly differentiated set of public and private goods (where “goods” is meant in the broadest sense, and not just something you can put into your Amazon shopping cart). Consider the fact that John Adams included the right to an education in the constitution for the Commonwealth of Massachusetts. The shallow lesson to be learned from this is that education is something so integral to the idea of democracy that it never will and never should be treated exclusively as a product to be sold on the private markets. The deeper lesson is that the idea of education—its value, even its very definition—is inextricably tangled up in deeper cultural notions and values that will be impossible to tease out with A/B testing and other engineering tools. This is why education systems in different countries are so different from each other. “Oh yes,” you may reply, “Of course I’m aware that education in India and China are very different from how it is here.” But I’m not talking about India and China. I’m talking about Germany. I’m talking about Italy. I’m talking about the UK. All these countries have educational systems that are very substantially different from the U.S., and different from each other as well. These are often not differences that a product team can get around through “localization.” They are fundamental differences that require substantially different solutions. There is no “education.” There are only educations.

But maybe I’ve gotten a little too abstract and philosophical for a practical-minded engineer or entrepreneur. Let’s get real. I went to Rutgers, a large state university in New Jersey. If somebody asked me what attributes of the “product” that was my college education were the ones that made me think it was a good value, I would give them a list that looks something like this:

• I got a scholarship, so it was practically free for me.
• They had a really good philosophy program and a pretty good linguistics program, both of which were areas where I really learned to think.
• I loved the diversity of the people who went there, which contrasted with the bucolic rural/suburban neighborhood I grew up in.
• I found a handful of teachers and more than a handful of students who inspired me to be curious and think harder.
• I was able to explore my musical side, which had no impact on my career (sadly) but a deep and lasting impact on my happiness.
• It was close but not too close to home, and my sister went there.

Nowhere on that list is “get a degree.” This is not to say that a degree was unimportant, but it’s not something that I, as a relatively privileged and relatively well-educated 18-year-old thought about. If you asked me which pieces were the most valuable, how much I would pay for each, and how I would react to “unbundled” services that met one or more of these needs, I wouldn’t know what to say. Moreover, I’m pretty sure that the working-class captain of his high school wrestling team with whom I shared a dorm room my freshman year would make a list of reasons why he went to Rutgers that would bear little or no resemblance to mine. Despite the fact that a college education has some consumer characteristics—you can shop for it and you can buy it—it is not an easily definable product. One college—heck, one class—can serve radically different needs for different students. So what, exactly, would you be disrupting when you disrupt college? What is the product that could replace what college does?

This is a big reason why Udacity failed in their SJSU pilot. Into an already experimental cohort, they threw some inner-city high school students at the last minute. There are lots of reasons why this was an irresponsible and stupid thing to do, but I want to focus on one in particular. The underlying assumption here is that those students need basically the same thing that the college students need, but maybe with a little more of something. Tutoring, maybe. This mentality is exactly the opposite of the promises we’ve heard about how technology is going to “personalize” education. News flash: Inner-city high school students don’t just have more educational needs than students who have successfully matriculated to SJSU; they have different needs (and different goals and motivations). Why would we assume that the same course would work for them?

What’s weird is that this mentality is also the opposite of what has made Silicon Valley great in recent years. The real revolution in software over the last decade has been in product design and development techniques that give us much better ways to understand the real and specific needs of well-defined classes of users. There is a range of these methodologies, but they all build from the common bedrock conviction that you don’t understand your users’ needs when you start. As Clay Shirky is reported to have said, “The first prototype isn’t meant to show a solution. It’s to show that you don’t yet understand the problem.”

If you want to help improve education as an entrepreneur, then start with that nugget of wisdom. Start by assuming that you don’t yet understand the problem, and that educators and students know more about the problems that need solving than you do. Use your skills to help them illuminate and elucidate the problems that they are trying to solve, and then work on your solution—not to “education”, but to a specific educational problem for specific actual humans. This is not to say that you can’t have a big impact. Education is in desperate need of help to untangle the mess of needs, goals, approaches, and institutional structures so that we can do a better job of helping more people. There are big challenges. But note the plural. There is not one hard problem. This a complex of many intertwined and poorly defined hard problems. Improving education isn’t like designing a better way to order a taxi, or building a better smart phone, or even inventing a self-driving car. It’s harder than any of those, because it is far messier than any of those. This makes it an incredibly gratifying space to work in as long as you don’t do so out of a fantasy that you and your entrepreneur peers are the heroes who are going to “save” it, after which you will be greeted as liberators. That way lies madness. And failure.