At the end of the day, there are no rules, only guidelines.
Thank you, Dean Ramirez and the distinguished faculty here today. And thank you to all the friends and family who have come out to celebrate this day. Thank you all for being here.
But most importantly: you. The Class of 2014. I gotta tell you guys: you look awesome. Downright amazing.
Now, I recognize that I’m the person standing between you and a selfie with your diploma, so I’m going to do my best to keep it short. And to start, I’m going to start with a confession: ever since Professor Getoor reached out and asked me if I’d be willing to do this, I’ve been dreading it. I mean really, really dreading it. I mean like as in final-exam-in-compilers dreading it. Read more…
My commencement address at the iSchool at Berkeley.
I would have never in my life thought I would have been asked to give a commencement speech. This year when I was asked again (last year’s is here), I was once again caught off guard. When I reflected back, I realized that I hadn’t taken the time to say thank you to the people who had really taken a chance on me.
The more I thought about it, I realized that it is a nice reminder to live a life where we dedicate our lives to taking risks on others. One of the people who has been putting that lesson into practice for his entire life is Mr. Knapp. To this day, I’m grateful for all that he did for me. What’s the biggest risk that someone took on you? And what’s the biggest risk you’ve taken on someone else? I’d love to know.
I can’t adequately express how honored I am to be your commencement speaker, but let’s set the tone right for the rest of today and start by being brutally honest with each other.
Right now, at least one of you is silently praying that I’m going to give this talk in 140 character snippets. Come to think of it, I suspect that it’s really Dean Saxenian who has her fingers crossed since she hasn’t read what I’m about to say today.
The Zero Overhead Principle can bring lessons from the consumer space to health care.
Recently I wrote about one of my key product principles that is particularly relevant for designing software for the enterprise. The principle is called the Zero Overhead Principle, and it states that no feature may add training costs to the user.
The essence of the Zero Overhead Principle is that consumer products have figured out how to turn the “how-to manual” into a relic. They’ve focused on creating a glide path for the user to quickly move from newbie to proficient in minimal time. Put another way, the products must teach the user how they should be used.
Just this weekend, I downloaded a new game for my son on the iPad, and he was a pro in a matter of minutes (or at least proficient enough to kick my butt). No manual required. In fact, he didn’t even read anything before starting to play. This highly optimized glide path is exactly what we need to focus on when we talk about the consumerization of the enterprise.
This week, the first Strata RX conference will focus on bringing data and health together. Just as in national security (the place where we came up with the Zero Overhead Principle to help combat the lack of tech adoption by overloaded security analysts), there is tremendous opportunity to apply lessons learned in the consumer space to the health care sector. We know the space needs disruption and it is a way to make constructive disruption with a rapid adoption cycle. Read more…
Smart data scientists can make big problems small.
Having worked in academia, government and industry, I’ve had a unique opportunity to build products in each sector. Much of this product development has been around building data products. Just as methods for general product development have steadily improved, so have the ideas for developing data products. Thanks to large investments in the general area of data science, many major innovations (e.g., Hadoop, Voldemort, Cassandra, HBase, Pig, Hive, etc.) have made data products easier to build. Nonetheless, data products are unique in that they are often extremely difficult, and seemingly intractable for small teams with limited funds. Yet, they get solved every day.
How? Are the people who solve them superhuman data scientists who can come up with better ideas in five minutes than most people can in a lifetime? Are they magicians of applied math who can cobble together millions of lines of code for high-performance machine learning in a few hours? No. Many of them are incredibly smart, but meeting big problems head-on usually isn’t the winning approach. There’s a method to solving data problems that avoids the big, heavyweight solution, and instead, concentrates building something quickly and iterating. Smart data scientists don’t just solve big, hard problems; they also have an instinct for making big problems small.
We call this Data Jujitsu: the art of using multiple data elements in clever ways to solve iterative problems that, when combined, solve a data problem that might otherwise be intractable. It’s related to Wikipedia’s definition of the ancient martial art of jujitsu: “the art or technique of manipulating the opponent’s force against himself rather than confronting it with one’s own force.”
How do we apply this idea to data? What is a data problem’s “weight,” and how do we use that weight against itself? These are the questions that we’ll work through in the subsequent sections.
Data science teams need people with the skills and curiosity to ask the big questions.
A data science team needs people with the right skills and perspectives, and it also requires strong tools, processes, and interaction between the team and the rest of the company.