Implementing Mobile access to MOODLE at Athabasca University

Billy Cheung
Athabasca University 1 University Drive, Athabasca Alberta, CANADA T9S 3A3 billyc@athabascau.cay

Brian Stewart
Athabasca University 1 University Drive, Athabasca Alberta, CANADA T9S 3A3 brains@athabascau.ca

Rory McGreal
Athabasca University 1 University Drive, Athabasca Alberta, CANADA T9S 3A3 rory@athabascau.ca

Citation information: Cheung, B., Stewart, B., & McGreal, R. (2006, July) Going Mobile with MOODLE: First steps. IADIS International Conference Mobile Learning 2006. Dublin: International Association for the Development of the Information Society.

ABSTRACT
This research consisted of an investigation into some of the technical and organizational implications of implementing the open source Moodle Learning Management System (LMS) for use on mobile devices. The research has provided insights into determining modalities for such an implementation. It also offers guidance in facilitating distance course delivery using mobile devices. A mobile delivery implementation was effected using the Moodle LMS. The problems in the accomplishment of this goal have been noted and recommendations compiled to aid others in their implementations. The research has identified some of the critical relevant features of the mobile device environment and operating system. KEYWORDS Mobile learning, mlearning, learning management system, LMS, MOODLE,

1. INTRODUCTION
Athabasca University Canadas Open University, recently chose the open source Learning Management System (LMS) Moodle <http:/moodle.org> for use throughout the university (Stewart, 2005) This research included an investigation into some of the technical and organizational implications of implementing Moodle for use on mobile devices. It also provides guidance for other researchers as they investigate their use of LMSs and Moodle in facilitating course delivery in mobile learning. This research involved the actual implementation of Moodle using mobile devices. The problems in the implementation have been noted and recommendations compiled to aid others in their implementations. The researchers have identified some of the critical features of the mobile device environment and operating system.

2. LITERATURE REVIEW
Learning Management Systems Wikipedia defines a Learning Management System (LMS) as a software package, usually on a large scale (that scale is decreasing rapidly), that enables the management and delivery of learning content and resources to students. At a minimum, the LMS, sometimes referred to as a Course Management System or Virtual Learning Environment (VLE) usually allows for student registration, the delivery and tracking of e-learning courses and content, and testing, and may also allow for the management of instructor-led training classes. Web-based tools, common in such systems, include email, content forms, chat groups, resource page templates, grade books, and interactive quiz applications. It is designed to facilitate teachers in the management of educational courses for their students, especially by helping teachers and learners with course administration. (Wikipedia, n. d.). Learning Management Systems are sophisticated web-based applications for learning that are used in increasing numbers by institutions and companies (Avgeriou, Papasalouros, Retalis, & Skordalakis, n. d.). They emerged from the Internet, with early descriptions of heterogeneous distributed databases for distance education (McGreal, 1995) or integrated distributed learning environments (McGreal, 1998). Alberta government researchers called for a virtual learning system for the WWW as soon as it became widely accessible (Advanced Education and Career Development, 1995). Washburn (1999) conducted an early evaluation of the first LMSs to become available. Brennan (2001) recommended the use of LMSs for controlling the content and tracking student activities. Ismail (2002) stresses the management role of an LMS, distinguishing it from a Learning Content Design System, a Learning Content Management System (LCMS), and a Learning Support System (LSS). Case studies have been conducted on different implementations (Courtney & Patalong, 2002; Dean, 2003; Lyons, 2000). FerlFirst (2005) provides information on other implementations. Other comparison sites are also available (Commonwealth of Learning, 2003; Western Consortium for Educational Technology (WCET), 2005; Landon, 2005). Koopal (2005) argues for the need for a scalable VLE to meet the needs of the University of Twente. Mobile learning Waycott and Kukulska-Hulme (2003) studied the use of PDAs outside the classroom. They focused exclusively on students experiences with reading course materials and taking notes on PDAs. Using PDAs for reading and note taking was discovered to be less than ideal. Students were getting lost in the documents, and were unable to make notes as comprehensively and easily as they could with a paper copy of the materials. The technological capacity of PDAs has increased dramatically in the past three years. Screens are bigger and better; systems have more memory; they have more multimedia capabilities; and there are more refined methods for inputting data. Given the constant advancements in this field as well as the plethora of possibilities, such as PDA models with unique capabilities, and a wide variety of systems and applications. These wide ranging differences provide researchers with more questions than answers on possible methods for delivering digital content to PDAs. According to (Clyde, 2004) the challenge is to identify the forms of education and training for which Mlearning is particularly appropriate, the potential students who most need it and the best strategies for delivering mobile education (p. 46). Hoppe, Joiner, Milrad and Sharples (2003) want to move the focus of Mlearning from content delivery to interpersonal relations; there is an imperative to move from a view of e- and m- learning as solely delivery mechanisms for content (p. 255). Rather, the goal is to optimize the quality of service, e.g. in terms of availability across time and space or in terms of multimedia support (p. 255). These researchers agree that we are just beginning to realize the potential of Mlearning (Clyde, 2004). As a result, more research, from a user perspective, is needed to discover the best strategies for maximizing Mlearning, including discovering what are the optimal features of a PDA for accessing learning content, and what form the content needs to take, especially in relation to the optimal use of an LMS.

3. MOBILE MOODLE
Athabasca University Canadas Open University is focusing its applied research agenda on the use of mobile devices for distance learning. The new mobile technologies support the convergence of the two types of distance education technologies: information media and communications media. Information media carries the distance education content, not only in text but also in many multimedia formats (audio, video, graphic). Communication media include multimedia interactivity options supported by mobile technologies to allow for interaction between and among learners, teachers and community members using a variety of modes including audio, text and video conferencing (Hulsmann, 2004). Together and converged, these types of media allow for the emergence of social applications that allow learners to become aquainted, work collaboratively, schedule learning and in other ways enhance learning. So, in addition to increasing accessibility, mobile devices have the potential to increase connections between students, tutors, and instructors and decrease isolation. Distance education and more recently elearning is in a state of constant change. Mobile learning (Mlearning) can be seen as the next stage in the development of distance learning (Keegan, 2002). Mlearning, or learning made accessible through PDAs, mobile phones, ultra notebooks, or other portable wireless devices, maximizes the idea of anytime, anywhere learning. To date, some research from a students perspective has explored which mobile

Choosing Moodle at Athabasca University technologies are the best to employ (White, 2004; Bull & Reid, 2003; Rieger & Gay, n.d.), or what applications enable an effective digital library (Fox, 2005) and what support systems need to be in place for Mlearning (Laroussi, n. d.). Further work from a teachers perspective has shown that Mteaching can also be efficiently conducted using mobile devices (British Educational Communications Technology Agency, 2004) .

4. MOBILE DEVICE
This research was conducted with a Palm Treo 650 mobile smartphone which also supports email, an organizer, messaging (SMS), web access, and a camera. This device has a powerful user interface, including a QWERTY keyboard, a 5-way navigation button, a stylus pen, as well as finger navigation. . It also has Bluetooth and wireless web connectivity. Compared to PDAs and ordinary cell phones, smartphones are characterized as devices with on-demand connectivity to the web via network providers, with operating systems designed for installing additional software. The Treo 650 uses the Palm OS operating system with 10Mb Ram. There are 28Mb built-in, of non-volatile storage, plus a SD memory slot for data and applications. The screen size for this device is 320 * 320 pixels. This TREO 650 comes with a web browser Blazer v4.0 and an email client VersaMail. There are also third party browsers available for this device, for example, Xiino from Mobirus, and Opera Mini. Our initial findings show that neither of the third party browsers is significantly superior to the Blazer v4.0. Testing on this device was limited to the software originally shipped and supported by the vendor. We also did additional testing with PDAs running the Palm OS and Microsoft Mobile OS. Unlike mini-size, full feature ultra computers like the OQO, Sony Vaio PCGU1 or MS Origami running Windows XP, our research focus is the mobile operating systems as well as the user interface. These mobile operating systems demand less computing power thus less-expensive hardware. On the LMS side, all tests were conducted using the current stable version of Moodle (v1.5.3). The Blazer 4.0 browser tested in this research supports XHTML, CSS 1, and some Javascript. The browser also supports cookies and SSL connections. We tested the device with the W3 CSS 1 Test Suite and the W3 CSS Mobile Profile Test Suite, but found some incompatibility. Most significant is the lack of support for an embedded [object]. Once we modified and removed the use of [object], the device passes on almost all the tests for CSS1.

5. MOODLE
The default template for the Moodle interface consists of a three-column layout. When browsed with a smart phone, the page is rendered nicely as a one-column layout. This is a perfect example for good, fluid layout design that works for both desktop and mobile devices. Fluid layout refers to the ability to change the layout of a web page on the fly, as opposed to a fixed layout that is static and is the norm in most Web pages (Jackson, 2005). Technically, this is achieved by the proper use of [float] positioning of block elements where the width is set using a percentage rather than by an absolute [value] such as pixels, centimetres or point size. Although the default template for Moodle is very mobile friendly, content that is input into Moodle can be non-fluid thus reducing the usability. Content developers should avoid using block elements and block positioning where the width is set by an absolute value. Moodle also has an option to display external pages within a frame. The Moodle navigation controls are displayed in the top frame and the external pages appear in the main frame. Although web development experts often recommend against the use of HTML frames (Developers of the Niagara Guide Network, n. d.). The frame display on the Treo 650 is very readable. The browser formats frames in a linear, stacked fashion. Mobile devices and their browsers often come with a default sans-serif font (e. g. Arial, Tahoma). On small screens there are limited pixels, making it difficult to display serifs at small point sizes (8-12pt). Sans-serif fonts are more readable than serifs (e. g. Times, Palatino), on screens at these smaller font sizes, which are usually more visually appealing on screen. The Treo 650 and the other smartphones used in this research, as well as other mobile devices tested used default sans serif fonts. Moodle issues The Treo was used to test the official modules that come with the Moodle application. In the standard distribution of Moodle, there are thirteen types of activity modules. These modules are integral to the Moodle application, as each of them provides unique functionality for online learning. Additional activity modules are also available in addition to the standard distribution of Moodle. See: http://download.moodle.org/modules/ In this research, only the standard modules were tested. To test the mobile readiness of these modules, Moodle Feature Demo course developed by the Moodle community was installed. Some modules were tested multiple times, with different settings each time. In some scenarios, a small part of the module was not functional on a mobile device. We documented these issues in the Remarks column. The results are available in Table 1. Table 1: Moodle module testing

MODULE
Assignments: Upload File Assignments: Online Text

RESULT REMARKS
Fail Working Issue #1: No support for file upload on the test device. Issue #2: No support for HTML editor,

Choosing Moodle at Athabasca University answer can only be plain text or hand coded HTML. Assignment: Offline Chat Working Fail Issue #3: forms without submit button. Does not work like desktop browser which hitting the return key is equivalent to submit a form.

Choices: anonymous results Choices: non-anonymous results Choices: updatable Choices: with a limited number of responses allowed Forums: Standard

Working Working Working Working Working Issue #4: Indentation for threaded discussion is lost due to small screen size. Issue 0 Issue #1, Issue #4 Issue #1, Issue #4

Forums: Each Person Posts One Discussion Forums: Single Discussion Glossaries: Teacher-Defined Glossary Glossaries: Learner-Defined Glossary Glossaries: Glossary of common terms HotPot: Multiple Choice

Working Working Working Working Working

HotPot: Crossword Lessons: Quizzes: Simple Quiz

Fail Working need some workaround

Issue #5: Javascript related problems Issue #6: Encountered an outof-memory error due to page size and complexity Issue #5 Issue #7: The Javascript checking function in Moodle prevent the mobile device to proceed to the quiz. After bypassing the checking function, device has no problem doing, submitting, and reviewing the quiz. Issue #8: Javascript timer does not work. Issue #7 Issue #7 Issue #7

Quizzes: Simple Quiz with timer Quizzes: Example Listening Quiz Quizzes: A quiz with a password Resources: A Text Page Resources: A Text Page (markdown formatting) Resources: A Web Page Resources: A Web Page (opens in a new window) Resources: A Web Link (opens in a new window) Resources: A Web Link (framed) Resources: A Web Link (with parameters) Resources: A Directory Resources: A Label

Fail need some workaround need some workaround Working Working Working Working Working Working Working Fail Working

The multimedia files do not work on mobile devices.

Choosing Moodle at Athabasca University SCORM Packages: Surveys: Short Answers Surveys: Multiple Choices and Drop Down Menu Wiki: Workshops: Fail Working Working Working Working Issue #2 Issue #6

6. UPLOADING DOCUMENTS
The Treo 650 demonstrated the ability to write and edit MSWord or Excel files using Document To Go. However, our research found that the mobile browser does not support uploading files via a web form. We believe the lack of such an uploading function is caused by the incomplete implementation of the HTML element [input type=file]. Without separate support for file upload, the assignment module in Moodle will not function on the Treo or on devices using Windows Mobile OS. The discussion module will also be affected as students cannot upload files to the discussion forum usng a mobile device. With the open architecture in Moodle, it is relatively easy to extend the assignment and discussion module and offer an alternative method for uploading files. One example is to use email attachments instead of file uploading. An online photo album service called Photobucket is currently offering this method to enable mobile photo blogging. <http://www.photobucket.com/tutorials/mobile_upload/mobile.html >

7. SCORM OBJECTS
SCORM (Sharable Courseware Object Reference Model) is comprised of different technical standards that facilitate web-based learning, enabling systems to find, import, and export learning or content objects (Advanced Distributed Learning, 2001). SCORM objects use Javascript extensively, which is a significant weakness for many mobile browsers. Although the smartphone we tested does support Javascript in a limited way, it appears that most SCORM objects cannot be loaded directly onto a mobile device. This problem is not related to the SCORM specifications nor to the Moodle architecture. The problem occurs because of reliance on the SCORM JavaScript API Adapter as the bridge between SCORM and Moodle. To overcome this problem, using a Pocket PC, one can use the Pocket SCORM Run-Time Environment (RTE), a standalone application that enables the delivery of SCORM content objects to certain small screen devices (Lin et al., 2004).. This application suppresses horizontal or sideways scrolling and formats the information, stripping most graphics to allow for simple vertical scrolling. Without such an application, importing and activating SCORM objects written in Javascript is not possibleCurrently Pocket SCORM supports PDAs and smartphones using Windows Mobile OS but not the Palm OS, so it was not possible to test this on the Treo.

8. HOT POTATOES TESTING

Hot Potatoes is a popular educational software suite made up of six testing or quiz applications including interactive multiple-choice; short-answer; jumbled-sentence; crossword; matching/ordering; and gap-fill exercises for the World Wide Web. Like SCORM, this application makes extensive use of Javascript. Unfortunately, this renders Hot Potatoes unusable on most mobile devices. On the other hand, Moodle comes with a quiz module that displays very well on a variety of mobile devices. Institutions using Moodle and offering courses for use by learners with their mobile devices should consider porting the existing Hot Potatoes quizzes into Moodle quiz.

9. BEST PRACTICE
There are some documents and applications made available by the WorldWide Web Consortium (W3C) that provide guidelines for mobile content developers. The Mobile Web Initiative has drafted a Mobile Web Best Practices document (Rabin & McCathieNevile, 2006). We found the draft document to be very useful and is a must-read for mobile web developers. The Mobile Web Best Practices document can be extended it with some of the most commonly used Moodle modules and content in context, such as: Object and Embed These are HTML elements that include multimedia files within a webpage. Alternative, multimedia clips can be offered as a downloadable file. We found out that the browser used in this research does not support [embed] and [object] tags in HTML. The missing support for embedded object also implies the lack of plug-in architecture, which is needed to display multimedia objects within the browser. Flash Flash Lite player versions 1 and 2 are available online or come pre-installed on many mobile phones and PDAs running Microsoft PocketPC OS. In addition to the lack of support for embedded objects in general, there is no player available for Palm OS based mobile devices. Audio Audio file formats like mp3 can be downloaded and played on smartphones. Although it is not possible to listen to a streamed audio feed, there is nevertheless significant potential for offering learning content as audio files. Video Similar to audio, the best way to deliver video is to offer it as a downloadable file. There are also some limitations for using video on mobile devices. For example, the Treo 650 supports formats like 3GPP2 and ASF. The resolution of video clips should also be reduced to fit the screen size without wasting wireless bandwidth. One compensation is that videos on a smaller screen require less pixels to appear sharp.

Choosing Moodle at Athabasca University Images In the default settings, mobile browsers resize large images to fit the screen. This is a neat feature which reduces scrolling. However, the resizing is not always beneficial for learning purposes. Sometimes important details of the image are lost after a resize. Also, pictures of different sizes are resized to the same screen width. This may cause a problem when a learning task requires the comparison of multiple images. We conducted tests attempting to maintain the original size of images shown in our smartphone, by providing the dimension in HTML and wrapping the image in a fixed size container, but without much success. Background colour and image Background colour and image are sometimes not rendered on mobile devices. If the page is viewed via a proxy server, these formattings are stripped away and a generic style is applied. Mobile content developers should bear in mind that the pages they develop should work without background color and image. There are also similar recommendations from section 5.3.7 of the Best Practice 1.0 Draft.

10. CONCLUSION
With the default settings, Moodle is already a very mobile friendly LMS. With the smartphone we used in this research, most Moodle pages will display appropriately. Course developers using Moodle should be aware of the limitations of mobile devices and the type of content they put into Moodle. We found those Moodle modules that rely on Javascript are not working. Also, mobile devices are still lacking support for multimedia and interactive learning objects. We also pointed out some different behaviors on mobile devices and their potential consequences. The research included an investigation into some of the technical and organizational implications of implementing Moodle for use on mobile devices. From a user's perspective, the research has provided us with experience in determining modalities for such an implementation. It also provides guidance for other researchers as they investigate their use of LMSs and Moodle in facilitating course delivery in mobile learning. To summarise, a mobile delivery implementation has been effected using the Moodle LMS. The problems in the implementation have been noted and recommendations compiled to aid others in their implementations. The researchers have identified some of the critical features of the mobile device environment and operating system. Athabasca University now supports the Moodle LMS and it is available to students and faculty from wherever they may be using a variety of mobile devices.

RELEVANT WEBSITES
Mobile devices Mobile/cell phones in education http://m.fasfind.com/wwwtools/m/2717.cfm?x=0&rid=2717 Moodle for Mobiles Project http://moodle.org/mod/forum/discuss.php?d=33033 SCORM http://reflectivesurface.com/weblog/2003/08/07/implementing-scorm-advantages-and-disadvantages Flash Lite for mobile device http://www.macromedia.com/mobile/ Mobile Web Initiative http://www.w3.org/Mobile/ Open Mobile Alliance http://www.openmobilealliance.org/ Scope of Mobile Web Best Practices http://www.w3.org/TR/2005/NOTE-mobile-bp-scope-20051220/ IADIS http://www.iadis.org/ml2006/dates.asp Mobile Web Best Practices 1.0 Working Draft http://www.w3.org/TR/2006/WD-mobile-bp-20060113/ Moodle information A list of articles about Moodle: http://moodle.org/mod/resource/view.php?id=102/ Humboldt State University website Moodle introduction: http://learn.humboldt.edu/login/index.php A Blackboard Moodle comparison: http://www.humboldt.edu/~jdv1/moodle/all.htm Basic corporate page: http://www-306.ibm.com/software/lotus/ School of Business demo page: http://sb.athabascau.ca/course/demo.nsf Main corporate webpage for WebCT: http://www.webct.com/ ECMA Standards Organization, 2000, Edition 3, ECMAScript Language Specification, http://www.mozilla.org/js/language/E262-3.pdf

REFERENCES