Stanford launches three free online computer science courses

Background

Stanford Engineering's three courses were not built from scratch for the web. They were among the school's most popular computer science courses, and the online versions were meant to give people outside Stanford access to the same faculty and the same material ^[1]. What was different was the scale and the openness: anyone who could get online could sign up, and the sign-up numbers went well past what a lecture hall, or a conventional distance-learning programme, can hold ^[1].

What happened

The three classes — Introduction to Databases, Machine Learning, and Introduction to Artificial Intelligence — officially started on October 10, 2011 ^[1]. Introduction to Databases was taught by Jennifer Widom ^[1]. Machine Learning was taught by Andrew Ng ^[1]. The artificial intelligence course was taught jointly by Stanford Research Professor Sebastian Thrun and Peter Norvig, Google's Director of Research ^[1].

The numbers

Stanford says 66,000 students signed up for Jennifer Widom's database class ^[1]. In Andrew Ng's Machine Learning class, 72,000 registered students had already watched more than 850,000 video clips before the first day of class ^[1]. The artificial intelligence course attracted an unprecedented 160,000 students from over 190 countries ^[1].

The activity did not stop at registration. Students in the AI course were collectively querying the course database at more than 7,500 times a second ^[1]. Online volunteers had begun translating the Introduction to Artificial Intelligence class into more than 40 languages ^[1].

What students got, and what they did not

Online students did not get Stanford credit ^[1]. What they gained, according to the school, was access to faculty and to three of Stanford's most popular computer science courses ^[1]. The teaching came from the people who normally teach those subjects on campus, but the record of the work completed carried no Stanford credit behind it ^[1].

A limitation in China

Andrew Ng said the Great Firewall of China was blocking the course video, so the classes could not serve China well at the time ^[1].

Why it matters

Taken together, the figures describe a demand that no campus can hold. The AI course drew 160,000 students from more than 190 countries ^[1]; 72,000 people in the machine learning class had watched more than 850,000 video clips before the first class had even met ^[1]; students in the AI course were hitting the course database more than 7,500 times a second ^[1]. All of this activity came from students who would receive no Stanford credit ^[1].

Two details point at how the audience behaved rather than simply how large it was. Volunteers, working online, had started translating the artificial intelligence class into more than 40 languages ^[1]. And the course sites were not passive pages being loaded: the database load and the video views were produced by people working through the material on their own schedule ^[1]. Stanford's own account of the launch treats the popularity and the activity around it as the headline ^[1].

What the three courses made visible was a mismatch between supply and demand in computer science teaching. A single campus can seat a few hundred students in a lecture hall; the same course, opened online, drew tens of thousands in each of the three subjects, and one of them drew six figures across more than 190 countries ^[1]. The courses were free, they were taught by Stanford faculty, and they awarded no Stanford credit — a combination that had not previously been tested at this size ^[1].