Education : Open Source at the Cutting Edge of Science Education
So I’m not your traditional computer geek. I’ve never played Dungeons & Dragons or World of Warcraft. I’ve never even seen a single episode of Battlestar Galactica. What I am is a nerd, a science nerd. I’m a PhD Chemistry student and my research has nothing to do with computers. However, open source software does have a long tradition at universities. I’m reminded of things like BSD (Berkeley Software Distribution) and the MIT license, the number of .edu domains hosting Linux/Unix OS mirrors, and the sheer volume of open-source software that started out as a grad school projects. The open-source development model fits in fairly well with the “academic freedom” that universities stand for. It is also fits well with the model of scientific inquiry. Ideas like peer review, constant empirical testing, and emphasis on experimentation play right into the way open-source software developers often approach their work.
So one would think that open-source software would be simply ubiquitous in science education. However, my experience, at least in the field of Chemistry is quite the opposite. You can find lots of hard-core research applications that grad students or research scientists might use, but when it comes to teaching undergraduates in particular, I very rarely see open-source software used. In my department in fact, it’s been non-existent until this semester. I’d like to share two personal examples of how open-source software is helping push university science education and a little bit about why I think open-source software is going to revolutionize science education in many ways.
This semester I’m the teaching assistant for a Physical Chemistry Laboratory class. I’m teaching mostly chemical engineers and they mostly feel it’s just a tad bit like watching an old black-and-white movie, boring, antiquated, and irrelevant. One of the things we’ve done this semester to liven things up a bit is add a lab involving a Scanning Tunneling Microscope (STM). I’ll spare you, dear reader, from all the gory details, but in short it lets us take pictures of molecules. It’s really pretty cool and the students get to work on a brand new instrument. The STM itself costs around $10,000 USD but the company wanted to sell us image processing software for ~$3,000 USD. The professor in charge of the class called up an STM expert at another university who had been using the same instrument for teaching and asked if the software was worth it. He said, “nope, just use this open-source program called Gwyddion“. Cool, we save $3,000 and get a better working program that is cross-platform, imports the native file format of the data acquisition program, and is easy enough for students to start using with only a few minutes of instruction.
There are two important aspect to me about this software. First, it was developed by scientists, for scientists. So much of the time when I use proprietary science software I can just tell that it was designed by somebody who knew more about programming than the science that the program was trying to accomplish. Gwyddion has a community of users and developers around it to support and guide the project. Second, its free and cross-platform. It may not seem like a huge deal, but a lot of times science software is > $1,000 USD which makes it impossible to let students take it home. Usually we make them check out CDs (and there are generally not very many) for a few hours at a time in the department’s computer lab. Students hate that and so they really never explore. Being able to say “you can download this for free at home, on Windows, Mac, or Linux, and play around with it” is really a big deal.
The second personal example I’d like to talk about is an up-and-coming molecular editing and visualization project called Avogadro. Like Gwyddion it is cross-platform and developed by scientists (covering biochemistry, molecular physics, and applied mathematics). I got involved with the project while looking for something to replace the software my department is currently using to teach first-year students about quantum mechanics.
Our current software was released in 1996 and cost something around $1,000. It crashes constantly, students hate it, and I end up teaching a “howto deal with computers” lab rather than the chemistry the students need to learn. I found that Avogadro had all the features needed for the first half of the lab, but was missing a couple key features for the second half. Since I had been in contact (and even contributed a patch here and there) with the developers on the mailing list and IRC I asked them if it would be feasible to add the features we needed. They were interested in the idea and started working on getting the needed code roughed out. To make a long story a bit shorter, starting in the fall semester we will have a program that will not aggravate the students, will have much better features, and they can install it at home to do homework with.
What I love about this story is that the open-source development model let me, as an educator, get involved with the development of the tools I need to help students learn some rather difficult material. The developers where interested not in what they were getting paid to code, but on what their users needs were. And as teachers use their software, they will continue to refine features that will drive education further.
I hope I’ve been able to give you a little glance into some aspects of open-source software that have great potential for improving science education. It is more modern because it is generally faster paced, more accessible because of low/no cost and often being cross-platform, and more science-like because it’s written by scientists and lets users get involved with design and development.