When cloud computing meets with Semantic Web: A new design for e-portfolio systems in the social media era

The need, use, benefit and potential of e-portfolios have been analysed and discussed by a substantial body of researchers in the education community. However, the development and implementation approaches of e-portfolios to date have faced with various challenges and limitations. This paper presents a new approach of an e-portfolio system design based on Private–Public (PrPl) data index system, which integrates cloud computing applications and storages with Semantic Web architecture, making semantic web-based visualisation and advanced intelligent search possible. It also discusses how the distinctive attributes of the PrPl-based digital asset management system can serve as a large-scale robust e-portfolio system that can address issues with scalability, sustainability, adoptability and interoperability. With such a new distinctive design, a large-scale deployment at a state or national level becomes possible at a very cost-effective manner and also such large-scale deployment with intelligent digital asset management and search features create numerous opportunities in education. Introduction A portfolio is a purposeful collection of student work that exhibits the student’s efforts and achievements (Paulson, Paulson & Mercer, 1991; Wiedmer, 1998) and progressive improvement over time (Barrett, 1994; Challis, 2005) in a life-long journey with academic and professional endeavours. With the advent of the digital age, the emergence of electronic portfolios (e-portfolios) appeared as a logical next step for portfolios. In fact, over half of US higher education institutions are using some form of e-portfolios (Green, 2008), and the popularity of e-portfolio has grown out from the higher education arena to flood into a broader population: Rhode Island implemented a state-wide e-portfolio system for its high school students(RINET, 2009), and Minnesota provides life-long e-portfolios for all residents (eFolio, 2009). Not only serving as a repository of digital artefacts or evidence of mastery and achievement, an e-portfolio also serves as an assessment tool for users and becomes a learning place where a student matures learning through self-reflection (Bhattacharya & Hartnett, 2007). These recognised benefits convinced many institutions of higher education in the US to college systemwide e-portfolio implementation initiatives such as e-portfolio California project funded by a grant from the California Community Colleges System Office (CVC, 2009). Furthermore, the ease and diversity of content creation afforded by social web tools, and the ubiquity of technology use amongst learners today, the range of e-portfolio activities and the British Journal of Educational Technology (2010) doi:10.1111/j.1467-8535.2010.01055.x © 2010 The Authors. Journal compilation © 2010 Becta. Published by Blackwell Publishing, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA. kinds of digital artefacts that could possibly be representative of an individual’s e-portfolio have increased substantially. For example, educators have reported and discussed the use of e-portfolio systems in creating a collaborative learning space (Greenberg, 2004), fostering peer assessment (Stevenson, 2006), student advisement and career planning (Lorenzo & Ittelson, 2005a), content visualisation with electronic concept maps (Kim, 2006), planning and managing instruction (Sherman, 2006), and improving the effectiveness of faculty (O’Brien, 2006). Despite the growing interest and use, there is a series of issues and challenges that austerely hinder the implementation and the wide use of e-portfolio systems firmly integrated in academic and professional activities. Therefore, this paper attempts to identify challenges and issues from current cases of implementations and development approaches with e-portfolio systems; introduces a new radical design approach as a viable solution; and shares potential implications of such design. Current efforts to implement e-portfolio systems Efforts in implementing e-portfolio systems in education have been pursued and well-documented in various European countries, Australia, as well as in the US. For example, the European Institute of ELearning (EIfEL) set up a Europortfolio group with a mission to ‘engage upon an orchestrated effort involving both educational and corporate institutions to define, design, and develop digital portfolio systems that meet the needs of all stakeholders’ (EIfEL, 2007). The Australian e-portfolio Project was awarded to a consortium of four universities—Queensland University of Technology, The University of Melbourne, University of New England and University of Wollongong—to, amongst other goals, ‘provide an overview and analysis of the national and international e-portfolio contexts, document the types of e-portfolios used in Australian higher education,’ and ‘identify any significant issues relating to e-portfolio implementation’ (AeP, 2008). In the US, the premier example is eFolio Minnesota, considered the first state-wide e-portfolio system in the US. eFolio provides every resident in the state of Minnesota a free lifetime e-portfolio with limited storage space (http://www.efoliominnesota.com/). Since it was launched in October 2002, over 90 000 residents and students in Minnesota have used the e-portfolio system. The number of registered users has grown linearly at a rate of approximately 1300 new users each month, and over the year ending in 2005, eFolio received an average of over 2 million hits and 67 000 unique visits per month (Cambridge, 2008). With reference to the functionalities of an e-portfolio system, Greenberg (2004) described the e-portfolio as ‘a network application that provides the author with administrative functions for managing and organising work (files) created with different applications and for controlling who can see the work and who can discuss the work (access)’. From a technical perspective, adopting an e-portfolio system typically involves setting up the e-portfolio application on a server, and providing a large data storage facility to house user files. Students then create accounts, add data files into the system and create different assemblages of the data files targeted towards particular audiences. In terms of the technical implementation of e-portfolios, there is currently great diversity in the technologies being used by various institutions. The Australian ePortfolio Project (2008) reported a range of tools and applications being used by Australian universities. Some universities used extensive applications like full e-portfolio systems, virtual learning environments and adapted learning management systems, whereas others simply employed student web pages and blogs. In the US, Lorenzo and Ittelson (2005b) observed four basic approaches in implementing e-portfolio systems: (1) home-grown, proprietary systems like the DU Portfolio Community system from the University of Denver and the UW Catalyst Portfolio tool from the University of Washington; (2) open source, publicly available systems like the VT Electronic 2 British Journal of Educational Technology © 2010 The Authors. Journal compilation © 2010 Becta. Portfolios pilot from Virginia Tech; (3) commercial, licensed systems like Avenet eFolio, which provides the e-portfolio platform for the Minnesota project; and (4) common hyper text mark-up language editors like Microsoft Frontpage and Adobe Dreamweaver. Existing implementation issues and challenges Despite numerous benefits of e-portfolios recognised throughout the academic community, only a handful number of institutions are reported as having mature campus-wide e-portfolio systems (Paoletti, 2006). In fact, many institution-wide implementation initiatives were stalled or are remaining at pilot stages on many US university campuses (Paoletti, 2006). The impeded acceptance of e-portfolios on many campuses can be explained by a number of barriers observed during e-portfolio implementation sites throughout the US and other parts of the world. Studies observe a range of challenges against a successful e-portfolio implementation at higher education institutions (Canada, 2002; Lorenzo & Ittelson, 2005a, b; Sherry & Bartlett, 2005; Tosh, Light, Fleming & Haywood, 2005; Wetzel & Strudler, 2005). In sum, such challenges include the insufficient technical infrastructure (hardware, software and IT support), lack of skills and knowledge among students and staff, demand on increased time commitment for users (especially on the part of teaching staff), and the problems with security and privacy of data. Especially, the myriad of technologies currently being employed, and the increase in range, file size and diversity of possible digital artefacts have given rise to a number of implementation issues and challenges: Scalability and sustainability Current e-portfolio systems generally provide users storage space in a centralised server for all of their data. This poses immediate problems of scalability and sustainability, especially for institutions with large populations of active users. With the increasing use of multimedia, and the ease of creation of media-rich content, there is a drastic increase in storage load imposed by any user on the e-portfolio system. For example, the 3 Mb storage space provided for each user in the eFolio Minnesota system would hardly be enough for the avid user today. In fact, Cambridge (2008) reported that users in the Minnesota system had given feedback that the ‘small amount of storage space allotted to them by the eFolio software impedes their ability to include substantial personalising multimedia content’. In addition to immediate storage requirements, questions such as storage growth rates and the duration of each user’s membership in the system after graduation from educational institutions have yet to be addressed. Data transportability As users import all their data into the centralised e-portfolio server of an institution, it is essential for users to be able to transport their information as they move from one educational institution to another. Most of the e-portfolio systems in use today are not interoperable. As Jafari (2004) described, ‘we do not yet have all the necessary interoperability requirements defined for various types of e-portfolios, and this is causing ever-increasing challenges for developers of e-portfolio software environments as the projects increase in size and complexity’. This challenge in data transportability has in turn given rise to issues with data ownership. Naturally, any user would like to export the data and to carry it with him when he moves on to other institutions, but who is the real owner of the data in the e-portfolio? Is it the author or the institution? Barriers to user adoption ‘If we build it, will they come?’—this question is a line adapted from the 1989 film, Field of Dreams. Patent (2007) asked this question in the context of an e-portfolio system—highlighting that ‘it is not a foregone conclusion that just because we build an e-portfolio into a course, students will want to use it, or find it beneficial in developing their learning in the ways it was designed to do’. The critical question we need to ask ourselves is whether there are barriers to the use of e-portfolios by learners. Social media ePortfolio 3 © 2010 The Authors. Journal compilation © 2010 Becta. An e-portfolio system is to provide a useful personal space where students can collect the digital artefacts that present evidence of their experiences and achievements. On the other hand, the learners of today, who are more accurately referred to as ‘digital natives’ are adept at using Web 2.0 tools. They already ‘rely heavily on communications technologies to access information and to carry out social and professional interactions’ (Kennedy, Judd, Churchward, Gray & Krause, 2008). ‘Web 2.0 tools tend to be relatively unstructured and are characterised by an ease of publishing, a high level of interaction, self-assigned semantics (tagging) and are often media rich’ (Cotterill, White & Currant, 2007). With learners already using (or fluent in) social and Web 2.0 technologies in interacting with each other and in showcasing their media artefacts, current e-portfolio systems require users to go through an additional process of duplicating the media artefacts which may already be residing in these diverse web applications, and uploading them onto the e-portfolio system. This is potentially time-consuming and laborious, and could pose as a barrier to user adoption. Introducing PrPl As highlighted earlier, with Web 2.0 technologies and applications, a significant amount of user data, or media artefacts, is stored in online storage services in the Internet cloud. Users today are inundated with a myriad of online services that offer data storage in the cloud for free. Such cloud computing services (eg, Google Docs, Facebook, Flickr, MySpace, Picasa, YouTube) allow users to share their data with anyone, anywhere and at anytime. Online blogging sites also offer free hosting services to users, and social networking sites provide free platforms for user interactions over the web. Furthermore, the technology trend is moving towards offering software as a service on the cloud, rather than as a standalone product. Software applications like Adobe Photoshop and Google Docs already operate on data stored in the cloud. Rather than operating an e-portfolio system apart from these online services, cloud computing offers a potential solution to the challenges to e-portfolio implementation—issues of sustainability, scalability, data portability and user adoption. Unfortunately, until recently, in spite of its potential, cloud computing has posed more challenges than solutions. Cloud computing services range from the narrow, like uploading pictures onto Flickr, to broad ones like Facebook, where users have the ability to define applications. The bulk of cloud computing services is currently application-centric, and requires the user to explicitly upload data to different websites. This results in data scattered across diverse devices and websites. Any one form of media can be stored across a variety of different services. With the potentially exponential rise in the amount of data stored across various services online, it is becoming increasingly difficult to keep track of one’s own data. Consequently, this poses problems of inconvenience and potential loss of data ownership and privacy due to reduced capacity to monitor and mange the data (Seong et al, 2009). PrPl (which stands for Private–Public) is a personal-cloud computing infrastructure being developed from an on-going NSF-funded project (POMI, 2009) lead by researchers at Stanford University. Tapping on the wide range of free cloud computing services available to the user, PrPl adopts a user-centric design approach to create a federated storage system out of existing web services and applications. It is a collaborative and semantically indexed data management system which functions as a simple index of data that users may already have stored in other services. PrPl therefore allows users to continue to take advantage of the free storage services—the PrPl infrastructure then serves to present ‘a unified, location-agnostic view of the data in a user’s personal cloud’ (Seong et al, 2009). There are two key concepts in the PrPl infrastructure: 4 British Journal of Educational Technology © 2010 The Authors. Journal compilation © 2010 Becta.