Network Infrastructure

1. What Is LON-CAPA?

• LON-CAPA stands for LearningOnline Network with a Computer-Assisted Personalized Approach.
• LON-CAPA is a distributed Learning Content Management and Assessment System.
• It is a project being developed and implemented by an enthusiastic group of faculty and professionals who recognize that Information Technology can play an important role in improving students' learning and understanding.
• Current members of the project include faculty from universities, colleges, and K-12 schools in the US, Canada, Asia, Africa, and Europe. The core development group is located at Michigan State University.
• The LON-CAPA software provides instructors with a common, scalable platform to assist in all aspects of teaching a course, from lecture preparation to administration of homework assignments and exams.
• It allows instructors to create educational materials and to share such learning resources with colleagues across institutions in a simple and efficient manner.
• It provides a sophisticated assignment engine that can create unique homework assignments and examinations for every student in a class.
• Its formative and summative assessment tools grade a broad variety of objective problems and assist in evaluation of essays.
• It provides prompt feedback for students and instructors, as well as statistical information on performance and on effectiveness of materials
• The LON-CAPA software is freely available and free (GNU General Public License), and may be modified and adapted.

Get a Demo Student Login

2. Content Re-usage and Granularity

Learning resources could be simple paragraphs of text, movies, applets, CAPA-style individualizing homework problems, etc. In addition to providing a distributed digital library with mechanisms to store and catalog these resources, LON-CAPA will enable faculty to combine and sequence these resources at several levels: An instructor from Community College A could combine a text paragraph from University B with a movie from College C and an online homework problem from Publisher D, to form one page. Another instructor from High School E can take that page from Community College A and combine it with other pages into a module, unit or chapter. Those in turn can be combined into whole coursepacks. Faculty can design their own curricula from existing and newly created resources instead of having to buy into a complete off-the-shelf product.

General overview of the resource assembly mechanism and the different levels of content granularity supported by the current implementation of this principle. The topmost puzzle piece represents a resource at the fragment level — one GIF, one movie, one paragraph of text, one problem, or one regular web page. Attached to the resource is metadata gathered at the publication time of the resource. Using the resource assembly tool described below, these fragments and pages can be assembled into a page. A page is a resource of the grain size which would be rendered as one page on the web and/or on the printer. Using the same tool, fragments (which would then be rendered as standalone pages), pages and sequences can be assembled into sequences. Sequences are resources which are rendered a sequence of pages, not necessarily linear. Examples are one lesson, one chapter, or one learning cycle On the third granularity level, fragments (rendered as standalone pages), pages, and sequences can be assembled into courses. Courses are a sequence which represents the entirety of the resources belonging to a learning unit into which learners can be enrolled. Examples are a University one-semester course, aworkshop, or a High School class.

To increase the utility of the materials, the number of hard-coded hyperlinks between the resources should be minimized. The actual combining and sequencing is part of the system functionality and driven by external "roadmaps", which are constructed by the instructors. With this mechanism, one and the same resource can be part of different courses in different contexts. An example from LectureOnline is a set of homework assignments on nomenclature of organic molecules that was constructed for an introductory organic chemistry class and re-used both in a food science and a biology class by different instructors. The soft-linking made it possible to import only the desired set of online assignments without effectively importing additional parts of the organic chemistry course through hard-linked menus or "next page" buttons that might have resided on that assignment page.

3. Curriculum Adaptivity

The electronic roadmaps also allow for conditional choices and branching points. The actual path through and presentation of the learning resources is determined by instructor-specified combinations of learner choices and system-generated adaptations (for example, if the learner does not pass a test, additional resources may be included). Each learner can have an individualized curriculum according to preferences, capabilities and skills.

Adaptivity Different students will be taking different pathes through the material.

These maps can be generated at different levels of granularity with a graphical tool. To generate higher-level maps, maps can be inserted into each other; in fact, a map is just another learning resource itself.

Any faculty participating in LON-CAPA can publish their own learning objects into the common pool. To that end, LON-CAPA provides a "construction space" which is only accessible to the author, and a publication process, which transfers the material to the shared pool — during the publication process, metadata [18,19] about the resource is gathered, and system-wide update notification and versioning mechanisms are triggered. To handle royalties and intellectual property issues, LON-CAPA will include Intellectual Property Rights Management (IPRM) mechanisms. An appropriate entity would have to handle the actual money flow, where the profit is generated from the sale of "virtual course packs" to learners for the network - either through a course materials fee or through standard distribution chains in campus bookstores.

4. Instructional Management System

To make LON-CAPA a valid Instructional Management System, it finally has to include spreadsheet and database functions to handle enrollment and grades, as well as tools for learner-learner and learner-instructor communication and collaboration. Optional links to existing administrative systems, such as those of the registrar's office, have to be implemented individually for every participating institution, while Kerberos [24] as an optional authorization mechanism is part of the distribution. Necessary management capabilities have been defined in a faculty focus group at Michigan State University in the past two semesters, and will be refined during beta-test.

While not necessary for the function of the resource pool, administrative and management capabilities as part of the overall system functionality greatly enhance immediate usability of the resource pool and overhead.

5. Scalability

Physically, the network consists of relatively inexpensive upper-PC-class server machines which are linked through the commodity internet in a load-balancing, dynamically content-replicating and failover-secure way.

There are two classes of machines in the network: library servers, which hold users' records and authoritative copies of resources, and access servers, which host sessions. Every participating institution needs to contribute at least one library server. For small institutions, a library server can double as access server.

From experience with LectureOnline, it is expected that a $2,000 access server at today's performance can handle up to 1,000 concurrent sessions. The network is designed so that the number of concurrent sessions can be increased over a wide range by simply adding additional access servers before having to add additional library servers, which at today's performance should be machines in the $5,000 range. Preliminary tests showed that a library server can handle up to 10 access servers fully parallel.

6. Open Source Code Base

Even by a conservative measurement of "internet time," this project runs for the equivalent of twenty years. It cannot be expected that the Michigan State University team of developers alone will be able to sustain the code-base of LON-CAPA (expected to be around 60,000 lines of code) over this duration without input from a broader development community. In contrast to a proprietary model, the internet-mediated cooperative open-source community of programmers and users has a fast turn-around on debugging and adaptation to new technologies, and usually leads to a broader adoption of the product (as demonstrated by the success of the Linux operating system). We especially expect this model to be successful for a system that operates in the academic community.

We intend to license the code-base under the GNU Public License (GPL) Version 2 [25], and open the code-base to any interested developers. MSU and eventually lon-capa.org will maintain the official code-base in an openly accessible repository (Concurrent Versions System (CVS)) [26]. This would parallel the operating system market, where commercial operating systems such as Windows, MacOS and Solaris, have recently found an open-source counterpart in Linux; the instructional management component of LON-CAPA would form an open-source counterpart to commercial systems such as WebCT and Black Board [5-7].

7. Non-Profit Service Organization, lon-capa.org

The project can only be successful if an interconnected network of faculty, academic institutions, and publishing firms can interact in a way that is sustainable, and the administration of Michigan State University is committed to guarantee this sustainability beyond the duration of this grant proposal.

The best way in the current thinking of the PIs is to create a business entity, which will take over the operation of the network at some point in time during the later phase of the grant. We envision a non-profit entity, which in many ways will act like a traditional custom course-pack publisher, yet operate on a non-profit base within the academic realm.

This proposed and to be investigated entity, lon-capa.org, would provide the necessary book-keeping, collect royalties from resource users and distribute them to the resource providers, handle marketing and PR, as well as keep the network itself running, provide technical support, and maintain the code repository for the tool. It would most likely be housed within the Michigan State University technology park.

While this scenario currently seems the most workable, part of this proposal is to investigate alternatives and/or details in close collaboration with the Business Advisory Board, the pilot users, and Michigan State University.

For example, the devotion of a fraction of MSU content royalties to project continuation would be another workable solution.

As this proposal is written, many institutions of higher education start to develop policies regarding the intellectual ownership of online instructional resources, which is very much still an open question in an academic culture which is increasingly striving to operate profitably [e.g., 27,28]. Intellectual property developed under this project will be disseminated in accord with the current policies of the Board of Trustees of Michigan State University. One of the opportunities this project offers is to bring together a diverse set of stakeholders to find workable and sustainable ways to operate in this new realm of academic publishing.

8. Regular Workshops and Conferences

We will offer regular technical workshops and user conferences to facilitate communication among the users and between the users and the developers. While the workshops will focus on technical aspects of the system and eventually be used as a forum also for the open-source development of the tool, the conferences will give an opportunity for educators to share about their experiences with the system in teaching and learning.