What is open source (OS)? It really describes two different things: a product and a process. The product is software that comes bundled with the code available and a license that permits users to download, modify, and redistribute it as they wish. The OS process is one in which a loose network of people from disparate places, communicating mainly via the Internet, who volunteer their services to develop a particular piece of software. Open source software that is considered successful, such as the Apache web server and Linux operating system, have vibrant communities of programmers working on them, mostly without pay [see related article: The Success of Open Source A Subject for UC Berkeley Researcher ].

Open source projects are often likened to a community of learners and researchers who share their findings and improve upon them for the benefit of the entire community. Not surprisingly OS has been popular among academics who share this value of building upon the knowledge of others. In some areas of the university where advanced technologies are commonly used -- in research and digital libraries, for example -- the use and development of OS applications is very common.

The tipping point of mainstream acceptance of OS by higher ed for teaching and learning applications came last January with the announcement of the Sakai project. Named after CHEF, the University of Michigan's CompreHensive CollaborativE Framework, as well as an actual chef from the "Iron Chef" tv show, Sakai is, at its core, a collaboration among four universities (MIT, University of Michigan, Indiana University and Stanford) and the uPortal consortium to build an enterprise-level, open source, learning environment (a course management system, portal, collaboration tool, and more). In its first year it grew to include 50 other institutions as partners, including five UC campuses (Berkeley, Davis, UCLA, Merced, and Santa Barbara), which each pays $10,000 a year to join the Sakai Educational Partners Program (SEPP), giving them early access to code, development staff, workshops, and other services.

This is where the Sakai project currently differs from "traditional" open source endeavors: it relies on structured and deliberate collaboration within a core set of resources instead of being open to public code submissions and relying on volunteerism. Sakai, in fact, refers to itself as community source, which it defines as:

The $6.8 million investment that universities and the Mellon and Hewlett Foundations are making in Sakai could be seen as a vote of support for the principles of open source. And for many campuses it is a clear signal that the current commercial market for learning technologies (and in particular, the two main course management system vendors: Blackboard and WebCT) is not meeting the needs of higher ed.

Although each of the UC campuses has its own unique contexts to consider when deciding whether to go the OS or proprietary route, they share some common motivations as well, mainly control and flexibility.

Control and Flexibility

A major complaint about proprietary software is that, because you can't see the code, you can't fix it or enhance it. You're stuck with a product the way the vendor created it until an upgraded version is put on the market. In successful OS projects, bugs are quickly fixed and new features are added as soon as they are ready because a network of like-minded folks is working on solving the same problems and is not beholden to a financial bottom line.

"I have found that the level of sophistication and the speed with which OS has evolved in response to needs of end users has been phenomenal," says George Michaels, an instructional consultant at UC Santa Barbara. "It's been amazing to me the rate of evolution that OS software has exhibited, which makes it very approachable, very responsive, and whenever bugs are found they are fixed quite rapidly and with a lot of notification. When you're involved with open source you're definitely plugged into a network of people making sure you're getting a best-of-breed product."

A project like Sakai puts the control firmly back into the hands of the people who work most closely with the ultimate end users -- in this case, the faculty -- thus the priorities of the development can be based on the needs that faculty articulate. This is the area in which the Sakai project might provide the greatest benefit to campuses, at least according to many of the UC people involved in the project. Simply put: the commercial systems, such as WebCT and Blackboard, do not provide the tools or user experience that faculty want. Also, there is great hope among the Sakai partners that because the system is being built by academics, it will better promote pedagogical uses of technology.

"The campuses at the core of Sakai have a lot of experience in developing OS learning applications," says Rich Kogut, Chief Information Officer of UC Merced. "That means we'll get products driven by real user feedback and not by commercial companies trying to sell other products. The ability to leverage this great developer community around a common framework will give us a plethora of products that we would never be able to get with commercial vendors."

UC Merced decided to join Sakai -- whose course management system it plans to launch when the first cohort of undergraduates arrives next fall -- because it would allow the campus to create a seamless and consistent web experience for its constituents. That is, they want all of their various applications and technology environments to communicate with each other so that users can get the information they want without having to navigate between systems and logging in multiple times. The fact that Sakai is based on open standards should help them achieve this goal. UC Merced already use the open source uPortal for its campus portal, and Sakai is being built to work on top of that, so the campus has a unique opportunity from the very beginning to use systems that will work together and to scale as the campus grows.

UC Santa Barbara does not currently have a standard course management system, and is evaluating Sakai to potentially become the first one. The campus has looked into purchasing commercial systems but has been turned off by "horror stories" of their expense and lack of flexibility, says Michaels.

"Usability for faculty and students is probably the number one factor for us," says Michaels. "And the extent to which it can promote good pedagogical practice is also important."

Likewise, UCLA is evaluating Sakai because of its potential to offer what faculty really want: ease of use and tools, often locally produced, that have pedagogical value. The campus uses several course management systems, from proprietary systems to homegrown products.

"Right now faculty are bounded by whatever their own staff can produce in terms of new tools or in terms of a vendor who is not necessarily motivated in the same way they are," says Ruth Sabean, Assistant Vice Provost and Director of Educational Technology. "They inevitably say they can't use WebCT or Blackboard because the interface is too complex. Having a community of faculty who are contributing to Sakai by providing feedback on tools, and having hundreds of developers contributing tools, I think will be a big motivator for faculty on this campus."

Costs

A common misperception about OS software is that it's free. Yes, most OS software is licensed at no cost, but, as with proprietary software, it requires programming expertise to implement and support, which the campus either must provide or purchase from a third party. In addition, OS software requires resources to modify the code for local needs. Thus, cost savings are not a given with OS applications; it's more about using resources in a way that aligns with a campus's goals.

Some campuses, such as Berkeley, would rather spend the money they would otherwise use to license proprietary systems (they currently license both WebCT and Blackboard) on a tool that more closely meets their teaching and learning needs, interoperates with other campus systems, and that allows them to develop the skills of their own team of programmers.

"I can be putting resources into the licensing costs of WebCT and Blackboard or I can be putting them into resources here, primarily human resources," says Mara Hancock, Associate Director of Educational Technology Services. "We want one system that my developers can focus on, and the faculty development folks can support, so that we can move beyond just how to use a tool to how to meet pedagogical needs."

At UC Davis, which is evaluating Sakai as a potential replacement for its homegrown course management system, MyUCDavis, staff development is perceived as a crucial benefit of using OS products. "Open source is a way to leverage the skills we've got instead of purchasing them," says Kirk Alexander, online media manager for teaching and research.

"Open Source to me is crucial because if you have any interest at all in doing programming it's a great learning environment because reading other people's code is a wonderful way to learn," says Mike Franks, a programmer in social sciences computing at UCLA who has developed an open source course management system called ClassWeb.

For campuses such as Berkeley, Davis, and UCLA that have relatively large programming staffs this approach makes sense. On campuses with fewer technology resources, however, relying on staff to provide development and implementation services as well as support might be prohibitively expensive. So if these campuses were to go with OS, they would need to either hire more staff or purchase support from companies such as Unicon that provide support for open source products such as uPortal and Sakai.

"At our campus, the cost of licensing WebCT is less than half of a programmer analyst II salary, which is what you would need if you went with Sakai," said Tony Wood , director of academic computing at UCSD. "We have used WebCT for six years and have trained 600 faculty to use it and over 200 courses a quarter employ it. I have to look at Sakai as an unproven package that would require a lot of training to use, and we'd have to give up a lot of features that we currently get with WebCT."

Shared Needs

An area where open source projects like Sakai could greatly benefit higher education is in minimizing redundancy of efforts. Within and among campuses people are working on the same problems; by working collaboratively to solve them there may be time freed up to work on other things, which could result in more and better tools. The tools built will also be more sustainable and scalable than if they were built for only one campus's use and distribution.

"Instead of creating redundant tools, we can develop them within the Sakai Educational Partners Program and you just don't have those opportunities with proprietary systems," says UC Berkeley's Mara Hancock.

In fact, UC Berkeley is contributing an online gradebook application to Sakai. The campus was already planning on building this tool when the Sakai project was announced and decided to build it to open standards so that it could be utilized by countless campuses, not just its own.

Because of its contribution of the gradebook, to which two programmers and a project manager are assigned, UC Berkeley is one of only a handful of campuses represented on the Sakai board.

Over the years UC Berkeley has established itself as an active member of the open source community and has been part of several OS projects, including the Chandler project from the Open Source Applications Foundation, the Open Source Portfolio Initiative, and the uPortal project.

"I think it's really about building something that's rich and built by people who understand what it means to serve faculty," says Hancock. "I'm really hoping that we'll see a vibrant community and product with Sakai as we've seen with other OS projects."