<a name="1_0"></a>RUSSEL

RUSSEL, (Reusable Support System for E-Learning) can be thought of as a lightweight open source learning content management system (LCMS). It includes a learning object repository that can be used to organize and tag digital assets for use in online training environments and a content discovery and assembly tool that can be used to add assets to vetted instructional design templates and to pass them on to developers and programmers for use in creating courseware. RUSSEL is intended for use by Subject Matter Experts (SMEs), Instructional System Designers (ISDs), training managers, training suppliers, and others involved in the creation and management of digital learning objects. It is integrated with the Learning Registry (LR), a national registry of online learning objects that can be accessed at http://learningregistry.org, and can be integrated with other registries, learning management systems (LMS), authoring tools, and repositories. With recent improvements made to this project, RUSSEL now includes:

User and group management and tools for creating collections of digital assets
Tools for importing and exporting SCORM packages
Tools tagging objects with metadata relevant to their use in military and other instruction
Tools for using resources in instructional design templates
Automated registration of objects, their metadata, and paradata in the LR. Paradata includes user ratings and comments (supported by RUSSEL) and usage data collected by RUSSEL.
APIs for integration with registries and repositories.

<a name="2_0"></a>Results of the RUSSEL Project Refactoring

RUSSEL was originally developed using the open source version of Alfresco as a back-end. The thinking was that an existing enterprise content management system could provide much of the required core functionality, freeing the RUSSEL team to focus on improving the User Interface and User Experience (UI/UX) and developing the specialized components that support asset reuse and instructional design in a DOD setting. Alfresco was selected after an extensive environmental scan and decision process, in part because a commercial (non-open source) version was also available and in DOD use. The use of Alfresco did, in fact, permit more progress to be made on innovative portions of RUSSEL, but several key limitations of Alfresco were exposed in the process. Overcoming these limitations required developing separate Java beans that interacted with Alfresco, and the resulting application still had limited ability to expose functionality through the equivalent of an enterprise service bus (ESB). The re-architecture of RUSSEL (under the DECALS project) using LEVR solved these problems and has made RUSSEL more extensible.

<a name="2_0_1"></a>Repository Services

RUSSEL offers standard learning object repository services. These include “Create, Retrieve, Update and Delete” (CRUD) services, version control, previewing, and meta-tagging. A tile-based drag-and-drop interface gives RUSSEL a different look and feel. The important aspects of RUSSEL as a repository are:

It is open source;
Other systems, including PALs, can access their functionality through RESTFUL web services; and
It contains tagging features that specifically support DOD education and training applications.

<a name="2_0_2"></a>Metadata Services

In RUSSEL, when you add an object, you are automatically taken to a screen that gives you the opportunity to edit its metadata. Some fields are pre-populated with automatically extracted metadata, and both systems are designed to integrate with third party systems that can automatically generate more fields. The existing keyword extraction uses Eduworks’ proprietary keyword generation service that is licensed to the ADL for use in PAL projects together with other semantic services. RUSSEL extracts resource types, size and duration (keyed to learners at a high school reading level), and auto-generate descriptions of resources from other extracted metadata fields. These descriptions are not high quality but serve as a starting point that can be edited, an approach that is called semi-automated metadata generation in the literature.

The RUSSEL project has emphasized automated and semi-automated extraction of metadata over purely manual entry. There are two main reasons for this. First, Authors and instructors tend not to add metadata to resources when they create them or add them to repositories. Users generally dislike filling out long forms when there is no clear personal benefit, and requiring extensive metadata serves as a barrier to submission. Second, automated methods can be more consistent than manual ones, especially with fields such as size or reading level that can be calculated directly or using a combination of formulaic and machine learning methods.

<a name="2_0_3"></a>Paradata

Starting about ten years ago, the notion of paradata started creeping into the vocabulary used by digital librarians and the LR. Originally used to describe survey data, paradata is now used to describe “social metadata” ranging from ratings and comments to usage statistics. In simple terms,

Metadata makes assertions about an object.
Paradata makes assertions about the use of an object.

RUSSEL collects paradata in the form of ratings, comments and usage statistics. This system allows users to rate and comment on resources. RUSSEL tracks where resources are used in instructional design templates and uses this as part of its internal search criteria. It also tracks how often resources are added to collections or projects.

This paradata is maintained internally and can be manually published to the LR. It is not automatically published because not all users may wish paradata for all objects to be stored in the LR and because pushing it to the LR on every update would use a lot of server bandwidth as currently implemented.

<a name="3_0"></a>Instructions for Setting up RUSSEL

These are instructions for getting the RUSSEL System installed and running on a VMware VM running Ubuntu Linux version 14.04.3. These instructions can also be used to setup the system software on a server running Ubuntu Linux version 14.04.3.

<a name="3_0_1"></a>Code projects in this Github repository required to build RUSSEL:

russel-ui
EwGWTLib
LEVR
scripts
- base-v2
- base
- russel

<a name="3_0_2"></a>Prerequisites:

VMWare Player 7.1.2 or higher OR
A web server running Linux
Ubuntu Linux 14.04.3 64 bit
Java 7
Tomcat 7
Solr
Eclipse Kepler v4.3 for Java EE Developers 64-bit and the GWT plugin

Download and Install VMware Player 7.1.2

<a name="3_1"></a>VMware VM settings:

CPUs: 2
RAM: 4GB
Vitualization Engine: Intel VT-x/EPT or AMD V/RVI with Virtualize Intel VT-x/EPT or AMD V/RVI box checked
VM OS: Ubuntu 14.04.3

<a name="3_2"></a>Install Ubuntu OS

Download Ubuntu 14.04.3 and install in the new VM

Ubuntu 14.04.3 LTS 64-bit installation in VMware player 2015

<a name="3_3"></a>Install Eclipse Development Environment

Download and install Eclipse Kepler v4.3 for Java EE Developers 64-bit on VM

Install GWT Eclipse Plugin

<a name="3_4"></a>Install Java 7:

sudo add-apt-repository ppa:webupd8team/java
sudo apt-get update
sudo apt-get install oracle-java7-installer
sudo apt-get instal

RUSSEL

Install / Use

README

Table of Contents