Universities' use of the Internet and software packages for their courses called "courseware" has become all but universal in the past few years. As universities and their professors work to take advantage of the opportunities technology and the Internet provide while maintaining the integrity of their classroom and teaching environments, they face the difficult choice of the software they will use to create the online classroom and university environment they want for their students. Three options present themselves: commercial courseware packages, courseware custom-developed by a university for itself, or the vision, courseware developed collaboratively by universities in open code. This section explores the pros and cons of each of these options by examining the major concerns expressed by Harvard University decision-makers in their recent choice to develop their own educational software platform.

During the dotcom years quite a few commercial companies have turned their attention to developing education-related software. This software generally allows a professor to create a website with a place for posting messages to the class, a syllabus, electronic access to readings (sometimes), and a threaded message discussion board. It is relatively easy to use and many professors and teachers have integrated it into their classes with success. The best of these software products were professionally developed, easy to use, equipped with the state-of-the-art educational technology features, and responsive to university and professor feedback and recommendations for change. However, since the implementation of commercial software platforms, some universities have begun to question whether they are going the right route.

In considering the choice of educational software platforms, four main concerns face university decision-makers. First, universities must be concerned about the long-term stability of the commercial products. Even before the cooling off of the e-boom, some educational software providers were there one year, gone the next. Now that problem looms even larger.

Second, universities understandably want control of their course platforms. Location of servers, security of data, and integration with specific university data management backends present significant issues. Intertwined with the concern about control over information is a deep concern about control over the content of the courseware. Ultimately, the company not the university will get to determine the features included in the package.

Third, cost is a factor. Many universities are limited in their options by budget constraints, making it difficult to get the best software or to create a customized platform from scratch.

Finally, many universities see potential in the use of the Internet to facilitate their classroom education that is not being realized with the standard education software platforms. Though quite a few of the exisiting commercial software plaforms are very creative as well as responsive to suggestions for change and augmentation, most currently incorporate tools that mimic the types of interaction that traditionally take place on paper or in a classroom. It is the hypothesis that technology can be leveraged to create tools, such as the rotisserie, that facilitate entirely new modes of teaching and learning interaction. Though a tool like the rotisserie might be incorporated by a commercial software company either because of its own educational research or because of a recommendation by a professor who has conceived of it, thus far at least the commercial educational software companies have not focused their energy on the creation of new, clever educational tools.

Those universities with sufficient funds available are still wary of developing their own platforms because of their inexperience in software development. Opting for the large outlay of money and uncertain results is a high risk proposition.

Given all these concerns, an open source, collaboratively developed platform seems substantively to be a promising option. Because the code is open source, stability concerns can be at least somewhat assuaged by the knowledge that the code itself could not be withdrawn even if it were maintained by an outside company that went out of business.

Control would be in the hands of the universities, individually and collectively. The universities themselves would be the arbiters of the design and features of the courseware. If a single university wanted to add, subtract or revise a feature of the courseware, the open code in their implementation of the software would allow them to do it themselves. Control over information could remain with each university or be outsourced as desired.

Costs would still be significant, especially at the beginning, because each school would have to adapt the platform to its own needs and would be responsible themselves for any additional development it sought. However, to the extent that many schools chose to use and improve the platform, the costs would low. In any case they would be no higher than a decision by a school to develop its own platform.

Finally, risk might be lower. Not only because the costs would be less, but also because the simultaneous and continuing efforts of multiple schools would make any one error in the development process fixable by any other school.

There are some major disadvantages to this option as well. The problem of inability for collaborative efforts to move quickly, effectively, and cohesively plagues open source software movements. Lack of structured, top-down organization leads to false starts and dead ends. Lack of accountability allows projects to languish. For a university in need of good courseware, time wasted is time lost. If it is contributing to development costs, it is money lost as well. The quality and completeness of the resulting system is no more sure-very possibly less sure, in fact-than self-development.

To truly benefit from a professor has to be a little adventurous-ready to try some things that will be different in many ways from the usual classroom experience and, at least that the outset, still in the phase of experimentation and development. Though the experimentation and change can be valuable for its own sake, the real value will come from the tools themselves.

Most professors enjoy their teaching but often need to focus on their research or find that a topic for a particular day of class is one on which they are not expert. provides for collaboration with other professors that make teaching and preparation easier and also remove many of the obstacles that prevent colleagues at different schools from working together.

The syllabus planner is the main vehicle for allowing this collaboration. Rather than piecing together a syllabus from paper copies from previous years and other professors, a professor will be able to look at the syllabi of other professors and literally drag and drop units onto her own syllabus. What's more, she will be able to plan a joint rotisserie or joint use of another tool with another professor when they are covering the same topic on the same date. The tools themselves will simply provide ways to interact with students-and get the students to interact with each other-that were not easily possible before.

Ideally, when experimenting with in their own classrooms, professors will have ideas for ways the tools could be improved or for new tools that could be added. Because of the open source nature of , it will be relatively easy for professors (with a little help from their universities or a couple of tech-savvy students) to build new tools that they can easily add in to their course website. In this way, can develop and grow to be more robust and to have a wider range of tools in the toolbox.