Oren’s Weblog

I realized that I’ve been carrying around some more impressions from the Sakai Educational Partners Program meeting last week in Denver without communicating them.

While it’s still too early to assume that the Sakai effort to create an open source collaborative learning platform will be ultimately succesful, there is certainly a lot of momentum gathering behind the project. It seems like there is a fairly solid (though still evolving) technical architecture being created, a lot of smart people from major institutions around the country are thinking about how to contribute specific tools to the project, and (perhaps best of all) the community is forming around the project and learning how to express common concerns and work towards common goals.

While I still have some major concerns, most notably the thorny issue of how really good teaching/learning tools that have been created in languages other than Java (like our own Catalyst tools from the UW) can be integrated into this platform, I think that there’s enough there there to warrant spending a considerable amount of time and effort with the Sakai folks.

Mara is a beekeeper, and she points out that her experience in listening to the level of the buzz is applicable to her work on the Sakai board …

Berkeley is planning to use their new gradebook in pilot mode in Fall 2004, and then integrate it into Sakai. It has been developed as a Sakai application, but for the initial deployment they are wrapping the single tool with a “psuedo-Sakai” layer.

MIT is also developing a Sakai gradebook. MIT and Berkeley will look at possible convergence at some point.

Berkeley has a nice graphing option in the gradebook that shows a box-and-whisker graph of grade distributions. If you’re only looking at a single student it shows you a bar graph instead.

The psuedo-sakai approach uses JSF, Hibernate, and Spring as basic technologies.

ETUDES is the course management system used at Foothill-DeAnza College, developed by a Computer Science professor in 1995. Foothill created the Etudes Alliance in 2002 to ensure long-term sustainability of ETUDES for community colleges, mostly in California. They began a redesign and redevelopment of the software in 2002.

They undertook a needs analysis among faculty, and identified a list of some 50 features that would be necessary in a CMS for community colleges – about 30 of those are present in the initial release of Sakai.

In 2003 Hewlett Foundation approached Foothill with an invitation to develop something equivalent to MIT’s Open Courseware Initiative – they went back to Hewlett to say that in order to do this they need to have a common CMS platform for the content, at that point proposing developing ETUDES as an open source effort for community colleges.

They ended up with a scaled back program to create 20 sets of open courseware during 2004 – created an organization called SOFIA – (Sharing of Free Intellectual Assets). In addition they submitted a second proposal to extend Sakai to meet the needs of community colleges, which Hewlett funded. They are contributing two developers into the Sakai effort. Their priorities are designing/building a simple content authoring tool, migrating current ETUDES users to Sakai, and faculty training adn support.

Willie points out that there are pressures on the Sakai effort that need to be addressed. The formation of alliances with other areas is critical – e.g. bringing Educause more into the effort. It’s more than just technology – there has to be a strong communication layer to this effort, both among the partners and outside – we need to work together to unify efforts to avoid unnecessary duplication.

Chris Coppola from the rSmart group is talking about the Open Source Portfolio Initiative

OSPI is currently at release 1.5 – was derived from U of Minnesota’s software. It’s a Java-based piece of software.

The big realization for me, about two minutes into Chris’ talk is that an electronic portfolio is centered around an individual – not a course, not an institution. The portfolio will ideally persist throughout an individual’s career and will want to move with the person as they wander through their life. Portability that would enable this kind of persistence is one of the visions for OSPI.

OSPI is up to about 1,200 members in 77 countries – mostly people looking at the demo, “kicking the tires”. The reality is that they don’t know a lot about what people are doing with the software.

Version 1.5 features an XML/XSLT presentation engine.

For version 2.0, schedfuled for Spring of 2005, they have received funding from Mellon, U of Indiana, and rSmart. One object is to make the development process more transparent to the community. Individuals will be able to subscribe to some number of “common interest groups” which will provide structures. Examples of “common interest groups” might be a chess club, or undergraduate biology majors, or (in Indiana’s case) the set of common rubrics (large scale student learning objectives) used for undergraduate education.

The goal is to have the Portfolio as a Sakai tool.

SAMIGO is an assessment tool that uses an Asynchronous, Web-based Interaction model:
- instructor asks questions
-student responds; gets feedback (immediate or delayed)
- instructor grades and makes comments

A Sakai tool for creating, distributing, taking, and grading assessments. It grew out of previous work at Stanford and Indiana.

Features an item or question bank for instructors to keep their questions to pull from.

It allows for file upload and audio recording as types of question types (in addition to all the usual suspects).

They decided to support the IMS Question and Test Interoperability spec as their base data schema.

I’m at the first Sakai Educational Partners Program meeting in Denver.

Ken WIner is talking about uPortal and it’s relationship with Sakai.

uPortal is the open source portal software that is a project born from the Java Architectures Special Interest Group (JA-SIG). uPortal began in December 1999 and has a lot of traction in the higher ed community (over 130 implementing organizations in at least 13 countries).

Two emerging standards for portlets (portal channels) – JSR168 (a java community process) and WSRP (a web services approach).

WSRP is a presentation-oriented web service – once you get the SOAP message back from the web service it already contains markup so it can be thrown right to the screen.

You can imagine uPortal acting as a WSRP consumer or producer. It allows the portal to communicate with other applications across heterogeneous technologies.

Pluto is the Apache project that is a portlet container, that is used in uPortal. In uPortal 2.3 (the current release) you have to use a “Portlet Adapter Channel” in uPortal.

SAKAI has thre different versions of interaction with uPortal:

- Embedded – Sakai runs as a portlet within a stock uPortal
- Injected – Sakai tools run as individual channels within a modified version of uPortal
- Integrated – Sakai tools able to run as inter-communicating processes within a future stock version of uPortal.