Sameer Verma interview on open education
Teaching the open source way: An interview with Sameer Verma
Dr. Sameer Verma first learned about open source software when a college friend gave him a weekend crash course in Linux. Now a professor of information systems in the College of Business at San Francisco State University, Verma has taken those lessons to heart—and is teaching his own students the open source way.
Recently, we talked with Verma about the challenge of open source pedagogy, about integrating open source technologies and values into the college classroom, about the benefits of learning open source project management, and about his work with One Laptop Per Child.
How and when did you first encounter open source software and open source philosophies? What role do these now play in your life?
I’ve been around computers since I was about 3. My father was a geophysicist in India, where he worked on a PDP 11 back in the 70s. He would take me to his office to show me the computers—blinking lights, punch cards, reel tapes and all. Later, while in college, I worked with VAX/VMS and CDC Cyber mainframes but never got to use any PCs. In fact, I used a PC for the first time in 1992, when I came to the US to go to graduate school.
In a business school, you don’t really see Unix as much. So, most of my exposure was on PCs with Windows and related software. We used a lot of proprietary software, but all of it was paid for by the university, so licensing never came up. I encountered open source when I had a problem and proprietary software was just too expensive! I needed a way to collect data through web-based questionnaires, but I didn’t have any money to buy the expensive software. I mentioned this to a friend. He suggested a Saturday afternoon exercise and gave me a crash course on Linux in exchange for a few beers. I still remember that it was Red Hat 5.2. in fact, I still have the CDs with me! Next thing I knew, I shaved a year off my data collection process! All this on a hand-me-down computer with 48MB RAM! I was sold. Good stuff that you can build yourself.
When I started at San Francisco State University, it became second nature to sneak in a little open source into everything I did. At first, it was a little Perl, some PHP, MySQL backends, and then a couple of desktops to try out GNOME and KDE and other window managers. Red Hat, SuSE, Debian, Xandros, Lycoris, Mandrake, Hancom, Ubuntu, I tried it all. Over the years I have enjoyed playing a role in the introduction of open source in our classrooms and on our campus through our implementations of Moodle and Drupal. I now run a project on campus called “The Commons Initiative." The initiative’s mission is to act as a collaborative hub for all things open source for students, staff and faculty on our campus. We also work with people at the Internet Archive, Creative Commons, Wikimedia Foundation and a few other groups.
Today, all my computers run Linux, both at home and at work. My desk, my backpack, my pocket, my car, all have Linux machines. Our family is completely immersed. My kids don’t recognize many Disney icons, but they do recognize Tux. Both my daughters have their own OLPC XO laptops. Even my mother uses Linux!
At San Francisco State University, you teach a course entitled "Managing Open Source." Tell us about that course.
The course is a bit strange. It is about free and open source software, but it doesn’t involve much coding. To top that, it’s in a business school. The course was something that a colleague thought of. He said that given that I spend so much time with this free stuff, I should maybe teach it. That was an "Aha!" moment. The hard part was trying to figure out whether this was going to be a computer science type course or a course in the business school. Open source is all about licenses that express intent. You couldn’t look at a page of code and tell me whether it was open or closed. Intent, license, and by extension, business models, all belonged squarely in a business school. So, it became a course about open source, and not just about writing code, and was offered by the Information Systems department in the College of Business.
The course can be roughly divided into three parts. In the first part, we cover the philosophical and cultural underpinnings of free and open source. We watch a bunch of videos including Revolution OS. Then, we look at the relationship between licensing and business models—reciprocating ones like GPL and academic ones like Apache. This is in fact something I have researched. That research is useful in class. In the third part, the students pick an open source project and examine its maturity using a multi-criteria approach called the Open Source Maturity Model (OSMM). This gives them a way of assessing a project based on its code, community, documentation, support, services and integration perspectives and then weigh it based on internal requirements. It’s basically a weighted scoring matrix used to assess a project on multiple aspects. The composite weighted score becomes a metric that can be evaluated against a rubric of decisions to experiment with the software, run a pilot test, or run it in full production within one’s organization, be it a bank, a hospital or a high school.
I am quite pleased to say that many of my students get a job because of this edge. Not many business school students have had exposure to Richard Stallman, the Apache license, and software maturity assessment methodologies!
What have you found are the most troubling barriers to open source education—not only barriers to education about open source, but also barriers to teaching and learning the open source way?
This is a tricky one. To begin with there is a perception problem. The status quo rules. Why change anything radically? If we teach using Microsoft Office, why bother with LibreOffice? This is a good question, and I hope one has a good answer before they switch, but there is a certain complacence in maintaining the inertia. There isn’t a perceived need to teach anything about open source. In fact, schools may use open source tools, but they won’t bother going into the nitty-gritty of where these tools came from and why these tools are free and open. There may be some mention about GPL in passing, but that’s it. It’s also amazing that in this day and age, people have told me that open source is not for real businesses. It’s a hippie thing—it used to be free love, now it's free software! The perception problem is real and troublesome.
The other part of the problem is the publisher angle. Anybody who teaches will get this, but if you haven’t taught a course at a community college or a university, you may not think of this angle at all. Most people think the student is the textbook “customer” in the classic sense. This isn’t really true. The student does not make the decision about a textbook or supplemental materials. The professor does! So, publishers court professors extensively.
So, let’s look at a course—its title, its purpose and its outline. The syllabus brings these together. Ideally, one should look for textbooks that suit the syllabus, but books with test banks, slides, supplemental materials, companion websites make it all too easy. So, the syllabus bends the the textbook’s will. It’s low hanging fruit. Now, if you wanted to take the path less trodden, you’d have to get a textbook, find your own support materials, and make your own exams, slides, and companion websites. That’s a fair bit of work, especially as teaching loads increase in public universities. This is really a problem of a support infrastructure. If we had a good way to tackle this problem, we’d see a lot more open source in schools.
Then there is lab support. Most labs come with Windows pre-installed. Installing Linux is yet another thing the IT staff does not want to deal with. I run a lab with Ubuntu in a clever (and somewhat subversive) way. This is a lab with 32 Windows computers. I run a server on the network that serves up Linux images via LTSP. So, when needed, this lab can be taken over and turned into a Ubuntu lab. Then, at the flick of a switch, it goes back into its miserable Windows self (can you tell I really don’t like Windows?). It works as advertised, but it took a lot of work to get it set up, both technically and politically. Not every school has the resources to make it happen.
It will take a concerted effort to get open source into the classroom. We will have to address all these issues in clever ways, so that we can take advantage of what we already have and make a case for why open source needs to have a place in the classroom.
What does open source have to teach us? Or, put another way: What is most valuable about an education in open source tools and principles?
Open source is incredible, both philosophically and operationally. Operationally, the cost of getting tools and help is minimal. Adding yet another parser or a compiler is trivial. Showcasing everything from desktops to the cloud is a lot easier. I built a Wordpress cloud in under ten minutes using Juju and LXC in my class. How cool is that?! I can talk about cloud computing all day, but nothing delivers the concept like building it yourself.
By experiencing the approach to problem solving, looking for answers in forums and IRC, students get a hang of how the power of community works. There is a learning experience about the balance between crowd and quality. There is also the invaluable lesson of getting the job done. Open source meritocracy isn’t about getting famous or rich, although one surely can, if they really want to. It’s about solving problems!
Do the students you encounter have much familiarity with open source? If not, how do you introduce them to it?
Some of them do, but most of them don’t really know it. They all use Firefox or Android, but they don’t really think of it as a process. They simply look at it as a product. The course (“Managing Open Source”) really shows them the “behind the scenes” view of why Firefox is important, or why Android on Samsung is so different from Android on HTC. Free (as in "zero cost") isn’t really a useful angle. Honestly, software in college is usually free, one way or the other. So, the zero cost angle doesn’t mean much.
The notions that help are: flexibility, capability, better use of older hardware, and a certain cool factor. Also realizing that companies like Google and Amazon and Facebook use a ton of open source, or open source might help me get a job, are all helpful in the adoption of open source.
The hardest part is to get the students to understand the whole concept of the availability of source code in an open form. Unless they work with code, they have no reason to care. However, once they get the concept behind the GPL or Apache or Creative Commons license, that sinks in and stays with them.
You have quite an extensive background with the OLPC project. Tell us about that.
This could easily become a much longer interview! The short of it is that I got involved in 2007 simply because I wanted the OLPC XO laptop. I saw it at OSCON 2007 and wanted one for myself! I found out that one couldn’t simply buy these machines, but you could get one through the developer program. I signed up for their developer program and got one. It was then that I truly discovered its power and capacity for change.
What started as a simple desire for a computer became a more involving preoccupation with the educational aspects, the technological innovations, and the social community outreach effort. It became clear to me that to solve digital divide problems, there had to be a better way than shipping old desktops to villages with no electricity! After all, why would people in places with no electricity or network access even want computers? OLPC was poised to change all that. I hopped on the bus and I am still on it. I have projects in Jamaica and India, and work with others in Madagascar, Tuva (of Richard Feynman fame), Pakistan, the Philippines, and many other parts of the world. I also organize the OLPC San Francisco community and we host the annual OLPC SF Community Summit in October. The project as a whole is at almost 3 million laptops in the hands of children in over 40 countries. That’s 3 million users of Linux, Python, and Sugar waiting to take over the world and make it better!
And what makes OLPC such an exciting project?
A part of my family lives in rural India—no running water, occasional electricity, no Internet access. And here I am, living in the thick of it in San Francisco, with the metaphorical faucet always on. It’s an incredible contrast. The people of this village are very close to my heart. I know many of them personally. I grew up learning much from them—some farming, some fishing, a lot of real life. I can milk a cow. I can prime a lister engine and run it. It's my turn to give back. So, after working with OLPC for a while, I started a project in my maternal village—Bhagmalpur. We couldn’t work with the local school. It’s too big for my personal budget. So instead, we modified our approach and went to households. We look at the youngest school-going kid in a household and make her the owner of the laptop. This way, we end up with a “one laptop per family.” The entire village has about 50 houses, and we now cover a third of it. The findings are interesting. The kids have learned an incredible amount with close to zero training! They are already creating their own content. At some point we hope to see if they will be able to create their own applications.
What are the most promising, near-future prospects for open source education, both locally and globally?
Open source as a concept is interesting in many ways. It’s a balance between quality and participation. That balance applies to many other areas, including education. The term “education” is becoming a broad brush that paints both traditional diploma-granting institutions and do-it-yourselfers. Eventually, the true meaning of education is one’s ability to solve problems. Whether it takes a Ph.D. or a Youtube video isn’t all that important. What is important is the solution.
The network plays a tremendous role in connecting the dots. This is why we are seeing an uptake in connected communities solving problems everywhere. In so many ways, the joy of Kickstarter and its like is really a spillover from the open source world. It may take longer, but we do see a point of convergence, where we get meaningful participation and high quality solutions. Software, content, design—a commons—where people have a sense of ownership and purpose. The localness or globalness of that domain is really dependent on the reach of the network and our own understanding of the culture that people bring with them. Sometimes a solution that works in San Francisco may not work in Bhagmalpur, but there is only one way to find out!