The Story of the O’Reilly School of Technology – Part 3

In this third installment of my series of blog posts, I’d like to tell the story of the O’Reilly School of Technology and the new Make: Mathematics project. I’ll also use this time and space to explain our overall mission and plans for OST in the future.

Sorry about the gap in between posts; I’ve been spending most of my time working hard to help develop a technology that will bring this story full circle shortly!

Part 3: History and Timeline

First, let me give you a short synopsis of the world of technology in education. The University of Illinois at Urbana-Champaign (UIUC) has a long history of using technology as an instructional aid. As early as 1960, U of I began experimenting with computers in education in a project called Plato. Many of today’s social networking technologies have roots than can be traced to the Plato Project. In fact, this year marked the 50th anniversary of the Plato Project. At UIUC, Plato was expanded upon and used throughout the 1970s and 1980s. It was eventually commercialized by the Control Data Corporation and was in use through 2009.

Meanwhile, in the 1980s, a successor to Plato called novaNET was being developed at the UIUC. It was a satellite-based system that allowed automated question generation. novaNET was eventually bought by Pearson and is still in use today.

In the early 1990s, the University of Illinois developed Mosaic, the world’s first web browser. Naturally, Mallard, the world’s first web based Learning Management System (LMS), was developed at the U of I as well, in its physics department.

Valuable lessons were learned from these experiments with computers in education at UIUC, mostly with negative results. It turns out that simply using technology does not guarantee improved learning outcomes. In hindsight, it’s clear that these technologies weren’t offering any real change in pedagogy, because the technology was just being applied to traditional teaching techniques. They were being used as another way to deliver information and quizzes. They didn’t change the learning behavior of students. They didn’t change the roles and relationships between the instructors and students.

In other words, they didn’t put the computer to work as a mechanism of student exploration with instructor coaching.

This remains the biggest problem with instructional designs that use the computer and internet today. Most educational technologies still simply replicate the same old teaching and learning patterns onto a new medium–and it doesn’t work.

But we can fix this problem. And I’m confident that history will eventually show that the most significant thing ever to happen in Educational Technology since the invention of the printing press, happened at the University of Illinois, Department of Mathematics, in 1989. Several events coincided perfectly in the late 1980s in Champaign, Illinois that led to creation of the Calculus&Mathematica Project. First, in 1987 Stephen Wolfram, who was then part of the mathematics department at the U of I, launched the Computer Algebra System (CAS) called Mathematica. This is a powerful program that supports mathematical editing, computation, and symbolic mathematics. Second, in 1988, the National Science Foundation (NSF) began offering grants to institutions for the purpose of Calculus reform. Third, Jerry Uhl and Horacio Porta, both of the University of Illinois, and Bill Davis of The Ohio State University, wondered collectively whether the new Mathematica software could be used to teach Calculus. (Jerry had been disappointed by past efforts to use the computer in education, and despised Plato and novaNet.)

After receiving a grant from NSF, the three of them created calculus lessons in Mathematica files called notebooks. They also built computer labs specifically designed for the project, where students could work on these lessons using Mathematica. There wasn’t a separate textbook. The content was integrated into Mathematica by virtue of being written in Mathematica notebooks. This turned out to be very important to the effectiveness of the paradigm.

The first Calculus&Mathematica courses at the University of Illinois were offered in 1989. Almost immediately, something unexpected happened to the structure of the courses. At first, the courses were taught during three lectures and two lab hours per week. Jerry noticed that his once popular lectures had barely any students present attending them now. He asked some students who happened to be in the lecture one day where the other students were; they took him to the computer lab, where the rest of the students were all working on their math lessons using Mathematica. Jerry, being very adaptable and open minded, decided right then that he would no longer lecture to teach mathematics.

(Aside: Please don’t get the impression that using this technology is the complete recipe for success. It’s not. The instructor is STILL the most important part of these courses. What changes is their responsibility and relationship to students. This is very important to understand and I will be discussing it in more detail in future posts.)

A year after they began teaching using Calculus&Mathematica on campus at the U of I, they started Netmath, a distance-education project that used the Calculus&Mathematica model to teach mathematics to high school students who lived in remote rural areas. Both of these projects are still alive today which speaks to the effectiveness of the methodology, as well as the passion of the people implementing the programs.

Other NSF funded reform projects haven’t done so well. Between 1988 and 1994, the NSF supported some 127 Calculus Reform projects. Of those projects the only ones that I know are still being used today are the Calculus&Mathematica (C&M) and Netmath projects at the University of Illinois.

The creation of Useractive, Inc.

As I mentioned in the previous blogs, I became passionate about this new paradigm, and was doing everything I could to spread the gospel of this new approach in math education. I went to conferences, and I met with teachers, professors, and administrators. But by 1996, I realized that this project wasn’t going to spread across the land to other schools and change education like I hoped. There were too many barriers, some of which were inherent in the technology, and some of which had to do with the structure of educational institutions themselves (I wouldn’t fully understand this until years later).

I did realize that if there wasn’t some sort of business supporting this paradigm and keeping it going, that it would eventually wither and die in the University system. There is no bureaucracy or budget within university math departments to start such a program, let alone maintain it. C&M only got off the ground at the U of I because the project received elusive NSF funding and benefited from the heroic work of its founders. I knew that when Jerry Uhl and Bill Davis retired, there wouldn’t be any of that kind of support left in their respective departments to maintain and grow these projects. And in fact, these courses ended at Ohio State shortly after Bill retired because no faculty remained to take the reigns.

In 1994, while working on the C&M project, I became interested in the internet, and specifically the new “World Wide Web”. At one point I picked up a copy of Learning Perl by O’Reilly, because I wanted to use Perl to build some CGI programs to make my web pages more functional. I became an immediate fan and follower of O’Reilly books. Randall Schwartz wrote a lot like Jerry Uhl and Bill Davis. “Learning Perl” was very conversational, funny, and used simple, well thought out examples to explain concepts.

Because I was in a university math department, I had a Unix account with Perl installed and I had friends who showed me how to access and use Perl. I wondered how people who didn’t have these tools and friends available, would learn Perl. So I set out to make a web-based tutorial that combined C&M’s technique of integrating content with software, and Randall Schwartz’s approach to Perl. I created a website called the “Web Workshop.” Within few months the site had 20,000 users building websites that used CGI on the server to create form-based interactive websites.

The Web Workshop was completely web-based, but still allowed people to program in Perl and test their CGI programs all from one page. The content and programming tools were located on a single web page. The content was preloaded in one textarea, and directly below it was the textarea and form to write Perl code, which was saved to the server to run. Output was displayed on the web as well. This way of combining content and tools made it possible to mimic the kind of interaction that happened in C&M. But still, something was missing. The site needed to an instructor. I was getting a lot of questions from students, so I designed a web-based feedback mechanism that I could use to test peoples programs and give them feedback, quickly. Because of the way it was designed, I could give feedback to hundreds of people a day all in a couple of hours.

In 1996, I met and fell in love with Tricia Mills who was a young, but very competent programmer and a manager in the C&M project. I moved to Champaign, got a job teaching at the University, and Trish and I got married. It wasn’t long before Trish had improved the work I’d done on the Web Workshop by making a more functional and flexible web based IDE called CodeRunner. In 1997, we started Useractive, Inc. a business dedicated to building and delivering online learning systems with the characteristics of both the C&M courses at UIUC and the Web Workshop courses I’d built.

We partnered with the University of Illinois to offer courses through their Department of Continuing Education, which provided us with enough income and momentum to add more partners and employees (including Kerry Butson, Trent Johnson, and Josh Nutzman). Because we were small and had limited finances, we put most of our efforts into building systems that focused on reducing the customer service overhead associated with getting students to work making things and then giving them instructional feedback. We were also forced to make teaching and administering our courses as efficient and effective as possible, without diminishing the teacher/student relationship. In other words, we had higher production costs than other online course producers, but much lower ongoing costs for the level of educational outcome we were achieving.

In early 2000, we made the decision to accept an investment from a local investment group and the state of Illinois.
Taking investment money turned out to be one the worst and best things we ever did. On the lousy side, the money and the pressure from the investment group compelled us to pursue many lines of business which led us down many dead end paths, which wasted a lot of time and money. On the positive end, by exploring those paths and dead ends, we learned a whole lot about the landscape of education. By trial and many errors, like a blind person might walk around in a new room until the entire room is mapped out and understood forever, we developed a clear sense of where we were.

We see a lot of people stepping into the business of education these days with the same kind of misconceptions we once had. They are making the same kinds of mistakes we did. Whenever I’m given the chance, I try to warn those people, but like me ten years ago, people with misconceptions about education tend to be passionate about their beliefs. I’ve decided that the only way to purge the misconceptions about education is to make mistakes, hopefully survive them, and move on.

I won’t describe the business aspects of the education market in this post; it would take us far afield of what I hope to accomplish with this particular series of posts. I do intend to discuss that topic fully in some future posts, because it’s important for anyone considering the profession of education to take the practical elements into account. But the business of education is a remarkably complex and subtle topic. Education is actually a myriad of markets, each with different set of producers, decision makers, end users, and purchasers. Part of the reason education has been so stagnant and resistant to change is because of the way in which these market forces dampen innovation. I will say for now, that the only hope for change in education is to combine teaching and learning innovation with a solid business plan that not only takes these market forces into account, but uses them as tools that require change. Change in education will only happen when economics and demand force it to do so.

Joining O’Reilly and the creation of the O’Reilly School of Technology

Over the years, I had worked with many different publishing companies. Those experiences completely soured me on those organizations. They were all backwards, traditional thinking, and void of any innovative spirit.

However, somehow I knew that O’Reilly was different. Like most fans of O’Reilly, I became a loyal fan after reading that first O’Reilly book. I followed O’Reilly’s story and movements over the years and saw how open and willing they were to think outside of the publishing box.

Since we were in the business of teaching people programming skills, and since O’Reilly is the most revered computer book publisher in the world, I knew a partnership of some kind could be constructive. On a lark, I sent Tim O’Reilly and Dale Dougherty an email. I was thrilled to learn that they were receptive to a discussion. It turned out that we had something in common; we all wanted to build products focused on the learner. It wasn’t long before we worked together to create an O’Reilly branded product called the O’Reilly Learning Lab. Eventually this O’Reilly-branded channel dominated our business.

Because of my accumulated experience, I knew that creating a school was key to achieving my ambition of changing education. I also knew that a school would need a strong brand. Students seeking education, seek to identify a school’s brand first. What better brand than O’Reilly to link to a school? O’Reilly already seemed to fuse an academic integrity to their products and their business.

In 2005, O’Reilly purchased Useractive. A year and half later we launched the O’Reilly School of Technology. We’ve grown slowly and carefully, making sure that we continue to build courses using the paradigm and spirit of those first courses in 1989. Web and internet technologies have evolved with giant steps recently and we are continuously building platforms that utilize these developments as fully as possible to reduce the students overhead of getting tools together to experiment and build things, and to make asynchronous instructor feedback as quick and meaningful as possible. (It turns out that synchronous feedback online is full of potholes that actually reduce educational value while increasing cost).

This historical perspective was the hardest for me to write (which may also account for some of the delay), and probably the least interesting to the reader (no Russian intrigue, no wacky professor descriptions, and a whole lot about early mistakes and failure!), but it’s the only way to get a complete perspective of our experience and motivations.

In next and final post of this series, and the one I can’t wait to write: our ambitions for Math education come full circle!

  • Ron Bannon

    Okay, so where’s the O’Reilly product? I was hoping to see something that we could actually use. I’d love to see C&M done the O’Reilly way—even with a book, and an LMS for schools to use.

  • scott

    Ron, it’s being beta tested in a Matrix and Geometry class at UIUC this semester starting Aug 23. Sometime during this semester we will release to the public.


  • Matthew

    Are we ever going to have some users forums dedicated to the OreillySchool programs to ask questions and give feedback? I see at there are forums dedicated to many books. It would be nice to have a forum dedicated to each class or each subject area just for OreillySchool.

    Also, it would be nice to be able to download all the lessons as a PDF file for future reference and study offline.


  • Scott Gray

    Matthew, we’re updating our LMS now. we want our forums to be useful in terms of code sharing. When we just had normal forums, people talked about cats a lot.

  • Scott Gray

    Python 3 is available in late November and the last one in Febuary.

  • Carolyn Oros

    I am interested in the upcoming Python certification and would like to take the two coursed that are available now; if I take the two classes now can they be applied to the python certification when the other two classes are available?


  • Mental Math

    Learning computer and its logic is not that easy. A student will learn and will effectively learn if the school he is into can be considered as high standard and have good mentors and facilities. And I am sure O’Reilly School of Technology is one of the best.