some notes I took at a presentation in Soda:
History
1992 - CAPA
X-Windows problem editing
Got Web student interface in 95
Informal sharing of content (ftp)
1997 - LectureOnline
Learning content management and individualized assessement
system for science and math
Sharing of content between courses
Completely web-based interface
Historical problems
- Culture of sharing content between institutions had developed, but
not reflected in archicture - Content granularity was one grain-size: one problem, one page at a
time - Scalability in peak workload times
- Failover security - single point of failure
- Insufficient cataloging (reuse means finding them)
1999 - LON-CAPA
- Sharing between courses and institutions
Today
- over 20 partner institutions
- 43,000 course enrollments per year
- useful and usable for "casual" faculty user outside science and math
- LON-CAPA is a full-feature course management system
- GNU GPL
Distributed
- geographically distributed network of constantly-connected servers
- Access servers: host user sessions-processing. Need no backup, etc. Automatic load balancing (7 at MSU)
- Library servers: every user and course has a home server in the network which holds all of their resources and data-storage backend. Library servers can double as access servers.
- if MSU servers are full, MSU students redirected to NDSU
Logical domains
- The network is logically divided into domains such as "MSU", "FSU", or "Publisher X"
- Domains limit flow of user information
- Domains can limit access to content resources (site license access)
- Domains limit the extent of user privileges
Authoring
authoring env is XML
an e-mail I received:
Subject: update about the LON-CAPA course management system From: Burks Oakley II <oakley@uillinois.edu> Date: Mon, 26 Jan 2004 09:13:40 -0600 X-Message-Number: 1 Greetings! Last week, I attended the LON-CAPA users conference in Washington, DC. LON-CAPA is an "open source distributed learning content management and assessment system" -- see: http://www.lon-capa.org/ LON-CAPA is an acronym, which stands for "Learning Online Network with a Computer Assisted Personalized Approach." I'm really very impressed with this system, which was supported earlier by the Sloan Foundation and the Mellon Foundation, and most recently by a major grant from the National Science Foundation. This project really has come a long way and it has huge potential. Here are some important thoughts: 1. LON-CAPA is a course management system - it has all the features of WebCT, Blackboard, Angel, etc., including a discussion forum, e-mail, calendar, gradebook, user authentication, etc. It supports a number of different languages (French, German, Spanish, Russian, etc.). 2. LON-CAPA is open source and distributed for free. It runs on generic Linux servers. Note that the annual license fees for WebCT and Blackboard at most institutions are now many tens of thousands of dollars, and increases yearly, which is inhibiting the growth of ALN. Especially when the licenses are often "per student enrollment". 3. Because it is open source, software developers around the world are developing and sharing code for it. A great example shown at the conference was from a chemistry professor in Israel who developed a really wonderful set of tools for students to use in entering chemical equations (and for the system to check if the equations are correct). He also developed a system for drawing complex organic molecules, and checking (in software) if they are correct. He then gave all his code back to the developers at Michigan State, and they integrated into the next release of LON-CAPA. Now people across the country are using these tools in their online courses. 4. LON-CAPA was designed around the management of "learning objects", such as homework problems, videos, audio clips, animations, applets, etc. The authors of these objects can publish them, and other users can download them directly into their own LON-CAPA courses. They also can build a collection of learning objects, and share them. By using material developed at a number of institutions, this has gotten around some of the faculty concerns of "not invented here." Sharing collections of objects extends to the course level also, so that someone at Florida State can download the entire introductory chemistry course from Michigan State. This means that there can be de facto national standards for the types of homework, quiz, and exam problems that students are expected to solve. Unlike MERLOT, the LON-CAPA learning objects all are written in the same language, which means they can be imported directly into any course. 5. LON-CAPA is being used extensively in high schools - mostly in Michigan, but other states as well. Today's high school students love learning online - and the MSU folks say that they now have some freshman entering with experience using the LON-CAPA system - what a great way to keep students in the math-science-engineering pipeline! The LON-CAPA folks are interested in working more closely with the Sloan Consortium. First of all, having institutions adopt this learning management system could lead to cost savings across the consortium (through the avoidance of license fees). It could lead to more course sharing and exchange, similar to Project SAIL at the League for Innovation, which was funded by the Sloan Foundation -- see: http://www.league.org/league/projects/sail/index.htm ). It could be used with blended learning and with completely online courses. It would give educators more control of the educational environment (such as adding the functionality for chemical equations and drawing molecules). It could address important issues such as the student pipeline in math-science-engineering and gender (women show higher increases in learning from LON-CAPA than men do). If you are interested in learning more about LON-CAPA, you should feel free to contact Prof. Edwin Kashy, who is the Co-Principal Investigator of the LON-CAPA project. His e-mail address is: kashy@nscl.msu.edu Prof. Kashy and his colleagues have published several papers about this system in the Journal of Asynchronous Learning Networks, and they also have published a Sloan-C effective practice -- see: http://www.aln.org/publications/jaln/v7n1/v7n1_kashy.asp http://www.aln.org/publications/jaln/v4n3/v4n3_kashy.asp http://sloan-c.org/effective/details4.asp?FS_ID=45 Please let me know if you have any questions about what I have written here. Best regards, Burks