This is a report to the program officers at the Hewlett Foundation concerning my onsite review of a complementary set of projects relevant to open content learning resources being funded by Hewlett at Utah State University under the leadership of Professor David Wiley. These projects are

I was joined in the review by Richard Baraniuk and Brent Hendricks of Rice University. They are leaders of the Connexions Project at Rice, also funded by Hewlett. Franklin Steen and another colleague from Harvard University observed the first hour of the meeting by video conference.

The visit began in the late afternoon of May 2, continued informally through dinner and ended after lunch on May 3. It consisted of conversation with the primary members of the group as well as formal presentations with extensive interaction between the team and the outside reviewers. The three reviewers on site also compared notes before parting and Frank also talked by phone separately to Rich and Dan several days later .

This draft is written by Atkins and does not necessarily include all of the perspectives of the others although it is based on part on conversations with Baraniu k, Hendricks, and Steen.

General Impressions

Wiley and his team are clearly driven by a deep personal commitment to the concept of open course content and its use to equalize access to knowledge in service of improving human condition. Their overarching goal is make a significant contribution to creating a global movement in this direction. They are not only producing functioning open source software, but also proactively marketing its adoption where ever they find opportunity.

I found the team to be very competent and fun to be around. The were very gracious hosts and the overall visit was interesting and good natured.

Although I do not have a copy of the original proposal they submitted (I have read the phase 2 proposal), my general impression is that they have made good progress to date.

There is much to commend about the project, the approach, and progress to date. In this written document, however, I will focus primarily on issues and concerns that I think deserve further attention. My comments are largely contextual, conceptual and strategic. These comments are given in a constructive spirit by someone also passionate about the general goals of the project and the “openness movement” in general.

Technical Critique

I will leave detailed technical critique to Rich and Brent. The software is built upon widely used, successful open source products Zope and Plone. The USU team software development philosophy is characterized by human-centered, iterative design including both in-house and external customers. My impression is that they work hard at getting user feedback to guide the iterative design. I think eventually they will need to give more attention to longitudinal studies of adoption and impact and perhaps bring in additional social science specialists to do so.

EduCommons

This work is strongly influenced and largely motivated by the MIT OCW Project. I know the MIT project pretty well, respect it, and have had extensive conversations with both Project Director Anne Margulies and former MIT President Chuck Vest about it. MIT OCW is an act of leadership and a type of self-assured benevolence on the part of MIT, and is having direct and indirect influence on many other higher education institutions world wide. Alliances of OCW activities are evolving. While all this is very valuable and the underlying policy decisions of MIT were bold, the project is in my opinion, only at the beginning in the application of cyberinfrastructure 1 to conduct of the mission of the university and must achieve levels of automation and reduced incremental cost to be more widely adopted.

The MIT OCW Project is also unique, or at least rare, in the level of external funding supporting it and the halo effect of the MIT reputation. How can other institutions go in similar directions with out large staffs investing an average of $15K per course? Does the world want material other than that provided by “the best”? And the bigger question for me: How can open learning resources and participation become a natural by-product (at zero or low incremental cost) of the routine use of cyberinfrastructure in service of learning, research, and societal engagement.

EduCommons is intended to reduce technical barriers and cost for creating MIT-type OCW websites, and to enforce a workflow model that supports quality control and scrubbing the content clean of intellectual property (IP) infringements. Initially at least it seems to focus on helping institutions move web-based course material open access with a more homogenous look and feel.

EduCommons is built on open source software components including Zope and Plone. It 1) uploads the output of an html authoring system (e.g. Dreamweaver) created by faculty and/or staff outside EduCommons; 2) expands the relative links to server-based links within a server database; and 3) enforces a workflow for processing the website through various levels of review, edit, and release; and 4) creates a transformed course website with a common, pre-defined look and feel. At least at present, the output is web-based course material intended to support traditional departmental-oriented courses.

The workflow model enforces a set of human roles with varying rights of review, edit, and publishing. This model is intended to provide an institution the means to assure academic and pedagogical quality, and also to manually attempt to assure that no material is used that violates terms and conditions of copyright or licenses. EduCommons is also the production environment from which course material is served to the world.

The philosophy of the OSLO team is that all resources emitted by EduCommons should be covered by a “educational” Creative Commons license, and that an institution will have people interacting with the workflow model to insure this is the case. There are not at present any tools provided to help those responsible know whether included material is copyright free or not. The USC need to be cautious about statements I heard along the lines that “everything coming out of EduCommons would be scrubbed IP clean.” That would only be the case if people, using the workflow model, made sure this was the case -- the system alone will not do the “scrubbing.”

This philosophy it seems to me carries the assumption that two different digital course resource systems would emerge within a university: one built entirely of creative commons material, and another built within the IP environment of the institution’s digital library/repository allowing access to copyright material to authenticated members of community.

Open Learning Support

The Open Learning Support (OLS) is intended to be software services to help people as individuals and teams make us of open content websites. At this point OLS consists of a threaded discussion group and some capabilities for rating an entry (kudo service) written by the team. This tool has not found extensive use to date but the team had some ideas about how this situation might be improved. There was also discussion about other capabilities that could be added to an OLS including reputation services and other capabilities of “social software.”

One of my overall recommendations (elaborated more fully later) is that this work needs to be better integrated with learning management systems, particularly those based on SAKAI. In this spirit, capabilities of an OLS should be compatible with or better yet, based upon these tools -- not standalone and distinct. The USC team needs to be careful to leverage open source work being done by others and not not invest is talents redundant work with respect related projects.

Utah State OCW Site

After the site visit I spent some time looking at the USU OCW site. The look and feel of the course material (produced I would assume with EduCommons) is good. I was also struck by the potential value of practical courses in irrigation systems (a real strength at USU) to developing countries. This helped me to realize that there could be real pay off from a diversity of specialized course material from a variety of institutions -- not just a few elite research universities.

Strategic Issues and Recommendations

I can well imagine that there is a market for the standalone EduCommon tool as it is now. The Utah team is being very bullish on encouraging adoption elsewhere. They may in fact eventually be overloaded with expectations for support by other institutions and therefore should give thought to how to scale up adoption in a less labor intensive way.

In the longer term, however, I think that open course material needs to be thought of as a special case of material available in a more general learning management system (LMS). The open content should ideally be derived at near zero incremental cost from the course material developed and supported within the institutional LMS. All open material (licensed by creative commons or on an open website elsewhere) is fully available and full citations are provided for other material that students in the open world can try to find as best they can.

But my hope would be that we could provide mechanisms so that over time more of the full content could be made legally available to the open world. Perhaps there will we ways to provide access to snippets and/or summarizations. Perhaps more journals will move to author-pay, open access reading. Perhaps Google and academic libraries will have success in moving orphaned material (in copyright, but out of print) to open access. Perhaps there will be legal ways to include broader consortia under fair use. There may at some point be a tipping point in the direction of more open academic context and if and as this occurs we need systems architected to pass on the fruits of this movement to the broader world.

My overall recommendation is that the USU work be linked and leveraged with other important work with the goal of better integration with the following:

Learning management systems based upon the SAKAI software environment for collaboration and learning.

Learning material authoring systems such as the Rice Connexions Project .

The idea of supporting seamless access to distributed but federated collections of open courseware also arose several times in the visit. I believe this is an important goal but if it is pursued in should be done in ways that leverage federated repository projects such as D-Space and Fedora. The USU team should also be aware of the open source Shibboleth Middleware Project and evaluate if and where middleware from this project, now being widely adopted in higher ed, might be useful in approaches to cross institution federated repositories.

In summary then, I think the USU group are competent and committed to the goal of broad access to learning opportunities. EduCommons and OLS are contributions to that goal, but I believe that they need to be strategically aligned and leveraged by a broader set of activities in this area, and not emerge as a standalone msystem and content repository.

Sharing Awareness of EduCommons

I intend to make the USU projects known to officials on the The Alliance for Equity in Higher Education with whom I have been working. This Alliance represents the nation’s largest association of minority serving institutions with a combined membership of over 350 institutions . I will also bring it to the attention of higher education officials in South Africa where I am providing on going consulting service through the W. K. Kellogg Foundation.