Hot on the heels of a new $37 million White House initiative designed to encourage data science work at American universities, education startup, Udacity, is responding with a data science initiative of its own. Today, the Palo Alto-based company launched an inexpensive, comprehensive degree track-style program that will allow anyone with basic computer science skills to get schooled in the wild and woolly world of data science by the companies that helped pioneer those concepts.
It’s a good thing, too, because as smart sensors, devices and networks proliferate, computing power and capacity expand, and everything becomes increasingly “connected,” we’ve officially entered the Data Era. The amount of data the world creates today, and the amount of data now available at our fingertips, is exploding. Exponentially. As a result, data analysis, data munging, cleaning, transforming and everything in between are now terms and jobs that every industry is familiar with — and data scientists of every stripe are in high demand.
But there’s a problem. While “Big Data” is the buzzword of the day, and many companies are scrambling to hire these elusive and expensive data wizards, the word on the street is that they’re in low supply. And the forecast doesn’t look good.
“There is a shortage of big data experts,” said Michael Rappa, the Director of Advanced Analytics at North Caroline State University in a recent interview with Federal Computer Week. “I don’t see the gap narrowing. Universities aren’t producing enough.”
TechCrunch readers are likely familiar with the work of data scientists in some capacity, even if they couldn’t pick one out of a lineup. But just in case, if Facebook or Google ads have ever suggested things eerily familiar to your personal life, well, you have your friendly neighborhood data scientist to thank for that. He or she likely played a role in architecting the very preference-predicting algorithms that brought you that ad.
Now, to be clear, data scientists come in many shapes and sizes — quantitative, operational, marketing, product, predictive modelers, statisticians, engineers, miners, visualizers, warehousers, machine learners and so on — and they perform many different roles within an organization. However, boiled down, a data scientist’s job is to mine reams of digital static and noise (content, user behavior, whatever it may be) in search of a signal. Data scientists, among other things, help identify patterns and glean insight from our ever-changing datasets.
Luckily, with the list of potential applications for data science in the business world growing long, and the need for data scientists becoming universal across industries, universities have started to wise up. For that reason, it may not surprise you to learn that Udacity isn’t the first to offer an online data science program.
In fact, after doing its due diligence as it looked to launch an online data science degree, UC Berkeley came to the conclusion that the MOOC providers, Udacity included, didn’t cut it. In other words, they didn’t yet offer the level of support they were looking for.
While the MOOC providers offered quality interactive content, Berkeley’s Dean of the School of Information told VentureBeat in July, they didn’t offer enough feedback. Partnering with veteran edtech startup, 2U, UC Berkeley launched its online master’s program in data science in July — after working with the startup to develop a “two-way delivery mechanism” that would allow student coursework to not only receive grades but peer-review as well.
There’s one slight problem, however. The online master’s program offered by 2U and Berkeley comes with a Bay Area-sized price tag of $60,000.
In contrast, Coursera offers a number of high-quality, free data science courses any student can peruse. However, the startup (so far) has yet to provide any real curriculum or the chance to earn a real degree. Furthermore, there are programs like Enstitute, services like BigDataUniversity, and a laundry list of open courses, certificate programs and tutorials available online for those interested in educating themselves on data science.
A number of schools even offer specific degree programs, like Georgia Southern and Northwestern, but generally the cost is in the thousands and courses are scattered. In other words, they’re not available in a comprehensive vocational curriculum, so those looking for a formal education in data science will likely end up frustrated and without any real credential to show employers. [For more, check out this awesome list of courses and programs offering a data science education in some form or another on Quora.]
But fear not, budding data scientists.
With its new program, Udacity will begin offering a series of comprehensive data science courses, and, going forward, plans to offer a Big Data certificate endorsed by the Open Education Alliance at a substantially lower price. Courses are free for anyone to take, but also give students the option to pay (starting at a few hundred dollars/month) if they want access to more instructional support. And, importantly, for those at risk of dropping out (and being a statistic in online education’s notoriously low-retention rates), there’s a good chance that this premium could lead to greater completion.
The first class on this track is called “Introduction to Hadoop and MapReduce” and is available starting today. The other selling point for Udacity’s new online data science degree program: The classes are co-developed between Udacity and “leading industry experts.”
While that can sometimes be an inflated way of saying “someone with a teaching degree,” this isn’t Udacity’s first rodeo. Its preliminary course, for example, Udacity built in collaboration with Cloudera, the veteran enterprise data management company. In the course, Cloudera experts and Udacity instructors start with “What is Big Data?” and then work through the fundamental principles of Hadoop and MapReduce, touching on important Big Data questions along the way.
Udacity also said via its blog that students will be able to study the free courseware for “Introduction to Hadoop and MapReduce” in self-paced mode, akin to a self-study MOOC. In January, the company will begin offering its full course experience and, continuing to add on to its for-credit classes launched this summer, the company’s Big Data and Data Science courses will include feedback and hands-on student projects — along with career mentoring and so on. This means that, going forward, Udacity’s program could start to look a lot more like 2U’s collaboration with Berkeley for example.
While the startup stumbled in its pilot with San Jose State University, this is, generally speaking, a path Udacity is familiar with, having partnered with Georgia Tech and AT&T to launch a $6K, totally-online master’s degree program in computer science earlier this year.
This new program, however, aims to that one step further (over the coming months, that is), eventually allowing anyone with basic computer science and statistical background to get schooled in a range of courses — from advanced machine learning, big data analytics and bayesian inference to computer programming (and can I get a wha-what from fans of R?).
Ultimately, with industry experts co-developing the courses, not only will students have an opportunity to get an affordable education, but, when all is said and done, they could just have a better shot at landing a lucrative job.
And, in the end, isn’t that what this is all about? Not the education, but the money?
No. Shame on you.