An E-learning 2.0 Environment – Principles, Technology and Prototype Tomáš Pitner (Masaryk University Brno, Czech Republic
[email protected])
Pavel Drášil (Masaryk University Brno, Czech Republic
[email protected])
Abstract: With the current shift in web technology called Web 2.0, many e-learning experts compare the traditional view of e-learning with the expected level of e-learning services and find common problems with present LMS platforms. Primarily, we will identify the most important properties of an E-learning 2.0 platform. Secondly, a prototype, Java-based opensource E-learning 2.0 platform L2 will be introduced and evaluated. Keywords: E-learning 2.0, LMS, Web 2.0, blog, wiki, REST, XML, Adaptive XML Inclusions, L2 Platform Categories: D.2.11, D.2.12, I.7.2, I.7.4, J.7, H.1.2, H.3.5, H.5.4
1
What is Web 2.0
Several prominent authors [O’Reilly, 05] currently identified the significant changes in the present trends in web use and technology. The most successful applications and services on the web like wiki, weblogs, Flickr, or many Google’s services (Gmail, Google Maps and others), have remarkable common characteristics: serviceorientation, collaborative nature, non-stickiness, and focus on data. O’Reilly outlined the core characteristics of a company (or service) that can wear the label Web 2.0: • It provides services, not packaged software, and with cost-effective scalability • It has control over unique, hard-to-recreate data sources that get richer as more people use them • It trusts users as co-developers • It harnesses collective intelligence • It leverages the long tail through customer self-service • The software is above the level of a single device • It has lightweight user interfaces, development models, AND business models The transition to Web 2.0, however, is not revolutionary. Legacy applications and services – the old fashion ones, even if they are not “Web 2.0 compliant” at all – can coexist further with the new ones. Web 2.0 is about its daily use. You use web applications in the work, at home, to communicate and share with friends, or just for fun. The web computing becomes
pervasive. The services can more and more be accessed from mobile devices on the go. As e-learning is mostly represented by web applications and services, it must cope with the Web 2.0 trends.
2
What is E-learning 2.0
E-learning experts [Downes, 06] compare the traditional view of e-learning with the expected (Web 2.0) level of e-learning services. Currently, e-learning is represented by: • (Reusable) learning objects that can be assembled together. • E-learning takes the form of online courses. • It is controlled by Learning Management Systems (LMS) that are usually monolithic, large-scale systems. • In the optimistic case, they are integrated with the institution’s information system. • There are already several good free, open-source LMS implementations. • LMS implement traditional theories of distance learning [Moore, 91] in online world. They aim at content organization and delivery; they provide training supported by quizzes, tests and discussions. However, many authors noticed that e-learning and LMS, as characterized above, does not fulfil current needs in pedagogy: • Learner- (person-/student-) centred learning – the student is not just the subject of teaching. They dislike being controlled by a LMS and live in the LMS world – closed and unreal. They learning must be fully in the hands of learner. The student more and more becomes a client of the learning service. • While Constructivism tries to force the learner to understand by meaningmaking tasks, it is not enough in the present chaotic world. The concept of Connectivism builds on finding the right connections. Again, legacy closed LMS do not help much in this aspect. • Learning at a higher level is not about gathering pre-existing knowledge anymore – and that was the area where LMS were successful. Web 2.0 leads to a new e-learning paradigm that is significant enough to be called E-learning 2.0 [Downes, 06]. In this paper, we will primarily try to judge why legacy e-learning platforms do not fulfil current needs. Consequently, we will present our vision how e-learning should be shaped (re-thought) to fit to the E-learning 2.0 era.
3
Where Legacy LMS Do Not Meet Current Needs
In this section, we will illustrate that present LMS – including the very popular opensource ones – belong predominantly to the Web 1.0 Generation – and do not meet current e-learning needs any more.
3.1
Weak Integration with the Internet
LMS do not exist in an isolated space, in their own universe. No matter how good content they provide, how sophisticated tools they offer, the users at both sites – teachers as providers, students as consumers – live and work in a much richer world. They use resources out of the LMS, generally accessible materials such as glossaries, Wikipedia pages, or just external web sites. They may need to discuss issues in larger communities via chat and not to be limited to the world of one LMS. They share documents or photos not just for learning, but also in their personal life. 3.2
Current LMS – Products, not Services
Even the open-source development results in a product that needs to be installed, maintained, patched, updated, upgraded – not in a service that is just used. Too liberal policy in contributing to an open-source project leads often to chaotic incorporation of many features that may be unstable, not mature enough or not well documented. Maintenance of such system might become a nightmare – especially for smaller educational institutions without a strong infrastructure and experienced technical staff. 3.3
Authoring Too Centralized
Legacy LMS are centralized – they resemble more Britannica Online than Wikipedia. Current experiences in e-learning clearly show that collaborative authoring systems such as Wikipedia offer better learning content than any closed team of authors – no matter how large and experienced the authoring team is. It is especially true in areas of common interest, in fields that are taught the same way all over the world – such as the computer science, or science in general. 3.4
Publishing and Content Management Not Dynamic Enough
E-learning is not about creating and publishing stable content in one step. It is about continuous improvement, modifications and quick publishing. With legacy LMS, publishing a resource often means creating it in an external authoring tool, uploading it into the LMS and associating necessary metadata with it. Any change in the learning material is cumbersome – it requires non-creative, repetitive work. As a consequence, the teacher often resigns and leaves the “stable” material untouched, including errors. Even publishing to personal web sites is not optimal. It is still too static, and requires specific knowledge of HTML and associated technologies. Full content management functionality is not needed in the e-learning area. The same applies for the content organization. The legacy CMS tend to organize documents into hierarchies, pre-designed by a “guru” who must have known the structure in advance. Once an e-learning course starts, the learning community tends to semantically self-re-organize the learning content by tagging it. This is called “folksonomy”, in opposite to (predefined) taxonomy.
4 4.1
Shaping an E-learning 2.0 Platform Introducing L2 Platform
The core idea is to let the users – both teachers and students – to do their job as ever before, not to force them to throw away tools they like and use, but to integrate them. To demonstrate and further investigate a platform conforming to E-learning 2.0 philosophy, we developed a prototype E-learning 2.0 platform called L2. In core principles, it embodies some similarities with Elgg learning landscape [Werdmuller, Tosh, 06] and goes further in terms of (remote) content adaptability, personalization, and more strict system architectural style. It is based on the following principles and technologies: • REST architectural style, as defined by [Fielding, 00]. • Restlet1 API [Jouvel, 06] implementation of REST architectural style for Java. • Adaptive XML Inclusions (AXI) [Pitner, Adámek, 05] enabling to adapt and integrate arbitrary XML resource available on the net. Applications of AXI in e-learning context have been shown in [Pitner, 04]. • Lightweight P2P Personalization [Pitner, 03] allowing to personalize (edit, annotate) arbitrary XML resource, especially of document character. L2 is written in Java and is available as open-source. Let us show its most important features and illustrate how they fulfil the E-learning 2.0 requirements. 4.2
Content creation, storage and delivery
This is probably the most important component in the stack of the e-learning infrastructure. L2 Platform does not force the user to store their favourite content on the platform but rather allows linking to it. The following figure depicts a page that integrates two RSS channels via Adaptive XML Inclusions. L2 is a platform implementing principles of Active Documents, [Heinrich, Maurer, 00]. L2 can access any content stored anywhere on the web, in Wiki, Subversion repositories and other places accessible via a URL.
Figure 1: Accessing and integrating remote content in L2 1
Restlet is a trademark and service mark of Noelios Consulting.
4.3
Blogging
L2 provides a simple blogging service allowing attaching arbitrary resources to each blog entry. If needed, the content of external blogs can be integrated too.
Figure 2: Tagged blog entry in L2 – REST tag
4.4
Glossaries and Searching services
One can find many very good dictionaries, glossaries on the Internet. Trying to provide those services (as some LMS do) on our own is wasting of resources and it is even not possible once you do not force users to store their content in the system. L2 is built on Restlet API allowing creating clients for such services, and integrating them. 4.5
Content syndication
Syndication is an important feature of Web 2.0. RSS, ROME, and Atom are wellestablished standards for content syndication. L2 Platform allows accessing and integrating the RSS feeds via Adaptive XML Inclusions.
Figure 3: Syndication in L2
4.6
Compatibility and reusability
However, the e-learning content creation and delivery should comply with established standards, especially for content metadata, such as AICC or SCORM. So, we should provide proxy services that can transform the required Web 2.0 content into the e-learning standard form. It is not, in principle, difficult, as there are already tools for
creating SCORM-compliant content packages from wikis, blogs or web pages, such as [eXe]. 4.7
Communication, Collaboration and Sharing
There are many web-based chats, Jabber servers, archives, and online discussion forums with archives. L2 Platform does not replace them – you can link to your favourite one and integrate its content wherever you need. The same applies for file-sharing services. Use what you like as you like. As long as your versioning/source code management system is accessible via URL – just link to the resources from L2. L2 offers a tagging service – any resource can be collaboratively tagged and searched. The figure 2 showed also tagging of blog entries (the REST tag is, of course, clickable…). 4.8
Adaptivity and Personalization
The Adaptive XML Includer incorporated in L2 has a RESTful interface; it may serve as a proxy for adapting any markup content (HTML, Docbook, DITA, xdoc, RSS) that can be accessed the REST way via a URL. Other types of content are not a priori excluded from adaptation. Input filters are provided for wiki pages, ordinary web pages, and text files. Adaptation axes include content adaptation for various output devices, like mobile phone or PDA. Adaptive XML Inclusion again helps to format the output into a suitable form, such as WML. Providing personalized, person-centred views of any content is the Holy Grail of many current research activities. But again, many of them concentrate on building either centralized systems for maintaining user profiles, usually requiring quite complicated bootstrap-phase until enough information about the users are collected. They also typically need a rich semantic background in the field of the personalized content. However, one can find also “cheaper” solutions. L2 builds on Lightweight P2P Personalization [Pitner, 03] enabling to personalise virtually any text (XML) resources accessible on the net. You can make annotations, remarks, drop the learning content no more needed.
Figure 4: Annotations in a Docbook document
Clicking on the pen icon makes a new annotation for this document and current user:
Figure 5: Entering a new annotation
5
Conclusions and Future Work
We saw the main characteristics of a desired E-learning 2.0 platform. So, the future LMS must follow these principles: • Must be built on services. • Must be simple, but extensible. • Should integrate other services available, either directly or via proxies. • Provide own services only in specific areas not covered by other Web 2.0 services available. • Respect the REST architectural style of services wherever possible, since it enables seamless integration into the web. We tried to respect these principles in L2 platform. The first tests have clearly shown the viability of the REST approach to build E-learning 2.0-aware e-learning applications. It has shown that the chosen architecture backed by Java Restlet API, Adaptive XML Inclusions and Lightweight P2P Personalization offers a unique, orthogonal, powerful, but still easy-to-use combination of tools. The next goal is to thoroughly test the system and the technology developed. It must be tested in a really large-scale distributed environment, under heavier load, and with various user groups. In longer run, the incorporation of semantic technologies is envisaged as a natural extension to the E-learning 2.0 platform prototype. The current prototype is a generic base where more sophisticated adaptation and personalization can be plugged in. Acknowledgements The research has been supported by the Czech National Program Information Society, project 'E-learning in the Semantic Web Context', No. 1ET208050401.
References [Bittorent] Bittorent System, available online http://www.bittorrent.com/introduction.html [Downes, 06] S. Downes, E-learning 2.0, ACM eLearn Magazine, 2006, available online http://www.elearnmag.org/subpage.cfm?section=articles&article=29-1 [eXe] eXe (E-learning XHTML Editor Project), available online http://exelearning.org [Fielding, 00] R. T. Fielding, Architectural Styles and the Design of Network-based Software Architectures, PhD Dissertation, University of California, Irvine, USA, 2000 [Heinrich, Maurer, 00] H. Maurer, E. Heinrich, Active Documents: Concept, Implementation and Applications, JUCS, Volume 6, Issue 12, 2000. [Louvel, 06] J. Louvel, Restlet API and Noelios Restlet Engine, Noelios Consulting, 2006, available online http://www.restlet.org [Moore, 91] M. G. Moore, Distance Education Theory, DEOSNEWS Vol. 1 No. 25, available online http://www.ed.psu.edu/acsde/deos/deosnews/deosnews1_25.asp [O’Reilly, 05] T. O'Reilly, What Is Web 2.0. Design Patterns and Business Models for the Next Generation of Software, http://www.oreillynet.com/lpt/a/6228 [Pitner, 03] T. Pitner, Lightweight Personalization of XML E-learning Materials, Berliner XML Tage 2003, pp. 361-369 [Pitner, 04] T. Pitner, Adaptive XML Inclusion for Effective Support of Hybrid Learning, in Proc. of I-KNOW 2004 Conference, Know-Center, Graz, 2004 [Pitner, Adámek, 05] T. Pitner, P. Adámek, Improving Document Reusability with Adaptive XML Inclusions, in Proc. of XTech 2005 Conference, Amsterdam, 2005, available online http://www.idealliance.org/proceedings/xtech05/papers/03-05-01/ [Porter, 05] J. Porter, Introduction to Web 2.0, available online http://www.squidoo.com/introtoweb20 [Werdmuller, Tosh, 06] B. Werdmuller, D. Tosh, Elgg learning landscape, available online http://elgg.org
