Create a book
Home
From Stellar Deliverable 6.3
1 D6.3 First iteration of STELLAR mash-up
Find the version before the submission of the deliverable here: PreDeliverable
2 DOW
The STELLAR WebSite will be initiated. It will contain a first set of mash-ups of general available web 2.0 and social software tools applied to TEL data sources. These mash-ups will be the basis for further refinement.
The official DOW part on the deliverable is: D6.3 First iteration of STELLAR mash-up. The STELLAR WebSite will be initiated. It will contain a first set of mash-ups of general available web 2.0 and social software tools applied to TEL data sources. These mash-ups will be the basis for further refinement.
The official DOW task description is: The task T6.1: TEL Science 2.0 Mash-up initiation and evolution: Within this task we will set up the mash-up infrastructure and make it available first within STELLAR and in a second step to the wider TEL community. We will develop a variety of mash-ups of mainstream tools and more specialized research oriented tools from the partners. These mash-ups will be available on the STELLAR web site (e.g. in the form of widgets to be easily embedded into web-applications). The work to be undertaken will involve the following steps for each of the yearly cycles: requirements gathering, design of mash-up, implementation of mash-up, evaluation of mash-up.
3 Latest activities
Stellar directory of services, widgets and tools Stellar Directory.
Publication feeds format introduction Publication feeds format introduction
Publication feeds format v1.0 Publication feeds format v1.0
Burst format discussion BuRST format adaption discussion
Roadmap Science 2.0 for TEL Roadmap Science 2.0 for TEL
WP6/WP5 integration scenarios for TELeurope widgets for TELeurope
Widget Cookbook Widget Cookbook
In the last few months, there is a strong trend to establish a simple API for retrieving research related data. Here is the work in progress development.
Draft for D6.5: http://www.stellarnet.eu/d/6/3/Draft_Outline_for_D6.5
Here you can find a Directory of BuRST feeds
4 Deliverable 6.3
Note: Automatically converted from word to wikitext with wikEd
Executive summary
This deliverable is D6.3, the ‘First iteration of STELLAR mash-ups’. It contains a vision, a collection of requirements, an architectural framework, and a first set of implemented mash-ups of general available web 2.0 and social software tools applied to TEL and TEL data sources within this framework. These initial mash-ups will be the basis for further refinement.
The work on science 2.0 mash-ups relates to our overall STELLAR project objectives in the following way. First of all, it serves as an infrastructure for sharing, discussing, and interacting thereby paving the way for more interdisciplinarity. Second, it serves the purpose of gaining more visibility for the work accomplished within the network. Additionally, it facilitates easy reporting and documentation of the activities pursued by the partners within and beyond the consortium.
This is strongly related to the work on evaluation (WP7) and project management (WP8). It also interfaces with our work on community outreach in our TELeurope.eu stakeholder platform (WP5): strategically, the infrastructure framework has been designed to allow for close collaboration. It services the work packages 1,2,3, and 4 with support in tools and practice.
It serves primarily the task T6.1 ‘TEL Science 2.0 Mash-up initiation and evolution’ in setting up an infrastructure and initial applications of this infrastructure and making them available first within STELLAR and in a second step to the wider TEL community. It is a first attempt to develop a variety of mash-ups of mainstream tools and more specialized research oriented tools from the partners. Some of these mash-ups are already publicly available via the STELLAR web site, others are soon to follow, additional ones in development.
The deliverable is structured into the following parts. First, a vision is sketched, summarising the current debate about Science 2.0 and mash-ups to support it. This vision is informed by the ‘Science 2.0 for TEL’ and ‘Mash-Up PLE’ workshops we have held at the 4th European Conference on Technology-Enhanced Learning. Within this chapter, requirements are elaborated using scenarios and use-cases. Second, an architectural framework is elaborated within which the mash-up developments move. A process accompanies this framework to allow for maturing those ideas and concepts into end-user ready widgets that have proven valuable in exploratory research. Third, a collection of initial building blocks for STELLAR’s Science 2.0 mash-ups is listed along two lines: infrastructure components (data sources and services) and use-case sized web applications – what we call Science 2.0 widgets. Finally, a conclusion rounds up the picture.
5 1 A vision of Science 2.0 mash-ups
The notion ‘Science 2.0’ is a rather young one. Not even old enough to have one single, established definition. Not to speak of an accepted set of schools of thought. Though, this new concept already has several parents trying to shape its meaning.
Waldrop (2008), for example, relates Science 2.0 to “new practices of scientists who post raw experimental results, nascent theories, claims of discovery and draft papers on the web for others to see and comment on”.
Shneiderman (2008) adds another aspect and sees in Science 2.0 also “new technologies [that] continue to reorder whole disciplines”, as “increased collaboration [is stimulated] through these socio-technical systems”.
Kieslinger & Lindstaedt (2009) stress that it focuses on “improving, enhancing, speeding up of feedback cycles” and request to start with defining “practices instead of technologies”.
Underwood et al. (2009) outline that Science 2.0 offers more than just efficiency optimisation (through improved workflows and better sharing possibilities): participation can be broadened beyond existing scientific communities. They envision a “technology enhanced participatory science”.
Gillet et al. (2009) see in Science 2.0 a concept that federates a variety of communication channels to ease internal communication within an existing scientific network and beyond.
Butler (2005) sees a key feature in this emerging new science web to be in the “dynamic interactions between users in real time”, while criticizing at the same time that the uptake of these new technologies in science is ironically slow.
Waldrop (2008) goes even beyond this and claims that ‘Science 2.0’ allows for a richer dialogue, such as collaborative brainstorming, meta conversations, or an open discourse of “critiquing, suggesting, sharing of ideas and data” among previously unknown parties. That way, Science 2.0 is supposed to allow for efficient crowdsourcing of ideas and the refinement of knowledge through an open debate.
Shneiderman (2008) asserts Science 2.0 even a paradigm change in methodology and focus: research takes place embedded in the real world through rigorous and large-scale observation with validity being proven empirically with the help of quantitative and qualitative analysis.
Although looking at this new concept from different angles, there seems to be an agreement amongst scholars to claim that ‘Science 2.0’ is more collaborative and collegial, more productive and efficient, focusing on offspring rather than offering, fostering engagement in both cooperation and assistance. And that Science 2.0 is about applying modern web technology (Web 2.0) to support and conduct research.
We have held two open workshops to further elaborate the Science 2.0 and mash-up concept for our field in the context of the European Conference on Technology-Enhanced Learning (ECTEL’09) in Nice, France. Annex D contains the call for papers and the programme overview for the Science 2.0 for TEL workshop, which had attracted five full papers. Annex E provides the call for papers and programme overview of the 2nd workshop on mash-up personal learning environments (MUPPLE’09), which had attracted 20 submissions and accepted nine full papers.
The subsequent sections elaborate in more detail, which Science 2.0 practices should be focus of our STELLAR Science 2.0 mash-ups. Furthermore, functional and non-functional requirements for an infrastructure are elaborated. Last but not least, the mash-up concept and approach is defined.
As depicted in Figure 1, the requirements elaborated subsequently in this chapter afford a distributed infrastructure that allows for the integration of legacy systems in form of a service-oriented architecture and with the help of widgets in order to create the desired science 2.0 mash-ups. This infrastructure will be further explained in Chapter 2.
Figure 1. Relation of use cases to the infrastructure framework
consisting of services and components and widgets.
6 1.1 Scenarios and use-cases
Not unlike Vannevar Bush’s scenario of Memex that served as an thought-provoking vision in hypertext research, the following scenarios and use-cases are intended to stipulate debate about Science 2.0 and are meant to inform both research and development – maybe just a tiny bit less ambitious. All of the use-cases have been evaluated to be realistic to reach within the lifetime of STELLAR.
We have elaborated the collection of scenarios and use-cases from two different perspectives: The viewpoint of a researcher in our field of technology-enhanced learning is complemented by use-cases taking the perspective of the STELLAR network. Both of them share that they are intended to support and maybe even re-invent research with the help of modern web technology.
The STELLAR perspective thereby relates to cases with a more overarching view, which may even be regardless of individuals.
An overview on the collected use-cases is given in Figure 2. As can be seen, that the use-cases touch upon the areas awareness, collaboration, reflection, and argumentation.overview-use-cases.pdf
Figure 2. Overview on the use-cases.
It should not stay unmentioned that many of the use-cases might also be applied in other research fields, as the scientific method is the common practice underlying all academic disciplines. It is, however, not our aim to serve all disciplines – and of course we humbly ask to excuse these eventual corollaries.
6.1 Use-case: publication alerts
The user would like to get an overview of papers in the field of TEL and be notified of new papers. He or she opens the ‘Publication Feeds’ tab on the widget platform and adds various standardized publication feeds from conferences/institutions he or she knows. The bibliographic data is loaded into the widget and presented in a well-arranged and visually attractive manner. The user is able to rate each publication and comment on it, which is used to adjust the interest profile of the user. Whenever there are new papers published, the widget is updated and the feed emitted by the widget reports a new paper.
6.2 Use-case: spotting publication trends
Researchers are usually very interested in current developments and new trends in their area of work. It is, however, often not very easy to keep up to date, and to find out about new trends without the investment of a lot of time. To shorten this time-consuming job, researchers have started to develop some automatic trend-crawlers. One example for such a trend tool is Google Trends[#_ftn1 [1]], which enables insights into the popularity of topics (qua the number of search requests executed for this topic in Google). Obviously, as research and research trends usually take place much earlier than their public perception, using log file analysis of generic-purpose search engines is not such a good idea (e.g. look in Google trends for ‘RDF’ or ‘javascript’, both becoming less popular over time although the technologies are more mature now and have more widespread uptake compared to previous years).
A researcher would like to be informed about trending topics in the field of TEL. She opens the trend widget and types in the keywords ‘reflection’. She is presented with a faceted browsing view. A tag cloud (or similar) aggregated from the keywords is shown to the user. She then restricts the data to certain years to see the changes in the tag cloud. To complete the picture a timeline with the publication dates of the publications is shown to the her.
As a result she gets a chart based on her publications covering all her publications dealing with reflection in TEL, additionally all added comments and rates. In a second chart she finds all further publications dealing with reflection from other people, with the corresponding comments and rates. Additionally she receives a tag-cloud, presenting all hot-topics concerning reflection. With these two presentations she finds out, who is also working on her area of interest as well as the trends in which her area of interest is developing.
Figure 3. Mock-up of a STELLAR Trend Widget.
6.3 Use-case: meeting recommenders
When a TEL researcher visits a conference, he wants to meet new interesting people. To get an idea who he could talk with on the conference that could be doing similar research as his own, he opens the conference author visualization and searches for his own name. He will be presented with a visualization showing authors with whom he shares references in his papers, so the topic of his research could be related to the research of these authors. Another option is to combine social networking information from the conference data with meeting recommendations for subsequent virtual follow-ups (using flashmeeting). Trained and tuned such a recommender system could help to spot unwanted fragmentation. In a similar manner, flashmeeting groups can be proposed and virtual attendance recommendations can be made.
6.4 Use-case: (conference) social network
A researcher attending a TEL conference likes to know more about the conference. He would like to know who else published in this conference that he might know and what kind of research is typical for this conference. To answer his questions he opens the conference browser and searches for his name or the name of researchers he knows. Information on the wanted author is displayed in an information panel, his geo-location is shown on the map and all his papers are shown on a timeline. He can browse through this data by clicking on his co-authors, other papers he wrote, etc. This will change the view around the selected piece of information. Another possibility is that a TEL researcher wants to understand more about the interrelations between countries publishing at a conference and about rankings of top authors and countries. He can open the conference visualization tool and explore the top publishing authors and top cited papers. He can as well get a world map and see which countries published most and together with which other countries. Similarly, the open archive can be used to allow for collaborative filtering in order to make recommendations about related readings.
6.5 Use-case: profiling the speaker
A researcher attending a presentation likes to know more about the speaker. He would like to check the paper that is presented, which other publications are authored by the presenter, or learn about other digital footprint such a blog or tweets left behind by the speaker. To get the desired information he uses his smart phone to scan the QR code (or Google goggles and face recognition) included in the speaker's presentation. A web page will load displaying the links to the slides and paper of the current presentation. Also, he can get links to the different social tools used by the speaker and a list of his previous publications. The researcher will also get the possibility through this application to contact the speaker or to recommend his work to colleagues.
Figure 4. Screenshot of the experimental ‘More!’ mobile application.
6.6 Use-case: follow people
Research is also about tracking other scientists working in the same area in order to, for example, to avoid redundancy. Researchers are supposed to know the community around their research area. However, communities are dynamic and change over time. Furthermore, different communities often work on similar topics. Therefore, it is essential for researchers to track scientists who work on similar topics but are not members of their own direct social circles.
A researcher wants to follow people who are working in the same field than she does. Therefore she uses the people tracker widget. This widget helps her to detect people working on similar topics than she does, but are not part of her own professional circles. The people tracker widget can be seen as a window to the world outside of the community. The widget tries to order the list of users according to their predicted relevance for a specific topic. The researcher can configure the topics she is interested. Furthermore, she can see for each user why the system provided the recommendation (e.g. because the user has published several microblog messages, which have been re-published or referenced by others for this topic).
Figure 5. Mock-Up of a follow people widget.
6.7 Use-case: reflection support
Learning from past experiences is an important research topic in TEL, being of concern not only to companies and organizations, but also to researchers themselves. To reflect over past activities is an important step in the whole research process. Therefore, the task of the Reflection Widget is to provide a researcher with all information available concerning her own research development. On the one hand, the presented data consists of what the researcher has published so far, where she has added comments or ratings to publications from other researchers. On the other hand, the data shows how she comes to decisions based on her own research and on her assumptions used for these decisions. The goal of the widget is to empower and to motivate the researcher by reflecting on her work practices and personal experiences in her past research. The Reflection Widget gives the researcher the possibility to reflect on her whole research development.
A researcher wants to reflect over her past research activities. She opens the Reflection Widget and finds all her publications listed in the reflection window based on a timeline. The visualization shows all of her publications, with added comments, ratings, or even stated goals (cf. Glahn, Specht, & Koper, 2009). If she has contributed some personal ideas or comments, this information will be displayed as well. Additionally all publications, where the researcher herself has added comments, opinions or ideas, will be presented. Thus, the researcher can reflect over her complete research-development and previously made comments and thoughts. If she wants to focus the presented data to a specific topic and she enters the keyword. The Reflection Widget presents all information according the given topic. This empowers her to reflect on a specific topic and empowers her to get and insert new ideas and thoughts.
Figure 6. Mock-up of a reflection widget.
6.8 Use-Case: streaming feedback
Researchers are usually very interested in getting feedback about their work. Some academic feedback methods, however, take a long time: with peer reviews of papers, researchers often have to wait for weeks if not months until they receive feedback. New forms of gathering feedback have popped up just recently with the rise of Web 2.0 technologies such as blogs and microblogs – but to the cost of centrality. To gather the feedback from distributed blog comments, tweets, and the like, a ‘personal window’ widget can be introduced that helps with collecting and presenting the distributed feedback about the researcher’s work. The presented feedback consists of different types: comments and ratings directly to publications from the feeds, links referring to publications used in blogs, twitter, etc. and uptake in form of citations.
A researcher would like to see feedback of her work. Therefore she uses the ‘personal window’ widget of the Stellar platform. This widget presents her the most current feedback stream (in chronological order) and several statistics about the received feedback. The feedback stream is a real-time feedback stream and is updates whenever new feedback is published. Furthermore, the researcher can filter the feedback by publications (i.e., showing all feedback for selected publications), by platforms (i.e., showing all feedback originating from selected platforms) and by user groups (i.e., showing all feedback generated by a certain group of users, e.g. Stellar community members).
Figure 7. Mock-up of a ‘personal window’ widget.
The feedback monitor widgets allow the researcher to quickly see feedback statistic over the last months.
Figure 8. Mock-up of a feedback monitoring widget.
6.9 Use-case: paper debating
Engaging in a one to one email exchange with another scientist can often lead to a point where it would make sense to open up the debate to the public, as there is a critical point on which it may be worth to enlarge the discussion and not to keep it private. This exchange can be carried into the open by posting a short position statement on the blog attached to the publication or in the professional blog of the researcher. It happens that other researchers interested by this paper or following this blog subscribed to its attached RSS feed, so they come to see the comment and engage in the conversation. Naturally this leads to opening up a forum. A small community of interest emerges, the forum is used for a dynamic discussion, but when a statement is mature enough it is on the blog that it is kept for a deeper argumentation. It seems that students are especially interested by these blogs attached to papers, which sometimes are difficult to understand without comments.
From these discussions it appears that there are different nuances in the understanding of the issues at stake. For example, computer scientist and learning scientist don't have the same meaning for the same words (e.g. ontology), or different words for possibly the same meaning (e.g. bog, gap). Second, the same phenomena happen between languages (e.g. connaissance vs. knowledge, limitacion vs. contrainte). Along the discussions these differences are documented in blogs and fora and can be harvested centrally to automatically build a multilingual, multi-disciplinary dictionary by tagging the paragraph containing definitions.
6.10 Use-case: collaborative writing
One of the big advantages of the penetration of Internet into almost all sorts of research activities are the increased collaboration opportunities. Of course this has resulted into a multitude of tools such as Wikis and other forms of online text processing. New forms (such as google wave) pop up and new protocols such as operational transformations help to create new types of collaborative applications.
A researchers needs support for collaboratively writing a paper together with a distributed team of scientists. He agrees on the structure during a video conference with his colleagues and assigns responsibilities. Reference meta-data is collected along the way of the search in scholarly databases and the joint wiki provides support to integrate references. Equations are used to outline the formalisms of the contribution and the wiki supports this. Before submission, the wiki is extracted into the template of the journal and the resulting word file is polished by the responsible editor.
6.11 Use-case: managing contact points
Scientific discourse takes place face to face and using digital means. To support scientists in optimising their travel plans and in negotiating upcoming meeting places such as conferences and workshops, a mobility tracker aggregates travel plans and allows for intelligent analysis of who would be available at which contact point.
6.12 Use-case: cultivating an open archive
In the STELLAR network, most institutions and also the individual scientists keep track of their publications. An open archive[#_ftn2 [2]] was proposed and implemented to collect all TEL-related resources in one place. To avoid the tedious process of manual updating the archive, every institution produces a publication feed instead. The open archive (and any other interested entity) subscribes to this feed and thus automates the process of filling and updating the archive.
A researcher has just started a new project on ‘YLearning’, the key ideas are motivation of learner and authenticity of learning environments. This is indeed related to his on-going work on personalisation of learning environments: he thinks that personalisation and motivation are tightly related, as motivation and authenticity are. Using Google Scholar (limited it to the last five years), an overwhelming big number of references could be obtained that made it impossible to make relevant use of them (9.650 for personalisation+motivation+TEL, and 15.200 for motivation+authenticity+TEL). Actually, adding TEL does not discriminate much. Moreover, the source on which Scholar makes its search is so general, ill defined and badly standardized, that it is very unlikely that it can help at this stage. The nature and quality of the resource you are exploring is as important as the preciseness of your request.
Alternatively, an open, domain-specific repository such can be used to relate with the state of the art. The same enquiry on the open archive is a bit disappointing. Among the 1.362 references available there, I get a few with each keywords (personalisation 8, motivation 27, authenticity 1 / authentic 12), and none with their combination. Actually, searching by keywords may not be the best approach since the knowledge about keywords evolves with our knowledge and the way we view problems. So instead, the researcher writes a short description of his ideas and uses the search-for-similar ideas-tool of the open archive, which can identify that some resources may address issues close to this particular research interest. Actually, 15 papers can be obtained for which abstracts can be requested. The researcher reads them and identifies a couple of papers of interest. After reading the full-texts, Google Scholar helps (via searching for the names of the authors) to discover another set of references.
6.13 Use-Case: follow news
Keeping up to date in even a specific research area can be a quite time consuming job. One strategy to cope with change is to browse through different web sources such as blogs, news sites, or even microblogs in order to spot the most relevant news items. Information filtering can help here: a researcher wants to see every day (or every week or month) the most relevant news in the field of TEL. Therefore, she uses the TEL News widget on Stellar platform. This widget tracks conversations on the Social Web and ranks them according to their relevance for the TEL community. The TEL relevance level of a news item increases if important TEL users have created or referenced the item. This means, that the TEL News site is automatically generated by analyzing the Social Web activities of stellar community members. The TEL news widget displays the five most relevant news items. The researcher can click on the evidence button next to each news item to display why the system considers the items as TEL relevant.
Figure 9. Mock-up of a TEL news widget.
6.14 Use-case: track discussion
The so-called scientific discourse can often be spread out across various papers and articles in a multitude of media. Keeping track of this discourse and the accompanying discussions, standpoints, and schools, is not often easy, but can be further supported using modern web technology. A feed aggregator allows the interested researcher to analyse artefacts of this scientific debate over time, focus on reply structures and argumentation.
6.15 Use-case: debate live
It is not only environmentally friendly to abstain from face to face meetings, it is also often more efficient (not necessarily more effective). A STELLAR flashmeeting system allows researchers to organise auto-moderated videoconferences with support for voting, minute taking, file exchange, white boards, concept mapping, and many others. When researchers need to meet, they negotiate their preferred meeting times, book a virtual meeting room, conduct and subsequently review the meeting.
7 1.2 A STELLAR science 2.0 mash-up infrastructure
The STELLAR mash-up infrastructure sets the foundations for Science 2.0 portals, which serve as a communication, collaboration, and information channels for the wider TEL community.
The mash-up infrastructure provides the technological basis to facilitate the TEL community to stay informed, exchange ideas, and to (net)work with each other. It uses Web 2.0 approaches to create an environment of dialogue, participation, and engagement (see (Andersen, 2007)).
The STELLAR science 2.0 infrastructure has to be seen within the diversity of the organically grown infrastructures of each partner. Every partner developed it independently from other partners’ networks. This has the benefit for each partner to be able to tailor it to institutional needs. This is especially important for TEL, a young area that grows fast and therefore needs a highly flexible infrastructure.
The STELLAR mash-up infrastructure is a platform of loosely-coupled distributed applications. It combines the strength of the decentralized infrastructure of the partners and provides mechanisms to use its application in parallel, to combine it to new applications and to tailor the applications according to user needs.
7.1 Characteristics of a science 2.0 mash-up infrastructure
Before we describe the technological components of this infrastructure, we picture the philosophy of the TEL Science 2.0 landscape to motivate our design decisions.
Broadly spoken, we see as the pillows of research the generation, communication, collaboration, documentation and retrieval/discovery of ideas. Especially so-called Web 2.0 applications (O'Reilly, 2005) bear potential to support these tasks. They are building on top of the web stack for the purpose of information publishing, referencing and sharing. With the development of the web and web browsers, new ways of interaction with the web became possible, which are now subsumed under the umbrella term Web 2.0. With the evolution of the web to the web of applications (Raman, 2009), web technology is used with a new philosophy in mind. The web changed from content provision to functionality provision in form of rich Internet applications.
In general, these web applications provide users a rich user experience similar to the one on computers. Users are able to use these tools without the need to understand the underlying technology. This enables more users to participate online. Academic blogs allow easily producing web content. The comment function allows others to reply instantaneously reducing the time of a feedback circle. Generally Web 2.0 applications follow the notation of an architecture of participation. This means that they try to reduce technical and access barriers and give multiple options for communication (dialogue and not monologue). Networking is supported with social platforms giving the possibility to have an up-to-date profile, to share information and to explore the social graph to find researchers with similar interests, helping building a community of researchers. Online document applications help to write distributed on research papers (like Wikipedia or google docs). These documents are seen as perpetual beta (living documents with dynamic content), under constant development updated when new information is available.
7.2 Requirements of a science 2.0 mash-up infrastructure
The TEL Science infrastructure follows the principles that make the Web 2.0 successful and tailor it to the specifics of TEL. The following dimensions are seen as guidelines or high-level requirements for the creation of the TEL Science 2.0 infrastructure. We will outline these considering the developer and user point of view.
Support a distributed infrastructure
One of the key ideas of the web is that everyone who wants to participate just needs to follow the basic client-server architecture to become a producer or consumer on the web. We see in this decentralized infrastructure at least four benefits:
Knowledge management: Infrastructures represent a large body of diverse and quite often complex technologies, which needs a highly specialized knowledge management process to maintain and extend. With a distributed infrastructure, each partner is able to take care about its internal knowledge management process helping new developers to make use of the technology.
Using existing infrastructures: Each partner has a historically organically grown infrastructure serving the specific needs of each institute. Compared to a common centralized infrastructure the institutes are thereby flexible to modify their own infrastructure to new situations. The goal of the TEL Science infrastructure is to use this strength and find ways to allow corporation between this heterogeneous infrastructures.
Ownership: The ownership of the tool belongs to the institute making success visible and responsibilities transparent.
Fault tolerance: In cases of system faults of parts of the infrastructure, only the affected parts will not be longer available. The other services will be unaffected.
While the infrastructure will be distributed, the user experience will abstract from this heterogeneity of technologies and will provide a seamless integration of the services and tools.
Be loosely coupled
A system is loose coupled from another, if the one does not rely on the internal working of the other. To exchange messages between these systems interoperability aspects, like the format of the message and the definition of interfaces are crucial. Because of its importance we outline dimensions of interoperability in an own chapter.
Openness
With openness, we understand that the developers of the services and widgets follow open and not proprietary standards, in terms of the webAPIs of the services, the service-widget-communication protocol, and the packaging of the widgets. This will lower the barriers to participate in and to learn about this new TEL Science infrastructure. And this allows to interface with legacy systems without the need of a revolution.
From a user’s point of view, openness also involves the transparency in the process of result generation. It will help users to decide to what extend they can trust a source, if they are able to reconstruct of the method of extraction and filtering.
Usability
Today’s web user interfaces are rarely distinct from desktop applications. Amongst others, this is due to Javascript toolkits like Dojo[#_ftn3 [3]], Scriptaculous[#_ftn4 [4]], Prototype[#_ftn5 [5]], and YUI[#_ftn6 [6]], which ease the development of rich internet applications. Carefully designed, these technologies provide enhanced usability, participation, and lower user barriers.
Allow for impact monitoring
To detect trends the usage of the tools will be monitored according to the guidelines of D6.1. The evaluation metrics will be aggregated and are openly available for the developers and the users to give them feedback about the usage of the tools. The evaluation process will be scheduled repeatedly. With every circle, the evaluation method will be tailored according to new results. This will help to tailor the development of the tools to the needs of the users.
Allow for service and widget discovery
Considering the distributed infrastructure a central point of information about existing widgets and services seems helpful to discover the independently developed solutions. Therefore, all tool and services for the TEL Science Infrastructure will be listed in a directory. Each directory listening will contain a description of the widget or service, how to instantiate a widget in a user platform, compliance with a user platform, their usage in combination with other tools and widgets (if applicable), a link to a showcase (if applicable), to the APIs (if applicable) or documentation (if applicable). Users will be able to comment and view statistics about the widget and services. The directory is the cornerstone of the infrastructure. It shows the provided widgets and servers, and their usage, which will help for further mashing of data services and widgets.
Facilitate identity management
At least two aspects are to consider. To allow users seamlessly to change between services of the TEL Science Infrastructure the login mechanism must be flexible enough to give access to the different systems with the need to login only once. In addition, a unique Stellar look-and-feel of all Stellar products can serve as identity point with Stellar. The presentation of the widgets should thus be designed according to the Stellar corporate design. A common Cascading Stylesheet for widgets, which are build on top of HTML can be shared amongst the developers.
8 1.3 Mash-Ups
With the raise of available APIs and structured data feeds (like RSS, Atom, and RDF) the web opens up for programmers to use and remix this data to new compositions. Mash-ups are usually seen as software applications that merge separate APIs or data sources (Zang, Rosson, & Nasser, 2008). They are the manifestation of the programmable web (Maximilien, Ranabahu, & Gomadam, 2008). They compose new applications based on existing data services and user interfaces. Mash-ups are “a combination of pre-existing, integrated units of technology, glued together to achieve new functionality, as opposed to creating that functionality from scratch” (Hartmann, Doorley, & Klemmer, 2006).
The goal of mash-up tools is to lower user barriers to remix content from heterogeneous data sources and provide user interfaces, which adequately visualize this mix without or near zero programming. Users can choose from a predefined set of available tools or widgets, which could be placed ideally with simply drag-and-drop on a workspace. Users should be able to aggregate, project and to build the cross-product (Raman, 2009) of the tools’ functionalities in a visual way. This should help users to manage information and gain new insights into data without the burden of being a skilled programmer. They can focus on their research.
Mash-up environments ease as well the development work. Through their provided mechanisms for authentication, user and group management, session and asset management, etc., developers can focus only on the functionality of the tool (e.g. data services, visualization, recommendation, communication, collaboration). The tools are usually from small development size, developed by one person. This allows distributed development of tools all over the STELLAR network and assigning authorship and responsibility to the developers. If these tools follow certain programming standards, they could be used in the same mash-up environment although they were designed separately.
Figure 10. Mash-up architecture and involed standards.
9 2 Architectural framework
The outlined characteristics and requirements and especially the idea of mash-ups form the base for the Stellar Science 2.0 mash-up architecture. The developers of the data services and widgets publicize them in the directory of widgets and services. They will provide through their own infrastructure the possibility for users to access these resources. Users will be able to look up in this directory interesting applications and will find the information needed to instantiate them in a range of supported mash-up environments. I could be also possible, that the mash-up environment can look up available tools and services through a REST call and display them in the mash-up environment so that the users can instantiate them per drag and drop. Three types of instantiation are possible:
- The directory could serve as widget engine, which will instantiate a widget with only a graphical user interface.
- The widget also provide a data service. In this case the directory can serve as widget engine and data are provided via a proxify() method to overcome the same-origin restriction of AJAX requests.
- The infrastructure of the institutions can take care about the proper instantiation of widgets and their data flow. In this case, the directory can serve as tutorial of how to implement the application.
Configurations of sets of tools could be provided as well. This will help inexperienced users to start quickly with a set of suitable tools, while they can modify the set of tools later.
Figure 11. Stellar Science Architectural Infrastructure
For our first version of the STELLAR science 2.0 mash-ups, we considered Elgg as showcasing platform for bringing together the developed widgets and services originating within Elgg and in the legacy systems of the partners. In Elgg, widgets and services can be delivered through the Wookie widget engine. We investigated this combination regarding interoperability dimensions. These form the baseline for further evaluation of applications.
10 2.1 Dimensions of interoperability
Interoperability standards are especially important in a context, where several widget and service providers operate in a loosely coupled distributed infrastructure. Interoperability “is a property that emerges, when distinctive information systems (subsystems) cooperatively exchange data in such way that they facilitate the successful accomplishment of an overarching task” (Wild & Sobernig, 2006). Interoperability has several dimensions (Palmér, Sire, Bogdanov, Gillet, & Wild, 2009), which are important for the success of the STELLAR science 2.0 mash-up infrastructure.
- Screen: Spatial organization of several widgets within a mash-up. Standards are: W3C Widget 1.0, Google Gadget API, OpenAjax Metadata 1.0., Open Alliance (Dojo), EMML.
- Data: Interoperability of data and metadata across widgets and underlying services. Standards are: RSS, RDF, Atom, JSON, Dublin Core, HTML5 D’n’D, OpenAjax Hub 2.0, Google gadget pubsub, SCORM, MediaRSS, webSlice Format.
- Temporal: Live updates of widget configuration, state or data. Standards are: XHR, COMET/Reverse Ajax, XMPP, webHooks, Salmon Protocol, PubSubHubBub.
- Social: Interoperability of user identity, profile information and list of friends. Standards are: OpenID, OAuth, OpenSocial API, FOAF, MashSSL.
- Activity: Degree of automatic orchestration of components. Standards are: WS-BPEL, IMS Learning Design Specification.
- Runtime: exchange one rendering and execution platform or its parts with another. Standards are: W3C Widget 1.0: Packaging and Configuration, OPML, Open Ajax Mash-up Reference Application.
Each of the dimensions bears four subcategories, which allow characterising platforms. A platform for example, which supports the runtime dimension, allows users to switch between other platforms, or a platform with distinct categories on screen, social and temporal prepares collaboration.
Palmér et al. (2009) investigated six different platforms (IGoogle[#_ftn7 [7]], Netvibes[#_ftn8 [8]], Moodle[#_ftn9 [9]] & Wookie[#_ftn10 [10]], Google Wave[#_ftn11 [11]], Afrous[#_ftn12 [12]] and G.ho.st[#_ftn13 [13]]). Their results show that none of the existing platforms satisfy all dimensions. Some of the platforms focus more on the collaboration aspect, while others more on personal customization. Both aspects would be important to have converged in one platform.
We will use these dimensions to outline the feature richness of Elgg, which was chosen by WP5 for the STELLAR stakeholder club[#_ftn14 [14]], together with Wookie to evaluate the interoperability of the underlying technologies of the STELLAR science 2.0 mash-up infrastructure. It will serve as an example platform to give interested parties insights into our TEL science 2.0 vision.
11 2.2 Choice of the showcasing platform
Elgg[#_ftn15 [15]] is a social network framework, which is extendable to a custom social networking platform. It provides blogging, file sharing, networking, new aggregation, activity streams, and it has a flexible plugin infrastructure.
We will use Elgg as widget container for Wookie widgets to mash together TEL specific data and TEL specific mini-applications.
The following table summarizes the Elgg and Wookie interoperability dimensions based on (Palmér et al., 2009):
| Dimension | Features | |||
| Screen | Shared screen: Integration of Wookie widgets with iFrame. On Server side integration with plugin. | Widget standards: Wookie widgets are W3C widget standard compliable. | Layout of widgets: The Elgg dashboard has three columns for the ordering of the widgets. Widgets could be added and removed | web desktop: Elgg does not mimic an Operation System. |
| Data | Inter-widget communication: Only between instances of the same widget. | Drag and Drop: Widgets are drag and drop able on the Elgg Dashboard. | Data manager: Elgg has a file repositiory, which could handle a wide range of data. | Linked data support: No |
| Temporal | Push data updates: Instances of the same widget can share data. | Push preference updates: Wookie allows to set preferences | Real time data updates: No | Data and preference history: No version history is kept |
| Social | List of friends: In Elgg you can specify with what person, you are friend or colleague | Friends server: No export or import of friends from other parties | Access control: Elgg has an access control mechanism. Wookie can specify, if a single user or a group of user can share information (using the shared data key). Authentication with OpenID in Elgg possible. | Independent groups: Elgg allows formation of groups |
| Activity | Manual guide: No set of instructions, which activity should be performed. | Flow enabled widgets: With Wookie you can specify the lifecycle of a widget with widget.stop, widget.resume. | Scripted inter-widget data flow: No | Recommendations: No |
| Runtime | Feed export and import: Elgg can generate several feed formats | Generic export and import: No: import possibilities for whole PLE configurations | External configuration: PLE configuration could not be stored in a general form in a separate service | Embedding: The entire Elgg can not be embedded into other PLEs, but Wookie widgets are embeddable in other PLEs like moodle. |
Table 1. Interoperability compliance of Elgg with Wookie.
The following spider chart assesses the Elgg/Wookie interoperability dimensions.
Figure 12. Elgg and Wookie and their features in the interoperability framework.
Elgg and Wookie have high interoperability in the social and space dimension together with a lower extent of the data dimension. These are the technical foundation to enable networking and the enrichment of the platform with a wide variety of widgets. However, enhancements especially in the activity, temporal, and runtime dimension would allow for more flexible mash-up environments with more powerful widgets.
Nevertheless, the combination of Elgg and Wookie seems a good starting point to test our building blocks. It will help to gain experience to refine and to evaluate other solutions.
Through the collaboration with WP5, this choice helps us to disseminate the science 2.0 mash-ups developed within STELLAR to our stakeholders.
Text Box: Widget IIIText Box: Widget IText Box: Widget II540px
Figure 13. The showcasing platform Elgg and Wookie
in interplay with partners’ legacy systems.
12 2.3 Development Process
This deliverable reports on an ongoing activity: it is the first iteration of the mash-ups (their ‘initiation’) which will be developed within the lifetime of STELLAR. Subsequent deliverables (e.g. d6.4, d6.5, d6.6, d6.7) will report on updates of this process of constant further development along the runtime of the project.
It should not go without mentioning that the underlying development process of the mash-ups aimed for within STELLAR is iterative by nature and involves phases of requirements gathering, design, implementation, and evaluation.
Therefore, we have agreed upon a maturing process that helps to develop promising research prototypes (providing and using the infrastructure) into market-ready applications. To keep software development agile and to allow for innovation, to keep iteration cycles small and close to the end-user, the semi-public release of ‘early betas’ for the interested seems to us like a strategy inline with the broader Web 2.0 philosophy – which helps us to sort out those exploratory prototypes fit enough to survive into end-user ready services.
This strategy ensures innovation while it does at the same time not interfere with the development cycles of the already more mature software products (such as, for example, the open archive). And it at the same time opens the development up for other partners to join.
13 3 Building blocks
As outlined above, work on the Stellar Science 2.0 support environments has been organised in two lines: infrastructure components form fundamental building blocks that service end-user directed (mini) web applications. The latter also called widgets. Each of these two types of building blocks shall be summarised in the subsequent passages.
Annex B reports upon them in more detail following a common description format. The structure of the description of the infrastructure components, – data sources and services –, is the following:
- Short description of the data source or service (what does it do)
- Considered use-case of the data source or service
- Why it is especially appropriate for Science 2.0
- Development stage: (idea, prototype, release candidate, approximated date until the tool is stable)
- Security concept if sensible data are involved
- Description of data:
- What data and what format (reference to standard) is used as input
- What data and what format (reference to standard) is used as output
- What data source is used (e.g. ectel, ed-media, dlbp, citeseer, …)
- Description of API
- A description how it works and technical specifications
- A list of parameters available on the API
- Description of the interface
- URL of the service or data source if available online
- Openness: exchange formats, source code, development process, …
- Evaluation: if applicable an evaluation strategy should be added. Measures can be selected from deliverable 6.2.
The description of the mini applications encompasses the following elements:
- Short description of the tool (what does it do)
- Considered use-case of the tool
- Why it is especially appropriate for Science 2.0
- Tool type: e.g. collaboration, awareness, communication, data aggregation, recommendation, visualisation, service finder
- Development stage: (idea, prototype, release candidate, approximated date until the tool is stable)
- Security concept if sensible data are involved
- Description of data:
- What data and what format (reference to standard) is used as input
- What data and what format (reference to standard) is used as output
- What data source is used (e.g. ectel, ed-media, dlbp, citeseer, …)
- Description of APIs (e.g. REST call, if available)
- Demo available at: URL
- Openness: exchange formats, source code, development process, …
- Screenshot(s)
- Evaluation: if applicable an evaluation strategy should be added. Measures can be selected from deliverable 6.2.
In the following section, we describe the building blocks of the STELLAR science 2.0 mash-ups in short. For an in depth description we refer to the annex.
14 3.1 Infrastructure components
This section gives an overview on the infrastructure components that are necessary to support web-based science support environments. The services envisioned so far encompass data collection in a standardised feed format, a geo-locator for scientists and their institutions, a conference paper database accompanied by a paper retrieval service, a graph visualisation component, and a set of transformation services for publication data. These server as first instances of generic infrastructure components and will be extended with future cycles of the STELLAR science 2.0 mash-ups.
14.1 A) Publication Feeds
In the STELLAR network, most institutions and also the individual scientists keep track of their publications. An open archive[#_ftn16 [16]] was proposed and implemented to collect all TEL-related resources in one place. To avoid the tedious process of manual updating the archive, every institution produces a publication feed instead. The open archive (and any other interested entity) subscribes to this feed and thus automates the process of filling and updating the archive.
Publication feeds are RSS 1.0 feeds enhanced with elements from the Semantic Web for Research Communities (SWRC) and Dublin core (DC) ontologies. The feeds are largely based on the idea of BuRST (Mika, 2005). Modifications were applied where the format was outdated or too broad for the use-case.
This infrastructure component allows us to open up the data feeding process to quickly include additional meta-data of conferences, journals, and institutional repositories. However, maintaining a rich variety of sources and high data quality will further be crucial and opening up deep-web archives poses a future challenge.
14.2 B) Geo-locator for persons and institutes
With geographical data, many different visualizations and mash-ups are possible, e.g. showing relations and data on a map. This allows users to understand for example interrelationships between authors or institutions in the TEL research field. A locator service would be able to find the geo-location of a person or a research institute, based on a set of input data about the person or the institute. With this data, it is then possible to show persons and institutes on a map or geographical data can be added to the statistical analysis.
14.3 C) Conference paper database
Some conferences made their proceedings available for use in the STELLAR project. These proceedings are mostly available in PDF. The PDF-files are parsed using Parscit. This returns XML-files. These are parsed and stored in a database. Some more structure is applied, e.g. matching the authors and e-mail addresses. This data is then stored in the database.
The database allows more advanced querying possibilities in SQL, than the XML-files. Querying the database directly provides faster access to the data for analysis. There is also an Hibernate ORM layer provided in Java. The database is primarily used by developers, by the [index.php?title=Paper_information_retrieval_service paper information retrieval service], and by people interested in data analysis.
14.4 D) Paper information retrieval service
Some conferences made their proceedings available for use in the STELLAR project. These proceedings are mostly available in PDF. We parsed the PDF's and put all information in a database, see (building block “conference paper database”). On top of this database, a RESTful service is provided, to retrieve the data and relationships for a given query.
14.5 E) Feed Transformation Services
An institution wants to publish a feed from its publications according to the publication feed format (see building block “Publication Feeds”). Since publication data is already available in several BibTex files, the BibTex Converter is first used to convert these files into publication feeds. The resulting individual feeds are then merged into a single feed with the help of the Publication Feed Merger. Since there are also publications not related to TEL in the feed, Publication Feed Filter with an appropriate filter file is applied. Afterwards the feed is ready for publishing.
A suite of helper tools aids institutions and individuals in producing, aggregating and refining [index.php?title=Publication_feeds publication feeds]. More precisely, these helper services are:
- BibTex Converter: transforms BibTex files to the publication feed format.
- Publication Feed Aggregator: Combines two
or more publication feeds, merges items by URI.
- Publication Feed Filter: sorts out those publications from a feed
that match certain keywords contained in a taxonomy file.
The services can be orchestrated with the help of DERI pipes[#_ftn17 [17]].
14.6 F) Fossa: FLEXible graph rendering with class
A larger network of researchers shows complex behaviour in co-publishing. To visualise these co-authorship relationships, data from the retrieval service is requested, converted into a co-authorship network, and visualised using the fossa renderer. It bases on the prefuse flare visualisation classes for flex.
Relational data can often be treated as a graph-like structure. Depending on the complexity of the underlying relationship network, these graphs can become quite difficult to visualise. One technology to support the flexible rendering of graphs is with the help of a force-directed lay-out. Fossa encapsulates such a lay-out class: it parses xml files with graphML mark-up as input and renders the contained nodes and edges into a visualisation on screen. The visualisation is written in flash.
15 3.2 Infrastructure Component Overview
These fundamental services form the basis for many of the use-cases envisioned so far and for many more mash-ups dealing with the support of scientists in their research and with the help of web technology. Data collection, storage, retrieval, and presentation services support the development of intelligent and more complex mash-ups. Additionally, we are exploring the use of RSS-aggregators and linked-data for distributed storage of data.
| Name'''' | Publication Feeds Format'''' | Geo-locator for persons and institutes'''' | Conference paper database'''' | Paper information retrieval service'''' | Feed Transformation Services'''' | Graph visualisation'''' |
| Type | Data format | Geo-coder | Database | API | API | Visualisation |
| Use-case | Feed Format | Conference social network | Open archive | Open archive | Publication feeds | Conference social network |
| Development stage | Proposal | Proposal | Prototype | Prototype | Prototype | Prototype |
| Security, Privacy, Copyright | n/a | Public service | (Meta-)data sharing only (to avoid copyright infringement) | Public service | Public service | Public service |
| Data | RSS 1.0, RDF, DC, SWRC, BuRST | Public (meta-) data | SQL-database | JSON | RSS 1.0, BibTex | graphML |
| API | n/a | REST-API | Hibernate ORM Layer available | REST-API | REST-API | REST-API |
| Openness | Open format | Open available | Data dump available | Openly available | Openly available | Open source |
| Evaluation | Using feedburner | - ground truth set - error login - usage tracking | n/a | - traffic - availability - performance | - traffic | - traffic - performance |
| Demo | http://i-know.tugraz .at/2009/papers/feed | n/a | http://groups.google.com/group/scitel20 | http://ariadne.cs.kuleuven.be/scitelrest/author?id=2212 | http://stellar.know-center.at/services | http://www.stellarnet.eu/kmi/graphml/fossa.swf?text=data/deliverables-wp2.xml |
Table 2. Overview on the data sources and services.
16 3.3 Widgets: (mini) web applications
Widgets developed on top of the infrastructure components can vary in size and maturity. The following overview gives an impression on where we currently are with our STELLAR Science 2.0 mash-ups. They are ordered by their development stage, i.e. the most mature web applications are listed first, followed by early beta releases and exploratory prototypes.
16.1 A) Open archive
The Stellar Open Archive (SOA) is a repository for resources (publications, video, tools, data ….) in the field of Technology Enhanced Learning. It comes within the scope of the Open Archive Initiative[#_ftn18 [18]]. On top of the repository, it intends to be also a valuable source of information for researchers and a place for scientific discussions. Thus, it provides also news and events threads from the domain together with blogging facilities.
16.2 B) FlashMeeting
FlashMeeting is an academic research project aimed at understanding the nature of online events and helping users to meet and work more effectively. FM has been used over four years by a great variety of communities of learners and knowledge workers across the globe, supporting online communities and creating new reusable learning objects[#_ftn19 [19]]. The FlashMeeting Project includes an application based on the Adobe Flash 'plug in' and Flash Media Server. Running in a standard web browser window, it allows a dispersed group of people to meet from anywhere in the world with an internet connection. Typically a meeting is pre-booked by a registered user and a URL, containing a unique password for the meeting, is returned by the server. The 'booker' passes this on to the people they wish to participate, who simply click on the link to enter into the meeting at the arranged time[#_ftn20 [20]].
16.3 C) Living deliverables: Wikis
A wiki is a new type of content management system, which provides different authors with the opportunity to easily work together on a joint text. The main challenges of a wiki are to collect and retain collaboratively created knowledge, to encourage users to contribute in this knowledge building community, and to serve as work and presentation platform. One of the main advantages of using a wiki for collaboration is that contributing is not restricted by place, time, or user.
16.4 D) STELLAR Blog
We have set up a WordPress at http://news.stellarnet.eu/ to support multi-author publishing of news like messages on the main portal. Wordpress provides a lot of facilities to syndicate content to other platforms that nicely allows for re-use in various other places – such as the STELLAR open archive and our TELeurope.eu stakeholder platform.
16.5 E) Nextspace Workspace
Nextspace is a Web 2.0 platform for online collaboration. It builds on the space metaphor and allows all users to share, edit, comment, tag, etc.
Nextspace has been successfully used as the main collaboration tool in the iCamp project, a STREP in FP6, where a consortium of 12 partners made extensive use of the platform. It serves as a kind of group blog supported by collaborative wiki editing for the postings, versioning, filtering, notification services.
16.6 F) Publication Feed Widgets
The publication-feed widgets are a collection of widgets for the visualization of and the interaction with [index.php?title=Publication_feeds publication feeds]. Specifically these are:
- Elgg Feed Module: Allows to add feed URLs to an Elgg group and visualizes publication feed items as blog posts. Feed items are stored on the Elgg platform and users can comment and rate publications.
- Elgg Exhibit Module: Visualizes publication feed items in a faceted browser and a timeline view.
The faceted browser allows browsing the publication feeds along the following dimensions: authors, publication years and keywords. Each dimension can be used as a filter criterion. These filtering mechanisms are complemented with full text search. The timeline visualization orders publications chronologically on a timeline and allows users to browse through them.
16.7 G) Deliverable repository
The deliverable repository consists of a set of tool helping to disseminate the results of Stellar. It consists of a deliverable repository form to gather meta-information of deliverables in a standardised way, a list based view page with commenting function and a scientific activity visualisation (composer proxy) based on the deliverable data.
16.8 H) Mobility tracker
The mobility tracker serves as a starting point for the mobility programme planned in WP4. It consists of an entry form, a list of travels, and a map visualisation using google maps. The purpose of the mobility tracker is to collect information on research exchanges and travels of partners in order to inform the project evaluation. Next steps would include a management section that allows to attribute budget to the applications for travel in the mobility programme and a matchmaker application that matches applicants with institutions offering to host a visiting researcher.
16.9 I) Podcasting Tool
The podcasting too records and disseminates podcast sessions from senior academics, visionaries and industry leaders. It aims to help TEL stakeholders with their capacity building according TEL. It consists of a podcast submission system and a presentation interface.
16.10 J) Conference Browser
The Conference Browser allows the attendees to explore the conference. Users can explore the conference in different dimensions, for example time (different years of the conference), location (location of the institutes of the authors), authorship, co-authorship, citations, etc.
16.11 K) Flashmeeting recommenders
This science proxy project is targeted towards increasing awareness and resolving unwanted fragmentation with the help of recommendations extracted from a set of data sources (ranging from publishing data from conferences such as EC-TEL to online communication tools such as Flashmeeting) and – on the intervention side – with the help of the networked communication instrument ‘Flashmeeting’ . This recommender system aims at supporting scientists in the field of TEL to increase the quality of their collaboration in the wider community of peers.
Scientific community structure is revealed in a number of interaction channels (events, authorship of papers, on-line meetings, etc). Data from different sources has to be integrated and combined in order to grasp an accurate impression of the actual state of the art. The basic idea of this set of aligned recommender tools draws from the possibility to build links from comparing relationship information in independent data sets, combining paper authorship and virtual meeting interactions. The recommender system is complemented with an interaction tool, Flashmeeting, which allows deploying the suggested recommendations into actual, successful collaborations.
16.12 L) Graaasp widget
Graaasp is a Web 2.0 application that can serve simultaneously as an aggregation, contextualization, discussion, and networking platform, a shared asset repository, as well as an activity management system. The Graaasp widget will display the spaces (an organizational unit) for the currently logged in user. Moreover, the user will be able to add papers into the existing spaces.
17 3.4 Widget overview
As outlined above, the presented (mini) web applications vary in size and maturity. Relating the widgets back to the areas touched by the use cases presented in chapter 1, the following overview on the widgets can be given (see Figure 14).
Figure 14. Overview on the use case areas covered by the presented widgets.
18 4 Outlook & Conclusion
With this first iteration, a framework has been created within which an initial set of fundamental infrastructure services and a collection of building blocks have been constructed. The framework and the building blocks follow the initial requirements elaborated in chapter 1, but they leave enough room for future innovation in subsequent releases.
Through the choice for widget 1.0 (and Elgg with Wookie as the runtime engine) for the widgets developed, we have paved the ground for close collaboration with our TELeurope.eu stakeholder platform – and we expect others to follow this model.
We see one challenge in integrating the more mature software applications into the open framework. The STELLAR open archive demonstrates, how this integration strategy can work: we have created one widget listing the most recent publications (and allowing for details on demand), thus breaking out one of the many functionalities and allowing to integrate this in other environments. Still, this leaves more work to be done for the next versions.
This is version 1. Version 2 influenced by the success monitoring will follow. We expect our next requirements engineering phase to – again – heavily influence design. We have scheduled a strategic STELLAR meeting for January 25th and 26th, 2010, in Grenoble, to further advance our science 2.0 mash-ups.
19 References
Andersen, P. (2007). What is Web 2.0?: ideas, technologies and implications for education. Citeseer.
Butler, Declan (2005). Science in the web age: joint efforts. Nature 438, pp. 548-549
Bos, B. (2004, February 26). Setting the scope for light-weight web-based applications. Retrieved November 30, 2009, from http://www.w3.org/People/Bos/webapps.html
Gillet, D., El Helou, S., Joubert, M., & Sutherland, R. (2009). Science 2.0: Supporting a Doctoral Community of Practice in Technology Enhanced Learning using Social Software. In Science2.0 for TEL Workshop. ECTEL 2009, France (2009).
Glahn, C., Specht, M., Koper, R. (2009). Perspectives on Tag Clouds for Supporting Reflection in Self-organised Learning, In: Proceedings of the Exploitation of Usage and Attention Metadata (EUAM 09) workshop at the Informatik 2009, Sep 28, 2009 in Lübeck, Germany.
Hartmann, B., Doorley, S., & Klemmer, S. R. (2006). Hacking, Mashing, Gluing: A Study of Opportunistic Design and Development. Retrieved from http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.66.1603
Kieslinger, B., & Lindstaedt, S. (2009). Science 2.0 Practices in the Field of Technology Enhanced Learning. In Science2.0 for TEL Workshop. ECTEL 2009, France (2009).
Maximilien, E., Ranabahu, A., & Gomadam, K. (2008). An Online Platform for web APIs and Service Mash-ups. Internet Computing, IEEE, 12(5), 32-43. doi: 10.1109/MIC.2008.92
Mika, P. (2005, March 14). Bibliography Management using RSS Technology (BuRST). Retrieved December 1, 2009, from http://209.85.229.132/search?q=cache:1EL8PyVji4UJ:www.cs.vu.nl/~pmika/research/burst/BuRST.html+peter+mika+Bibliography+Management+using+RSS+Technology&cd=1&hl=en&ct=clnk&gl=uk
O'Reilly, T. (2005, September 30). What Is Web 2.0. Design Patterns and Business Models for the Next Generation of Software. Retrieved November 29, 2009, from http://oreilly.com/web2/archive/what-is-web-20.html
Palmér, M., Sire, S., Bogdanov, E., Gillet, D., & Wild, F. (2009). Mapping web Personal Learning Environments. In F. Wild, M.Kalz, M. Palmér, D. Müller: Mash-Up Personal Learning Environments. Proceedings of the 2nd Workshop MUPPLE'09, Nice, France, September 29, 2009, CEUR Workshop Proceedings, ISSN 1613-0073, online CEUR-WS.org/Vol-506/.
Raman, T. V. (2009). Toward 2W, beyond Web 2.0. Commun. ACM, 52(2), 52-59. doi: 10.1145/1461928.1461945
Shneiderman, B. (2008). COMPUTER SCIENCE: Science 2.0. Science, 319(5868), 1349-1350. doi: 10.1126/science.1153539
Shneiderman, B. (2007). web science: a provocative invitation to computer science. Commun. ACM, 50(6), 25-27. doi: 10.1145/1247001.1247022
Underwood, J., Luckin, R., Smith, H., Walker, K., Rowland, D., Fitzpatrick, G., Good, J., Benford, S. (2009): Reflections on Participatory Science for TELSci2.0, In: Science2.0 for TEL, Workshop at the 4th European Conference on Technology-Enhanced Learning (ECTEL’09), Nice, France
Waldrop, M. M. (2008). Science 2.0. Scientific American, 298(5), 68.
Wild, F., & Sobernig, S. (2006). ICamp D3.1 Interoperability Framework Draft for the Distributed Open Virtual Learning Environment. Retrieved from http://www.icamp.eu/wp-content/uploads/2007/05/d31___icamp___interoperability-framework-draft.pdf
Zang, N., Rosson, M. B., & Nasser, V. (2008). Mash-ups: who? what? why? In CHI '08 extended abstracts on Human factors in computing systems (pp. 3171-3176). Florence, Italy: ACM. doi: 10.1145/1358628.1358826
20 Annex A: Terminology
Widget: Mini applications: web app:
Are usually “small, platform independent programs that are downloaded on demand and executed inside a client program, such as a browser. They are thus like Java applets, but more “script-like” than “program-like” and therefore easier to write in many cases (though harder in others). They have a clearly separated user interface, that allows webapps to be easily adapted to different devices.” (Bos, 2004). It is often a combination of a user interface or view and/or a service.
User interface: Is the part of the tool with that the user can interact. It is part of the client.
Service: The service provides the data for the visualization and is often implemented on the server.
A web widget is a miniature web application designed to work within a particular kind of framework, the widget engine.
Widget engine
A widget engine is designed to enable widgets to be used in a wide range of web applications. An example is the Wookie engine.
Widget container
A widget container is a web application that contains widgets. Widget containers take care of all the main business of a web application (e.g. user log in or managing content). The widget engine supplies some of these functionalities. Examples are Wordpress, Elgg, Moodle, and Sharepoint.
21 Annex B: Building Blocks
22 Infrastructure components
This section gives an overview on the infrastructure components that are necessary to support web-based science support environments. The services envisioned so far encompass data collection in a standardised feed format, a geo-locator for scientists and their institutions, a conference paper database accompanied by a paper retrieval service, a graph visualisation component, and a set of transformation services for publication data. These server as first instances of generic infrastructure components and will be extended with future cycles of the Stellar Science 2.0 mash-ups.
22.1 A) Publication Feeds
Bibliography Management using RSS Technolgy (in short: BuRST) is a feed format developed by Peter Mika describing bibliographic metadata (Mika, 2005). For STELLAR we propose an adapted version of this standard for the automated exchange of publication metadata.
22.1.1 Use-case
In the STELLAR network, most institutions and also the individual scientists keep track of their publications. An open archive[#_ftn21 [21]] was proposed and implemented to collect all TEL-related resources in one place. To avoid the tedious process of manual updating the archive, every institution produces a publication feed instead. The open archive (and any other interested entity) subscribes to this feed and thus automates the process of filling and updating the archive.
22.1.2 Appropriateness for a modern Science 2.0
One of the prerequisites for speeding up feedback cycles in Science 2.0 is the open access to publications for the whole community (Kieslinger & Lindstaedt, 2009). RSS feeds are a popular means for lightweight exchange of information in the Web 2.0 (Gillet et al., 2009). Publication feeds therefore have the advantage to provide critical data for the introduction of modern Science 2.0 in a form collectively readable by the existing Web 2.0 infrastructure.
22.1.3 Development stage
This is a proposal. Comments and suggestions are welcome. Please join the BuRST format adaption discussion.
22.1.4 Tool type
Feed format
22.1.5 Data
BuRST is based on RSS 1.0 with RDF extensions. An item of a BuRST feed essentially consists of three parts (see listing below):
- The RSS description, mainly for display within generic RSS applications, like Google Reader or Firefox.
- The burst:publication Tag linking the RSS description with the RDF description.
- The RDF description for semantic markup of publication metadata (currently with DC 1.1, SWRC 0.3, and BuRST 0.1.).
For a exact specification, please refer to the BuRST format adaption discussion for the moment.
22.1.6 Demo
See http://i-know.tugraz.at/2009/papers/feed for a demo.
22.1.7 Openness
This is an open format.
22.1.8 Evaluation
Publication feed usage can be evaluated by using Feedburner.
22.2 B) Geo-locator for persons and institutes
With geographical data, many different visualizations and mash-ups are possible, e.g. showing relations and data on a map. This allows users to understand for example interrelationships between authors or institutions in the TEL research field. A locator service would be able to find the geo-location of a person or a research institute, based on a set of input data about the person or the institute. With this data, it is then possible to show persons and institutes on a map or geographical data can be added to the statistical analysis.
22.2.1 Use-case
Many different use-cases are possible. However, mainly, the use-cases can be generalized as enriching existing data sources with geo-graphical data. When analysing the TEL research field, we can add an extra dimension to the data by using geo-location data. This allows showing interrelations of persons, institutes and countries.
22.2.2 Appropriateness for a modern Science 2.0
Geographical data is very important to generalize data to different granularities. This service will be used as input for multiple Science 2.0 mash-ups and visualizations.
22.2.3 Development stage
Proposal stage.
22.2.4 Security
This is a publicly available service. No authorization applies.
22.2.5 Data
Publication (meta-)data, personal data, data from different Web 2.0 and Semantic Web sources, like SIGMA[#_ftn22 [22]] and the Yahoo! or Google Maps geo-coding service for getting standardized geo-metadata.
22.2.6 API definition
It provides a REST API, with an HTTP GET-method to retrieve the geo-location. The input for the service will be a set of data available about the institute or person, e.g. the name of the institute, publication data, and possible addresses. The output is an address, the city, an ISO-standard-based country code and geo-coordinates.
22.2.7 Openness
Openly available.
22.2.8 Evaluation
- Evaluate the service with a ground truth set: The quality of the locator can be evaluated against a labelled ground truth set. The result of the evaluation will result in an accuracy percentage. The hard part in evaluating this is that we will need a labelled ground truth data set. This probably needs to be manually created.
- Improve the service through the applications: The service could also be improved and evaluated by the results from the mash-ups and visualisations build on top of the service. When users or developers of the mash-up applications notice errors in the locator, we could improve this.
- usage tracking: From deliverable 6.2, we would like to track following measures: traffic sources, availability and performance.
22.3 C) Conference paper database
Some conferences made their proceedings available for use in the STELLAR project. These proceedings are mostly available in PDF. The PDF-files are parsed using Parscit. This returns XML-files. These are parsed and stored in a database. Some more structure is applied, e.g. matching the authors and e-mail addresses. This data is then stored in the database.
22.3.1 Use-case
The database allows more advanced querying possibilities in SQL, than the XML-files. Querying the database directly provides faster access to the data for analysis. There is also an Hibernate ORM layer provided in Java. The database is primarily used by developers, by the [index.php?title=Paper_information_retrieval_service paper information retrieval service], and by people interested in data analysis.
22.3.2 Appropriateness for a modern Science 2.0
Analysing TEL conferences can provide a great insight into the TEL research field and research parties. It can uncover for example new insights in collaborations between researchers and provide statistics on researchers and institutes.
22.3.3 Development stage
There is a database available of all proceedings of the EC-TEL Conference[#_ftn23 [23]] (from 2006 until 2009).
22.3.4 Security
We are not allowed to share the PDF-files. We can only share the generated data.
22.3.5 Data
A database dump is available upon request or downloadable from the SciTEL group[#_ftn24 [24]].
An Entity-Relationship schema of the database can be seen in the next picture.
Figure 15. Entity relationship schema of the conference database.
22.3.6 API
There is a Hibernate ORM layer in Java available on the SciTEL group. Hibernate allows to communicate with a relational database in an object-oriented way.
22.3.7 Openness
Openly available in form of a database dump.
22.3.8 URL
http://groups.google.com/group/scitel20
22.4 D) Paper information retrieval service
Some conferences made their proceedings available for use in the STELLAR project. These proceedings are mostly available in PDF. We parsed the PDF's and put all information in a database, see (building block “conference paper database”). On top of this database, a RESTful service is provided, to retrieve the data and relationships for a given query.
22.4.1 Use-case
This service allows developers to develop an application that makes use of the conference paper data more quickly.
22.4.2 Appropriateness for a modern Science 2.0
Analysing TEL conferences can provide a great insight into the TEL research field and research parties. It can uncover for example new insights in collaborations between researchers and provide statistics on researchers and institutes.
22.4.3 Development stage
Very basic version is available. More methods will be provided when applications need it.
22.4.4 Security
This is a publicly available service. No authorization applies.
22.4.5 Data
Currently the REST service returns information about an author, his affiliation and the papers published.
22.4.6 API definition
The service is written in Java as a REST service making use of JAX-RS (JSR-311). Extending the service is very easy and only requires implementing one Java class. Currently only two methods are implemented. One retrieves all the author names in the database and the other one retrieves information about an author and his papers. As input to the service GET-method, you need to provide the name of the author and you can provide an interval of publication years. This will limit the returned papers of the author. The service returns results in JSON format.
An example of the query response for searching for all author names can be seen in this JSON code example:
| [{"author":"Daniel Weiss","email":"dweiss@chimereguide.org","authorId":95},
{"author":"Teofilo Romera","email":"teo@comp.leeds.ac.uk","authorId":154}, {"author":"Julika Matravers","email":"julika@comp.leeds.ac.uk","authorId":157}, {"author":"Luis Canas","email":"lcanas@gsyc.escet.urjc.es","authorId":156}, {"author":"Michel C Desmarais","email":"michel.desmarais@polymtl.ca","authorId":109}, {"author":"Chris Tebb","email":"chrispy@comp.leeds.ac.uk","authorId":152}, {"author":"Christoph Meinel","email":"meinel@hpi.uni-potsdam.de","authorId":1181}, {"author":"Yusuke Hayashi","email":"hayashi@ei.sanken.osaka-u.ac.jp","authorId":260}, {"author":"Kit Logan","email":"K.Logan@ioe.ac.uk","authorId":391}, {"author":"Harvey Mellar","email":"H.Mellar@ioe.ac.uk","authorId":393}, {"author":"Nicolas Manin","email":"Nicolas.Manin@laposte.net","authorId":676},...] |
Table 3. Query response in JSON.
An example of the query response for information of one author can be seen in the JSON code example in Error! Reference source not found..
| [{
"author": "Erik Duval", "authorId": 2212, "dblpUrl": "http://dblp.uni-trier.de/db/indices/a-tree/d/Duval:Erik.html", "email": "Erik.Duval@cs.kuleuven.be", "papers": [ { "abstract": "Technologies that solve the scarce availability of learning objects have created the opposite problem: abundance of choice. The solution to that problem is relevance ranking. Unfortunately current techniques used to rank learning objects are not able to present the user with a meaningful ordering of the result list. […] ", "affiliations": [ { "address": "Celestijnenlaan 200A, B-3001, Heverlee, Belgium", "affiliationId": 149, "city": "3001 Heverlee", "cleanName": "Dept. Computerwetenschappen, Katholieke Universiteit Leuven", "country": "BE", "latitude": 50.86394, "longitude": 4.68038, "name": "Dept. Computerwetenschappen, Katholieke Universiteit Leuven" }, { "address": "Va Perimetral Km. 30.5, Guayaquil - Ecuador", "affiliationId": 148, "city": "Guayaquil", "cleanName": "Information Technology Center, Escuela Superior Politcnica", "country": "EC", "latitude": -2.20382, "longitude": -79.89746, "name": "Information Technology Center, Escuela Superior Politcnica del Litoral" } ], "authors": [ { "author": "Erik Duval", "authorId": 2212, "dblpUrl": "http://dblp.uni-trier.de/db/indices/a-tree/d/Duval:Erik.html", "email": "Erik.Duval@cs.kuleuven.be" }, { "author": "Xavier Ochoa", "authorId": 2214, "dblpUrl": "http://dblp.uni-trier.de/db/indices/a-tree/o/Ochoa:Xavier.html", "email": "xavier@cti.espol.edu.ec2" } ], "citations": [], "filePath": "/Users/sten/Documents/research/summerproject/correctedXMLs/2007/fulltext-19.txt.xml", "keywords": [ { "keyword": "Topical Relevance", "keywordId": 218 }, { "keyword": "learning objects", "keywordId": 141 }, { "keyword": "Relevance Ranking", "keywordId": 217 } ], "paperId": 109, "title": "Relevance Ranking Metrics for Learning Objects", "year": 2007 } ] }] |
Table 4. JSON wrapped query response.
22.4.7 Openness
Openly available.
22.4.8 URL
The two available methods are publicly available: Retrieving all authors:
http://ariadne.cs.kuleuven.be/scitelrest/authors
This returns all authors and their email and database id.
Retrieving data of one author and his publications:
http://ariadne.cs.kuleuven.be/scitelrest/author?id=2212
This returns all data available in the database on one author, his relationships and all his publications
Retrieving data of one author and his publications between a start and end year:
[http://ariadne.cs.kuleuven.be/scitelrest/author?id=2212&startYear=2006&endYear=2009 http://ariadne.cs.kuleuven.be/scitelrest/author?id=2212&startYear=2006 &endYear=2009]
This method allows limiting the amount of publications returned based on the publication year.
22.4.9 Evaluation
From deliverable 6.2, we would like to track following user-tracking measures:
- Traffic sources
- Availability
- Performance.
This is currently not implemented.
22.5 E) Feed Transformation Services
A suite of helper tools aids institutions and individuals in producing, aggregating and refining [index.php?title=Publication_feeds publication feeds]. More precisely, these helper services are:
- BibTex Converter: transforms BibTex files to the publication feed format.
- Publication Feed Aggregator: Combines two
or more publication feeds, merges items by URI.
- Publication Feed Filter: sorts out those publications from a feed
that match certain keywords contained in a taxonomy file.
The services can be orchestrated with the help of DERI pipes[#_ftn25 [25]].
22.5.1 Use-case
An institution wants to publish a feed from its publications according to the publication feed format (see building block “Publication Feeds”). Since publication data is already available in several BibTex files, the BibTex Converter is first used to convert these files into publication feeds. The resulting individual feeds are then merged into a single feed with the help of the Publication Feed Merger. Since there are also publications not related to TEL in the feed, Publication Feed Filter with an appropriate filter file is applied. Afterwards the feed is ready for publishing.
22.5.2 Appropriateness for a modern Science 2.0
As already mentioned in [index.php?title=Publication_feeds publication feeds], open access to publication data is essential for Science 2.0. To kick-off the opening of archives, easy-to-use tools for institutions are needed. Services are especially apt for that since they are the cornerstone of Web 2.0, allowing for loosely coupled systems and simple syndication (O'Reilly, 2005).
22.5.3 Tool type
Data aggregation, awareness
22.5.4 Development stage
Perpetual beta: In the tradition of Web 2.0, these services will mature with user feedback and insight generated from statistics in short release cycles.
22.5.5 Security concept
There is no personal data involved. Copyright issues will not arise as only metadata is transferred through the server.
22.5.6 Data
- Bibtex Converter: takes any bibtex file as input and outputs a publication feed.
- Feed Merger: takes two or more publication feeds as input and outputs a single publication feed.
- Feed Filter: takes a publication feed and a filter file as input and outputs a filtered publication feed.
22.5.7 APIs
BibTex Converter takes the following parameters:
- bib: The URL to the BibTex file.
- keyword_del: If BibTex records contain a keywords, this parameter is used to define the standard delimiter.
- pub: The name of the publisher.
- channel: The resulting feed's channel name.
- desc: The resulting feed's channel description.
- domain: The publisher's domain. There are two possibilities how the converter builds the URI of a publication feed item. By default, it takes the publisher's domain and appends the item identifier of the BibTex entry. If, for example, the publisher's domain is "http://stellarnet.eu/" and the BibTex identifier is "Lindstaedt:2003a", the resulting publication feed item URI will be "http://stellarnet.eu/Lindstaedt:2003a". This is overruled, if any of the BibTex items has defined a URL in the "note"-field using the following syntax:
note="\url|[URL]|",
for example note="\url|http://stellarnet.eu/Lindstaedt:2003a|".
Publication Feed Merger takes the following parameters:
- feeds[]: The URLs of the feeds to merge.
- pub: The name of the publisher.
- domain: The publisher's domain.
- channel: The resulting feed's channel name.
- desc: The resulting feed's channel description.
Publication Feed Filter takes the following parameters:
- feed: The URL of the feed to filter.
- tax: The URL of the taxonomy file. The taxonomy file must contain all the concepts in the taxonomy separated by line feeds. For a demo see: http://stellar.know-center.tugraz.at/data/tel.csv.
22.5.8 Demo
The services can be accessed via a server at the Know-Center:
http://stellar.know-center.at/services
for the converter.
- http://stellar.know-center.at/services/merge_feeds.php for the merger.
- http://stellar.know-center.at/services/filter_feed.php for the filter.
There is a frontend to the BibTex converter:
http://stellar.know-center.tugraz.at/html/convert.php
There is also a DERI Pipes Installation for orchestration at:
http://stellar.know-center.tugraz.at:8080/pipes/
22.5.9 Openness
The services are free for everyone to use. There is no registration or API key required. The source code will be made publicly available at a later stage.
22.5.10 Screenshots
Figure 16. Sample DERI pipe for service orchestration.
22.5.11 Evaluation
Each server access is logged for later analysis.
22.6 F) Fossa: FLEXible graph rendering with class
Relational data can often be treated as a graph-like structure. Depending on the complexity of the underlying relationship network, these graphs can become quite difficult to visualise. One technology to support the flexible rendering of graphs is with the help of a force-directed lay-out. Fossa encapsulates such a lay-out class: it parses xml files with graphML mark-up as input and renders the contained nodes and edges into a visualisation on screen. The visualisation is written in flash.
22.6.1 Use-case
A larger network of researchers shows complex behaviour in co-publishing. To visualise these co-authorship relationships, data from the retrieval service is requested, converted into a co-authorship network, and visualised using the fossa renderer. It bases on the prefuse flare visualisation classes for flex.
22.6.2 Appropriateness for a modern Science 2.0
Many aspects about science networks are not visible from the raw data, but need visualisation techniques (or other ways of supporting the analysis).
22.6.3 Development stage
The precompiled swf-file is available and so is the flex source-code.
22.6.4 Security
n/a (depends on the security mechanisms of the data provider)
22.6.5 Data
n/a (has to be provided)
22.6.6 API
The graphML files can contain the following node attributes to configure the visualisation: colour, linecolour, size, alpha (i.e. transparency), label (i.e. character string), zoom (e.g. ‘nextData.xml’). Thereby, size manipulates node radius and node mass. The edge attributes that can be configured are: colour, alpha (i.e. transparency), weight, label (i.e. character string). Thereby, weight manipulates the tension, restLength, and linewidth.
To render a graphML file, the url has to be handed over to the rendering service:
http://www.stellarnet.eu/kmi/graphml/fossa.swf?text=[url]
The graphML format is described here in more detail:
http://graphml.graphdrawing.org/
An example of data wrapped into graphML follows here:
Figure 17. graphML example.
Rendering, for example, the co-authorship relations between contributors, editors, and the internal reviewers within one work package of the previous network of excellence Prolearn, results in a visualisation such as depicted below in Error! Reference source not found..
Figure 18. Co-authorship relations depicted in a visualisation.
22.6.7 Openness
Source code available as open source. Compiled swf-file available.
22.6.8 URL
http://www.stellarnet.eu/kmi/graphml/fossa.swf?text=data/deliverables-wp2.xml
23 Widgets: (mini) web applications
Widgets developed on top of the infrastructure components can vary in size and maturity. The following overview gives an impression on where we currently are with our STELLAR Science 2.0 mash-ups. They are ordered by their development stage, i.e. the most mature web applications are listed first, followed by early beta releases and exploratory prototypes.
23.1 A) Open archive
The Stellar Open Archive (SOA) is a repository for resources (publications, video, tools, data ….) in the field of Technology Enhanced Learning. It comes within the scope of the Open Archive Initiative[#_ftn26 [26]]. On top of the repository, it intends to be also a valuable source of information for researchers and a place for scientific discussions. Thus, it provides also news and events threads from the domain together with blogging facilities.
23.1.1 Use-case
- From the perspective of a researcher in TEL: Use-cases 1 & 2
- From the perspective of the STELLAR network: Use-case 1
- Widgets under consideration.
23.1.2 Appropriateness for a modern Science 2.0
- Dissemination of scientific production
- Communication/collaboration: blog and forum attached to every resource
- Recommendation: users based clusters of similar resources are proposed
23.1.3 Development stage
Mainly stable. Some features still prototype.
23.1.4 Security
No sensible data involved regarding bibliographical references. Deposit under the responsibility of the depositor regarding copyright issues. Submitting requires authentication (account owner).
23.1.5 Data
- input: Data approx. equivalent to a Dublin core set. Format: html forms, BibTex import. Feeds submitted by Stellar institutions.
- output : OAI-PMH compliant when the level of description is sufficient. RSS 2.0 feeds. And enriched RSS1 feeds (according to the format suggested in the builing block: “Publication Feeds”). Example query: http://oa.stellarnet.eu/open-archive/rdf?search
- source used: online submission, Conference repositories (such as ECTEL, ICCE ...), repositories such as DBLP, any relevant source at least partly processable. Feeds submitted by Stellar institutions.
23.1.6 API definition
OAI-PMH interface, Base URL: http://oa.stellarnet.eu/open-archive/oai. The Open Archives Initiative Protocol for Metadata Harvesting (OAI-PMH) is a low-barrier mechanism for repository interoperability. Data Providers are repositories that expose structured metadata via OAI-PMH. Service Providers then make OAI-PMH service requests to harvest that metadata. OAI-PMH is a set of six verbs or services that are invoked within HTTP[#_ftn27 [27]].
The SOA is currently a 'Data Provider'. It may have to become also a 'Service Provider' in order to harvest metadata coming from other OAI-PMH compliant institutional repositories (such as EPFL InfoScience).
From another perspective, considering the “Publication Feed” building block and its possible use by the consortium, the SOA will off course have to process these feeds coming from institutions and thus develop a specific interface to accept and process the submitted feeds.
23.1.7 Openness
Openly available. Based on AlphaComplex CMS: open source software (CECILL licence).
23.1.8 Evaluation
The use of Stellar OA should be evaluated in two ways: as a target repository (upload of bibliographical references and original material) and as a valuable source of information (including visits, participation in online discussion of resources and download of material).
The key indicators are described in Deliverable D6.2 (see section “Monitoring the success of the open archive).
Running tools, to ensure the evaluation baseline, are currently the archive database itself, Awstats (local log analysis), Google Analytics (script included) and FeedBurner (feeds submitted)
Additionally the efficiency of the Facilitators enrolled at every Stellar partner institution can be measured/estimated.
Lastly, a set of interviews with representative Stellar members will complete the evaluation.
23.1.9 Demo
Demo available at URL http://oa.stellarnet.eu
23.1.10 Screenshot
Figure 19. Stellar Open Archive (stand alone).
| 246px | 252px |
Figure 20. A Stellar Open Archive widget in Elgg.
23.2 B) FlashMeeting
FlashMeeting is an academic research project aimed at understanding the nature of online events and helping users to meet and work more effectively. FM has been used over four years by a great variety of communities of learners and knowledge workers across the globe, supporting online communities and creating new reusable learning objects[#_ftn28 [28]]. The FlashMeeting Project includes an application based on the Adobe Flash 'plug in' and Flash Media Server. Running in a standard web browser window, it allows a dispersed group of people to meet from anywhere in the world with an internet connection. Typically a meeting is pre-booked by a registered user and a URL, containing a unique password for the meeting, is returned by the server. The 'booker' passes this on to the people they wish to participate, who simply click on the link to enter into the meeting at the arranged time[#_ftn29 [29]].
23.2.1 Use-case
Helen, a researcher in TEL wants to meet up with Mark, Henry, and Claudia in the STELLAR Network of Excellence to discuss the next steps according to the current development in their work package. Helen enters the FlashMeeting website and books a meeting for a certain time and duration. The attendees of the meeting are informed via e-mail. When the meeting starts, all researchers enter the meeting with their browser. At first they have a discussion on what the next steps should be. Every participant in the meeting has headphones, a microphone, and – although not absolutely necessary but highly recommended – a webcam. Whoever wants to speak presses the “Start Broadcasting” button. Others can interrupt or join the queue. People having short comments just type them into the chat window.
During the discussion Mark uploads a file that he deems important for the discussion. Claudia contributes a link to a website with interesting information. Helen listens carefully and summarizes all the thoughts on the whiteboard. Henry does not quite agree and makes some amendments to Helen’s summary. Afterwards they have a vote on the collaboratively constructed work plan.
All activities are stored within the system, and give those people who had not had time to attend the meeting the possibility of replaying it. It also helps Helen to write the protocol, to revisit some of the discussion points, and to make further use of the material.
23.2.2 Appropriateness for a modern Science 2.0
FM is a collaboration and meeting tool on the web and therefore bridges the local distances of the participants of a meeting. Additionally the storage of the meeting and the possibility of replaying and making further use of it are important features for an interconnected but distributed scientific community.
23.2.3 Development stage
Released
23.2.4 Security concept
Several authentication options available.
23.2.5 Data
XML transcript of the meeting available.
Figure 21. Transcript of a flashmeeting.
23.2.6 Demo
http://fm.ea-tel.eu/fm/flashmeeting.php?pwd=demo-demo
23.2.7 Openness
Participation free for everyone, booking for STELLAR members or EA-TEL members.
23.2.8 Screenshots
Figure 22. Enhanced attendance information in flashmeeting.
23.3 C) Living deliverables: Wikis
A wiki is a new type of content management system, which provides different authors with the opportunity to easily work together on a joint text. The main challenges of a wiki are to collect and retain collaboratively created knowledge, to encourage users to contribute in this knowledge building community, and to serve as work and presentation platform. One of the main advantages of using a wiki for collaboration is that contributing is not restricted by place, time, or user.
23.3.1 Use-case
Use-case 1: A STELLAR scientist wants to work on a specific research-area in TEL together with other STELLAR researchers. Therefore she sets up a WIKI, starts to fill in her topics of interest, and adds all information she already has. Then she adds her new pages to her watch-list to be automatically informed by the system, when this page is modified or updated by other users. She sends out an email to the other researchers, and informs them about the wiki. The invited researchers open the wiki and can easily enter their additional information and suggestions to the current topics, but they can also broaden the wiki by inserting new topics. As a result, the researchers will get a continuously growing content management system with collaboratively created pages, covering all information concerning TEL.
Use-case 2: A STELLAR researcher wants to write a document, for example a deliverable, together with other scientists. She puts the starting content into the wiki and adds the pages to her watch-list. The other scientists are then able to enter their parts, like paragraphs, images, further topics etc. into the wiki. When the document is finished and all contributors are satisfied with the result of the collaboration, the researcher can export the necessary content of the wiki into a PDF document.
23.3.2 Appropriateness for a modern Science 2.0
A wiki is a very important tool for collaboration between researchers. The wiki can be used as basis for different types of collaborative work as well as for collection and providing knowledge in the area of TEL.
23.3.3 Development stage
The MediaWiki is open software can easily be downloaded at http://www.mediawiki.org/wiki/Download.
23.3.4 Security
Authorization is necessary for editing pages as well as adding watches. OpenID support is enabled. No authorization is necessary for read access.
23.3.5 API definition
For the usage of the MediaWiki a complete definition of the API can be found at http://www.mediawiki.org/wiki/API.
23.3.6 Openness
This is open source software.
Screenshots
Figure 23. STELLAR wiki for a deliverable.
23.4 D) STELLAR Blog
We have set up a WordPress at http://news.stellarnet.eu/ to support multi-author publishing of news like messages on the main portal. Wordpress provides a lot of facilities to syndicate content to other platforms that nicely allows for re-use in various other places – such as the STELLAR open archive and our TELeurope.eu stakeholder platform.
23.4.1 Use-case
A STELLAR scientist wants to report a news item such as a workshop coming up. She goes to WordPress and writes a new blog posting containing the call for papers. When the workshop is finished, she returns to the blog to post a photo and a summary of the event. Many other STELLAR scientists in the open network of excellence have subscribed to the feed, so they are automatically notified about this news item. Additionally the feed contents are shown on various institutional web pages and in several TEL-focused portals.
23.4.2 Appropriateness for a modern Science 2.0
Allows for re-use of news messages in various places, supports aggregation, quick way to get a news item out and build up a significant number of subscribers.
23.4.3 Tool type
Communication, collaboration.
23.4.4 Development stage
Released.
23.4.5 Security concept
WordPress authentication / authorization routines.
23.4.6 Data
News messages RSS feed generated possibilities to display other feeds in a widget.
23.4.7 APIs
Editing API, Pingback API, many more plugins available.
23.4.8 Demo
http://news.stellarnet.eu and http://news.stellarnet.eu/wp-login.php
23.4.9 Openness
Open source.
23.4.10 Screenshots
Figure 24. The STELLAR blog (admin mode).
23.5 E) Nextspace Workspace
Nextspace is a Web 2.0 platform for online collaboration. It builds on the space metaphor and allows all users to share, edit, comment, tag, etc.
23.5.1 Use-case
Nextspace has been successfully used as the main collaboration tool in the iCamp project, a STREP in FP6, where a consortium of 12 partners made extensive use of the platform. It serves as a kind of group blog supported by collaborative wiki editing for the postings, versioning, filtering, notification services.
23.5.2 Appropriateness for a modern Science 2.0
Workspace is a combination of various social software elements, like free tagging, RSS feeds to syndicate content with other websites
23.5.3 Development stage
Running version available that has been used extensively, but certain improvements are currently under development.
23.5.4 Security
Access can be restricted via username and password; there is also the option to publish content from the workspace to the web.
23.5.5 Data
RSS feeds, iCal feeds API Openness
Open Source Licence.
23.5.6 Evaluation
The Nextspace workspace is evalutated in a case study description on the use of workspace in iCamp[#_ftn30 [30]].
23.5.7 Demo
Demo is available here: http://nextspace.zsi.at/spaces/STELLAR.
Screenshot
Figure 25. STELLAR Nextspace.
23.6 F) Publication Feed Widgets
The publication-feed widgets are a collection of widgets for the visualization of and the interaction with [index.php?title=Publication_feeds publication feeds]. Specifically these are:
- Elgg Feed Module: Allows to add feed URLs to an Elgg group and visualizes publication feed items as blog posts. Feed items are stored on the Elgg platform and users can comment and rate publications.
- Elgg Exhibit Module: Visualizes publication feed items in a faceted browser and a timeline view.
The faceted browser allows browsing the publication feeds along the following dimensions: authors, publication years and keywords. Each dimension can be used as a filter criterion. These filtering mechanisms are complemented with full text search. The timeline visualization orders publications chronologically on a timeline and allows users to browse through them.
23.6.1 Use-case
Use-case 1: The researcher would like to get an overview of papers in the field of TEL and be notified of new papers. He or she opens the “Publication Feeds” tab on the widget platform and adds various standardized publication feeds from conferences/institutions he or she knows. The bibliographic data is loaded into the widget and presented in a well-arranged and visually attractive manner. The user is able to rate each publication and comment on it. Whenever there are new papers published, the widget is updated and the feed emitted by the widget reports a new paper.
Use-case 2: A researcher would like to be informed about trending topics in the field of TEL. She opens the "Publication Visualization" tab on the widget platform. She is presented with a faceted browsing view. A tag cloud (or similar) aggregated from the keywords is shown to the user. She then restricts the data to certain years to see the changes in the tag cloud. To complete the picture a timeline with the publication dates of the publications is shown to the user.
23.6.2 Appropriateness for a modern Science 2.0
The subscriber widgets are the logical continuation of [index.php?title=Publication_feed_publisher_services Publication feed publisher services] on the end-user side. Whereas the services aid the producer in generating a publication feed, the subscriber widgets let the recipient consume and manipulate these feeds in the style of Web 2.0. Users can collectively contribute to the data base by adding their own feeds, they can help in identifying good publications by rating them, and interact with each other by leaving comments. The visualization widget provides them with filtering and searching facilities for the aggregated data.
23.6.3 Tool type
Data aggregation, awareness
23.6.4 Development stage
Perpetual beta: In the tradition of Web 2.0, these Elgg modules will mature with user feedback and insight generated from statistics in short release cycles.
23.6.5 Security concept
There is no personal data involved. Copyright issues will not arise as only metadata is transferred through the server.
23.6.6 Data
- The Elgg Feed Module can handle RDF/XML and RSS data.
- The Elgg Exhibit module can handle RDF/XML Publication Feeds. The following input formats are accepted: RDF/XML, JSON Exhibit, BibTex, Google spreadsheet, Excel spreadsheet, EditGrid spreadsheet (see http://simile.mit.edu/wiki/Exhibit/For_Authors). The babel translation service (http://simile.mit.edu/babel/) is used to transform the different data formats into the JSON Exhibit format which can be visualized by the tool. The tool can output the visualized data RDF/XML, HTML, Exhibit JSON, semantic wiki text or tabulator separated values.
23.6.7 APIs
The Elgg Feed Module builds upon Blogextend and the Simplepie RSS Feed Integrator[#_ftn31 [31]].
The simile exhibit library[#_ftn32 [32]] and the timeline library[#_ftn33 [33]] are used by the Exhibit module. Both libraries use the babel translation service to convert incoming data into the Exhibit JSON format.
23.6.8 Demo
To be announced soon
- Elgg Feed module: http://iisv027.joanneum.at/elgg/pg/blog/group:17
- Elgg Exhibit module: http://iisv027.joanneum.at/elgg/mod/visitelf/pages/vis.php
23.6.9 Openness
All Elgg modules will be availble as as open source modules.
23.6.10 Screenshots
Figure 26. Publication Feed Subscriber Widget from the Elgg exhibit widget.
23.6.11 Evaluation
Widget usage will be logged with the extensive logging facilities already built into Elgg.
23.7 G) Deliverable repository
The deliverable repository consists of a set of tool helping to disseminate the results of Stellar. It consists of a deliverable repository form to gather meta-information of deliverables in a standardised way, a list based view page with commenting function and a scientific activity visualisation (composer proxy) based on the deliverable data.
23.7.1 Type
Awareness, communication, data gathering and visualization
23.7.2 Use-case
The deliverable repository combines the use cases “publication feedback” and “publication visualization” and could be used as input for the use cases “automated update of the open archive”, “personal window” and “TEL community news”.
23.7.3 Appropriateness for a modern Science 2.0
The comment function of the deliverable repository enables quick feedback about the work of the authors. With the produced feeds, syndication and recommendation become possible. Especially the RDF feed contains detailed data about the deliverables making the usage of Semantic Web applications possible. The visualization helps researchers gaining deeper insight into the social structure of the projects deliverable review process.
23.7.4 Development stage
Released.
23.7.5 Security
Authentication required to store the deliverables, but not for viewing them.
23.7.6 Data
- input: Forms used to gather metadata of the deliverable, like title, editors, contributors, and reviewers.
- Output: RSS and RDF.
23.7.7 API definition
n/a
23.7.8 Openness
To store the deliverable, authentication is necessary. Open available are the deliverables and the scientific activity visualization.
23.7.9 Evaluation
web statistics will observe the usage of the deliverable repository.
23.7.10 Demo
Deliverable repository list: http://www.stellarnet.eu/repository/deliverable_repository_list/
Deliverable visualization: http://www.stellarnet.eu/repository/composerproxy/
23.7.11 Screenshot
Figure 27. Composer Proxy.
23.8 H) Mobility tracker
The mobility tracker serves as a starting point for the mobility programme planned in WP4. It consists of an entry form, a list of travels, and a map visualisation using google maps. The purpose of the mobility tracker is to collect information on research exchanges and travels of partners in order to inform the project evaluation. Next steps would include a management section that allows to attribute budget to the applications for travel in the mobility programme and a matchmaker application that matches applicants with institutions offering to host a visiting researcher.
By now, the mobility tracker supports four scenarios:
- Mobility tracking of meeting journeys.
- Mobility tracking of researchers and staff exchange.
- Mobility details of meetings.
- Mobility details of presentation.
23.8.1 Use-case
A STELLAR scientist registers a trip or exchange including details on the nature of the stay. A map visualisation shows where STELLAR partners have been going.
23.8.2 Appropriateness for a modern Science 2.0
Ultimately, the tracker should allow managing the integration instrument “mobility programme”.
23.8.3 Tool type
Awareness, (recommendation)
23.8.4 Development stage
Working application is in review. Next steps will follow.
23.8.5 Data
Trip information, details on the nature of the exchange.
23.8.6 APIs
Uses google maps.
23.8.7 Security
The aggregated map of all trips is openly available. Adding of journeys requires authentication.
23.8.8 Demo
Add a meeting: http://www.stellarnet.eu/repository/meetings_geotag_form/
Add a meeting trip: http://www.stellarnet.eu/repository/mobility_geotag_form/
Add research and staff exchange: http://www.stellarnet.eu/repository/exchange_geotag_form/
Add presentation: http://www.stellarnet.eu/repository/presentations_geotag_form/
Overview of the stored mobility data:
http://www.stellarnet.eu/repository/mobility_geotag_map/
http://www.stellarnet.eu/repository/mobility_geotag_list/
23.8.9 Openness
To be discussed.
23.8.10 Screenshots
Figure 28. Mobility tracker.
23.9 I) Podcasting Tool
The podcasting too records and disseminates podcast sessions from senior academics, visionaries and industry leaders. It aims to help TEL stakeholders with their capacity building according TEL. It consists of a podcast submission system and a presentation interface.
23.9.1 Type
Communication, awareness.
23.9.2 Use-case
The pods of the podcasting tools could be used as input for the use cases “publication feedback”, “More! About the speaker”, “automated update of the open archive”, and “TEL community news”.
23.9.3 Appropriateness for a modern Science 2.0
Podcasts are made available through web syndication and allow conveniently downloading new episodes, which could be fetched by podcatchers. Compared to other ways of downloading media files, this allows to automatically identifying new released pods, which could be stored on the local computer or mobile application for later replay.
23.9.4 Development stage
Released.
23.9.5 Security
Submitting of podcasts requires authentication. No authentication required to listen to the podcasts.
23.9.6 Data
- Input: Forms used to gather metadata of the podcasts.
- Output: Podcast feed.
23.9.7 API definition
n/a
23.9.8 Openness
Podcasts will be both open available and after registration.
23.9.9 Evaluation
web statistics are used to track the usage of the podcast tool.
23.9.10 Demo
Podcast: http://www.stellarnet.eu/index.php/podcast/
23.9.11 Screenshot
| 157px | 377px |
Figure 29. Podcast Presentation Interfaces.
23.10 J) Conference Browser
The Conference Browser allows the attendees to explore the conference. Users can explore the conference in different dimensions, for example time (different years of the conference), location (location of the institutes of the authors), authorship, co-authorship, citations, etc.
23.10.1 Use-case
A conference attendee who wants to learn more about the conference proceedings or the conference and its participants itself. For example, a scientist wants to discover relations between researchers and countries. He goes to the application and starts searching for an author. The visualization then shows the network of co-authors and additional information – researchers are able to explore the conference data in a visual way.
23.10.2 Appropriateness for a modern Science 2.0
This tool could allow you to find interesting new peers to contact at the conference and could place the user's research in the bigger picture of the conference.
23.10.3 Tool type
Recommendation, visualisation, data browser.
23.10.4 Development stage
Under development.
23.10.5 Security concept
No critical data.
23.10.6 Data
The data is provided by the building block “conference paper database” and “paper information retrieval service”. In the future, the geo-locations of authors and institutes will be provided by the building block “geo-locator for persons and institutes”. Other data sources to enrich the paper database will be SIGMA[#_ftn34 [34]], linkedIn, etc.
23.10.7 APIs
This application will not provide any API's for external developers.
23.10.8 Demo
This includes new development, but a related version can be seen here: http://ariadne.cs.kuleuven.be/ectel/
http://ariadne.cs.kuleuven.be/edmedia/
23.10.9 Openness
Source code will be made publicly available.
23.10.10 Screenshots
Figure 30. ECTEL single author network.
Conference data provide information about relations of authors and coauthors, allow insight about participation of countries, and reveal statistics about the number of publications of authors, and countries (Ochoa, Méndez, and Duval 2009). The knowledge networks visualization gives an aggregated view to the underlying conference data.
Figure 31. ED-Media single author network.
Below you can find additional mock-ups of the new tool. A map and an information panel occupy the major part of the screen. The map shows the location of the topic in the information box and possible locations of relations with the topic (e.g. author & co-authors). At the top left a search bar is located, to search for authors and papers. At the bottom two timelines of all papers that were found is shown and the references of these papers.
When the user clicks on some information in the information panel, the map or the timeline the view of the information panel can change to for example another author or information about the selected paper.
Figure 32. Conference browser.
23.10.11 Evaluation
- conversion and abandonment, traffic sources, referring websites, inbound links from social networks, loyalty
- Depending on the number of users it can be worthwhile to extend the evaluation to monitoring online buzz. This has to be decided.
- A usability study will also be done to detect major flaws in the user interface design.
23.11 K) Flashmeeting recommenders
This science proxy project is targeted towards increasing awareness and resolving unwanted fragmentation with the help of recommendations extracted from a set of data sources (ranging from publishing data from conferences such as EC-TEL to online communication tools such as Flashmeeting) and – on the intervention side – with the help of the networked communication instrument ‘Flashmeeting’ . This recommender system aims at supporting scientists in the field of TEL to increase the quality of their collaboration in the wider community of peers.
Scientific community structure is revealed in a number of interaction channels (events, authorship of papers, on-line meetings, etc). Data from different sources has to be integrated and combined in order to grasp an accurate impression of the actual state of the art. The basic idea of this set of aligned recommender tools draws from the possibility to build links from comparing relationship information in independent data sets, combining paper authorship and virtual meeting interactions. The recommender system is complemented with an interaction tool, Flashmeeting, which allows deploying the suggested recommendations into actual, successful collaborations.
Flashmeeting[#_ftn35 [35]] is an academic research project aimed at understanding the nature of online events and helping users to meet and work more effectively. FM has been used over four years by a great variety of communities of learners and knowledge workers across the globe, supporting online communities and creating new reusable learning objects.
23.11.1 Use-case
Recommenders made possible with this approach range from community forming to group activity support. They include but are not limited to:
- Defrag meeting recommender: from the co-authorship network and the co-citations therein, a recommender can identify when authors are working on the same topic (=keywords) but with different co-authors and different literature. This can be a strong indicator of unwanted fragmentation. A recommender can be trained that proposes to hold a 'getting to know each other' Flashmeeting that may initiate desired defragmentation.
- Group proposal recommendation: existing cliques can be discovered from graph components, recommending their members to form a group for supporting the management of joint meetings.
- Group closing recommendation: lack of activity in a group may indicate that it no longer exists as such. Confirming group disappearance would be necessary for keeping the server tidy and an accurate map of existing active communities.
- Group access recommender: when raising awareness about existing groups for a given individual, the participation of his/her contacts in a certain group is a strong indicator about the interest of the group for such a person. Recommendations for joining a given group based on contacts’ membership can help to avoid missing information.
- Meeting invitation recommender: awareness of community specific events can also be improved. Based on the known participants in the event as well as their contact relations, recommendations can be made for potential attendants.
- Defrag group recommender: communities are far from homogeneous. Sub-groups can emerge, particularly in big communities, which are connected by a small set (two or three) of members acting as bridge builders between otherwise disconnected components in the interaction graph. Alerts about such structural dysfunctions including the provision of solutions such as joint virtual meetings can help to mend them and improve effective collaboration inside the global community.
23.11.2 Appropriateness for a modern Science 2.0
Through the use of such a recommender system, unwanted fragmentation can be spotted more easily (and eventually resolved). Tool type
Awareness, (recommendation)
23.11.3 Development stage
Early prototype (algorithms work)
23.11.4 Data
EC-TEL conference meta-data, flashmeeting data
23.11.5 APIs
n/a
23.11.6 Security
Security model not yet defined.
23.11.7 Openness
To be discussed.
23.11.8 Screenshots
Figure 33. Unwanted fragmentation spotted in the EC-TEL conference meta-data.
When analyzing the ECTEL co-authorship and co-citation graphs with the defrag meeting recommender algorithm, it becomes visible that Mohamed Amine Chatti, a researcher from RWTH Aachen University, Germany is an isolated member connected only to his 2 co-authors. A textual analysis of the content of his paper reveals that he is working on the automated annotation of learning materials. If that topic is searched inside the ECTEL collection of papers, it appears that there is a strongly related group of researchers working in the same field (centred on Alexandra Cristea). The location of Mohamed and Alexandra group could be seen in Error! Reference source not found.. The recommender could suggest Alexandra and Mohamed to meet.
24 Annex C: Widgets for Wookie & Elgg
24.1 Installation
First you have to set up the Wookie engine. You will find a step by step installation here: http://incubator.apache.org/wookie/downloading-and-installing-wookie.html.
The same applies for Elgg. Go to http://elgg.org/downloads.php and download the latest release. Find the installation instructions here: http://docs.elgg.org/wiki/Installation.
Elgg can use Wookie widgets through the Elgg plugin system. First install the Wookie Widget Plugin (http://community.elgg.org/pg/plugins/hoisl/read/323321/wookie-widgets-21). Rename the base folder “wookie” to a name reflecting your plugin. You can modify the languages/en.php the array values, to brand your widget and add further languages files for internationalisation (see http://docs.elgg.org/wiki/Engine/Internationalisation#Adding_language_files_to_plugins_.28and_best_practices.29).
For general plugin configuration information, see http://docs.elgg.org/wiki/Configuration/Plugins.
Set at the beginning of the view.php in the view folders the URL to the Wookie engine, your API key, and the type of widget (see the following section).
For the development process of widgets within Elgg turn off the simple cache and the view filepath cache, to avoid that changes become not instantly visible. Set the options in the Adminstration menu under the “Site Administration” submenu.
To enable the Wookie plugin, go to the Administration page to the submenu “Tool Administration” and enable your plugin. The name of the plugin is the name of your plugin folder, e.g. “wookie”. Further descriptions of the plugin can be set in the manifest.xml of your plugin.
24.2 Wookie
24.2.1 Wookie Engine
The Wookie Server allows delivering widgets, which follow the W3C widget specification. The Wookie server administration interface provides functionalities to add widgets and to remove them, to set a white list for accessing data (both in a restricted are), to request an API for the widget, and to instantiate a widget.
24.2.2 Getting an API key
Through the public administration interface of the Wookie server everyone can request an API key entering an email address (the latest version of Wookie seems to have a bug. It says that there is a problem to create the API key. Nevertheless you can find the API key in the database).
24.2.3 Accessing data
White lists for remote content could be set in the administration interface of Wookie. If the widget wants to access for example a data feed from a remote server, the white list has to be set in the Wookie engine. This is due to the restriction of many browsers to allow communication only from the server from where the widget is coming from (same server restriction).
An example for a white list entry is: http://news.kmi.open.ac.uk/rostra/rss2.php?r=11
24.2.4 Instantiate a Widget
You don’t have to worry about how to instantiate your widget. The Elgg Wookie plugin described earlier sends a proper request to the Wookie engine to instantiate the widget. Nevertheless the details for the call to the Wookie engine and the response could be useful to extend the Elgg Wookie plugin to your needs.
The API call “getwidget” instantiates a new widget or returns the previously used instance. The administration interface helps to generate an example call. Five parameters could be specified:
?requestid=getwidget
&userid={userid} (unique string for a user)
&shareddatakey={ shared data key} (any string)
&api_key={API key} (string retrieved from the administration interface)
&servicetype={service type} (string, e.g. ‘chat’, ‘forum’ or ‘vote’)
24.2.5 Example:
The shareddatakey parameter allows users to share the same instance of the widget. If another user wants to share the same instance of the widget the call has to include the same shareddatakey string of the first user.
24.2.6 Example:
24.2.7 Return values
The call returns and XML fragment with information how to display the widget:
<url> url of the widget instance
<height> height of widget
<width> width of widget
<maximize> {true, false} to flag if the widget can be maximized to full screen or if it should be fixed to the default height & width in the client application.
24.2.8 Example
<widgetdata>
<url>
</url>
<title>Default chat widget</title>
<height>358</height>
<width>500</width>
<maximize>false</maximize>
</widgetdata>
This information could be used for the <iFrame> of the target container (e.g. Elgg). Wookie loads the start.html of the widget package into this iFrame.
For more information, look into the Wookie Widget Server API specification: http://tencompetence.cvs.sourceforge.net/viewvc/*checkout*/tencompetence/wp6/org.tencompetence.widgetservice/docs/widget_web_api_09.doc.
A Wookie REST API draft is listed here: http://cwiki.apache.org/WOOKIE/wookie-rest-api.html.
24.2.9 Creating widgets
The Wookie Widget Developer’s Guide is a comprehensive introduction of the Wookie widget development process.
See http://getwookie.org/Widgets_files/widget_dev_guide_1.pdf (for Wookie engine 0.8) and
http://tencompetence.cvs.sourceforge.net/viewvc/*checkout*/tencompetence/wp6/org.tencompetence.widgetservice/docs/widget_dev_guide_09.doc (for Wookie engine 0.9).
The documentation of Wookie is work in progress. New documentation is in development. For the actual state of these look here: http://incubator.apache.org/wookie/.
25 Annex D: Workshop ‘Science 2.0 for TEL’
The workshop took place on Tuesday, September 29th, 2009, from 9:00 to 13:00 in the Novotel Hotel, 8/10 Parvis de l'Europe, 06300 Nice, France. It was a workshop at the ECTEL conference, the fourth European Conference on Technology Enhanced Learning ‘Learning in the Synergy of Multiple Disciplines’.
25.1 Proceedings
The following accepted full papers are available from the STELLAR open archive:
- Science 2.0: Supporting a Doctoral Community of Practice in Technology Enhanced Learning using Social Software (Gillet D., El Helou S., Joubert M., Sutherland R.)
http://oa.stellarnet.eu/open-archive/browse?resource=2196_v1
- Social networking in scientific conferences - Twitter as tool for strengthen a scientific community (Ebner M., Reinhardt W.)
http://oa.stellarnet.eu/open-archive/browse?resource=2197_v1
- Reflections on Participatory Science for TELSci2.0 (Underwood J., Luckin R., Smith H., Walker K., Rowland D., Fitzpatrick G., Good J., Benford S.)
http://oa.stellarnet.eu/open-archive/browse?resource=2198_v1
- Visualizing Research Patterns in the Field of e-Learning (Salman Khan M., Ebner M., Taraghi B.)
http://oa.stellarnet.eu/open-archive/browse?resource=2199_v1
- Science 2.0 Practices in the Field of Technology Enhanced Learning (Kieslinger B., Lindstaedt S.)
http://oa.stellarnet.eu/open-archive/browse?resource=2200_v1
25.2 Programme
The programme included:
09h00-09h20 introductions, welcome, goal, logistics
09h20-09h45 Muhammad Salman Khan, Martin Ebner, Behnam Taraghi,
Visualizing Research Patterns in the Field of e- Learning
09h45-10h10 Denis Gillet, Sandy El Helou, Marie Joubert, Rosamund Sutherland, Science2.0: Supporting a Doctoral Community of Practice in Technology Enhanced Learning
10h10-10h35 Barbara Kieslinger, Stefanie N. Lindstaedt, Science 2.0
Practices in the Field of Technology Enhanced Learning
10h35-11h00 coffee
11h00-11h25 Jan Reichelt, Mendeley
11h25-11h50 Joshua Underwood, Rosemary Luckin, Hilary Smith, Kevin Walker, Duncan Rowland, Geraldine Fitzpatrick, Judith Good, Steve Benford, Reflections on Participatory Science for TELSci2.0
11h50-12h15 Martin Ebner and Wolfgang Reinhardt, Social networking in scientific conferences – Twitter as tool for strengthen a scientific community
12h15-12h30 Erik Duval, Xavier Ochoa, Sten Govaerts, A
visual exploration of ECTEL
12h30-13h00 discussion & wrap-up
25.3 Call for papers
1st Workshop on Web2.0 approaches, tools and technologies to support research on Technology Enhanced Learning (TELSci2.0) at the 4th European Conference on Technology Enhanced Learning (EC-TEL09), Nice, France, September 29 - October 2, 2009, http://stellarnet.eu/science2ectel/
25.3.1 SCOPE
In Technology Enhanced Learning (TEL), the use of web2.0 technologies is now actively being researched, under banners such as "Personal Learning Environments" or "Open Learning Environments" and the like. In this workshop, we want to discuss how we can leverage the same opportunities for our research on TEL.
Indeed, as researchers in Technology Enhanced Learning we already know how to include things like blogs, wikis and forums into the heart of our work to enhance collaborative working, but a full "Science 2.0" framework might provide us with a much more powerful framework to make our research more effective.
This workshop aims to bring together all those who want to turn a vision of the e-scholar and e-scientist of the 21st Century into reality in our own domain of TEL.
The core of significant development in a ‘2.0 world’ is
- leveraging web2.0 technologies (that rely on the ‘social graph’) for improving the effectiveness and efficiency of how we conduct science;
- the capacity to share and reuse data, and to benchmark scientific activity in the international community.
Key questions for this workshop include:
- How do we mesh social web2.0 technologies with the scientific workflow?
- How do we share the data of our activity wider in our community?
- How do we visualize that data and how do we provide researchers with powerful new frameworks for their scientific endeavour?
Early generic tools for science2.0 are beginning to appear:
- http://www.scivee.tv/,
- http://www.academiccommons.org/,
- http://www.academia.edu/,
- http://www.researchgate.net/,
- http://www.mendeley.com/,
- http://www.escidoc.org/,
- http://www.authormapper.com/,
- http://www.surfspace.nl/wiki/display/widgetswetenschappers/Home,
- http://www.plos.org/,
- http://www.openwetware.org/,
- http://www.galaxyzoo.org/,
- http://bosch.informatik.rwth-aachen.de:5080/AERCS/
- http://www.rkbexplorer.com/
- http://ariadne.cs.kuleuven.be/edmedia/
- etc.
However, most of these focus more on the sciences. There is some discussion in blogosphere about how we could make use of science2.0 opportunities for TEL. Yet, there has not been a dedicated workshop or conference on this theme. That is the missing piece that this workshop wants to fill in, in the hope of moving forward this idea and accelerating the evolution that has considerable promise of improving how we conduct research on TEL.
25.3.2 FORMAT
The workshop will be highly interactive: presentations will be clustered and presenters will not only discuss their own work, but also to explicitly comment on the work of the other presenters in their session. We will make all the papers available beforehand, so as to facilitate this.
Participants will also be asked to do a 5 min vlog posting to the workshop web site before the event. (This will be a requirement for registration!)
We will ask the editors of the IEEE Transactions on Learning Technologies to host a special issue of the re-worked papers after the workshop.
25.3.3 ORGANISATION
This half-day workshop is supported by the STELLAR EU Network of Excellence in Technology Enhanced Learning; http://stellarnet.eu
The workshop is organized by
- Erik Duval (K.U.Leuven, Belgium)
- Peter Scott (Open University, UK)
- Stefanie Lindstaedt (KnowCenter, Austria)
- Nicolas Balacheff (UJF, France)
25.3.4 SUBMISSIONS
Authors are invited to submit original unpublished research as papers (4-8 pages). Including demonstrations is explicitly encouraged. All submitted papers will be peer-reviewed by at least three members of the program committee for originality, significance, clarity, and quality.
All questions and submissions should be sent to: erik.duval@cs.kuleuven.be
25.3.5 IMPORTANT DATES
- Paper Submission: July 5, 2009
- Results Notification: July 20, 2009
- Camera Ready Submission: September 1, 2009
- Workshop Date: September 29 or 30, 2009
25.3.6 PROGRAMME COMMITTEE
- Martin Ebner (Technische Universitat Graz, Austria)
- Jean Marie Favre (Universite Joseph Fourier, France)
- Barbara Kieslinger (Zentrum fur Soziale Innovation, Austria)
- Chad J. Kainz (University of Chicago, USA) (tbc)
- Ralf Klamma (Rheinisch-Westfalische Technische Hochschule Aachen, Germany)
- Erica Melis (Deutsche Forschungszentrum fur Kunstliche Intelligenz, Germany)
- Xavier Ochoa (Escuela Superior Politecnica del Litoral, Ecuador)
- Andreas Schmidt (Forschungszentrum Informatik, Germany)
- George Siemens (University of Manitoba, Canada)
- Katrien Verbert (Katholieke Universiteit Leuven, Belgium)
- Martin Weller (The Open University, UK)
- Fridolin Wild (The Open University, UK)
- Martin Wolpers (Fraunhofer-Institut fur Angewandte Informationstechnik, Germany)
25.3.7 ABOUT EC-TEL09
After three successful EC-TEL conferences in 2006, 2007, and 2008, the Fourth European Conference on Technology Enhanced Learning provides a unique forum for all research related to TEL, among them education, psychology, and computer science. The contributions will cover the design of innovative environments, the implementation of new technological solutions, results of empirical studies on socio-cognitive processes in learning, and field studies regarding the use of technologies in context.
EC-TEL is a competitive and broad forum for TEL research in Europe and beyond. In its specialised accompanying workshops and the highlighting main conference programme, EC-TEL09 provides unique networking possibilities for participating researchers throughout the week and includes project meetings and discussions for ongoing and new research activities supported by the European Commission. See http://www.ectel09.org/ for details.
26 Annex E: Workshop ‘Mash-Up PLEs’
This is annex documents the programme, proceedings, and call for papers of the 2nd Workshop on Mash-UP Personal Learning Environments (MUPPLE-09) held at the 4th European Conference on Technology Enhanced Learning (EC-TEL09), Nice, France, September 29 - October 2, 2009.
26.1 Programme
9:00 introduction
Surprise welcome message by one of the organisers Marco Kalz, Daniel Mueller, Fridolin Wild, Matthias Palmer.
9:05 reports (20 min each)
Mohamed Amine Chatti, Matthias Jarke, Zhaohui Wang and Marcus Specht. SMashup Personal Learning Environments
Debating Space10
Behnam Taraghi, Martin Ebner and Sandra Schaffert. Personal Learning Environments for Higher Education: A Mashup Based Widget Concept
Debating Space10
Hendrik Drachsler, Dries Pecceu, Tanja Arts, Peter van Rosmalen, Hans Hummel, Rob Koper, Edwin Hutten and Lloyd Rutledge.
ReMashed – Recommendation Approaches for Mash- Up Personal Learning Environments in Formal and Informal Learning Settings
Debating Space10
10:30 coffee break
10:45 reports (20 min each)
Matthias Palmér, Stéphane Sire, Evgeny Bogdanov, Denis Gillet and Fridolin Wild.
Mapping Web Personal Learning Environments
Debating Space10
Scott Wilson, Paul Sharples, Dai Griffiths and Kris Popat.
Moodle Wave: Reinventing the VLE using Widget technologies
Debating Space10
Stéphane Sire, Matthias Palmér, Evgeny Bogdanov and Denis Gillet.
Towards Collaborative Portable Web Spaces
12:15 lunch break
13:15 reports (20 min each)
Dominique Verpoorten, Wim Westera and Marcus Specht.
Infusing reflective practice in eLearning courses – Can widgets help?
Debating Space10
Joanna Wild, Fridolin Wild, Marco Kalz, Marcus Specht and Margit Hofer.
The MUPPLE Competence Continuum''''
Debating Space10
Christian Glahn, Marcus Specht and Rob Koper.
Reflection support using multi-encoded Tag-clouds''''
Debating Space10
14:45 coffee break
15:00 stories (15 min each)
Hans Poldoja and Mart Laanpere. Conceptual Design of EduFeedr — an Educationally Enhanced Mash-up Tool for Agora Courses''''
Debating Space5
Vladimir Tomberg and Mart Laanpere. RDFa versus Microformats: Exploring the Potential for Semantic Interoperability of Mash-up Personal Learning Environments
Debating Space5
Roland Klemke and Birgit Schmitz. Interoperability issues for formal authoring processes, community efforts, and the creation of mashup PLE
Debating Space5
Zuzana Bizonova. Generalized Model for Interoperability of Data Based on Model Driven Architecture
Debating Space5
Effie Law, Daniel Müller and Anh Vu Nguyen-Ngoc.
Differentiating and Defining Portlets and Widgets: A Survey Approach
Debating Space5
17:00 open forum, resume, & agenda setting
26.2 Proceedings
The proceedings are available per open access here: http://CEUR-WS.org/Vol-506/
26.3 Call for papers
This is the original call for papers published to attract contributions.
| 131px 143px 96px 133px 119px 96px openscout-small.png |
2nd Workshop on Mash-UP Personal Learning Environments (MUPPLE-09) Interoperable Widgets, Services, and Microformats to facilitate Competence Development
held at the 4th European Conference on Technology Enhanced Learning (EC-TEL09), Nice, France, September 29 - October 2, 2009
RATIONALE
A change in perspective can be certified in the recent years to technology-enhanced learning research and development: More and more learning applications on the web are putting the learner centre stage, not the
organisation. They empower learners with capabilities to customize and even construct their own personal learning environments (PLEs).
These PLEs typically consist of distributed webapplications and services that support system-spanning collaborative and individual learning activities in formal as well as informal settings. These PLEs typically complement Learning Management Systems (LMS) with additional widgets, services, and data integrated from and with organization-external learning tools.
Consequently, the aim of this workshop is not to discuss the concepts ‘PLE vs. LMS’, but to focus more generally on how learning experiences can be enriched using mash-ups of widgets and services with microformats and how technology can help to respond automatically to competence level, need, or context. Moreover, the investigation of necessary competencies to deploy mash-up technologies, is dedicated special attention in this workshop.
Technologically speaking, this shift manifests in a learning web where information is distributed across sites and activities can easily encompass the use of a greater number of pages and services offered through web-based learning applications. Mash-ups, the ‘frankensteining’ of software artefacts and data, have emerged to be the software development approach for these long-tail and perpetual-beta niche markets. Core technologies facilitating this paradigm shift are Ajax, javascript-based widget-collections, and microformats that help to glue together public web APIs in individual applications. Interoperability is the enabler to allow these different components to be worked together facilitating the achievement of the underlying learning task.
In a wide range of European IST-funded research projects such as ROLE, iCoper, Stellar, LTfLL, Mature, Palette, OpenScout, and TENcompetence a rising passion for these technologies can be identified.
This workshop therefore serves as a forum to bring together researchers and developers from these projects and an open public that have an interest in understanding and engineering mash-up personal learning environments (MUPPLEs).
TOPICS OF INTEREST (but not limited to):
- Visions: Architectures, Frameworks, Strategies, Models
- (Collaborative) Authoring Tools
- Data Interoperability: with e.g. Microformats, streaming data, mixed media data
- User Interfaces: Concepts, Metaphors, Workflows
- Development Methodologies
- Innovative Widgets and Services: e.g. for instruction, game-based learning, self-reflection, personal information
- Interoperability Standards for widgets, content recombination, configuration
- User Studies & Evaluation Methods: evaluating e.g. performance, usability, specific design features, training methods
WORKSHOP FORMAT
The aim of this workshop is to bring together the various research and development groups in technology-enhanced learning that currently focus on the development of the next generation learning environments – learning environments that put the individuum centre stage and empower learners with design capabilities by deploying modern mash-up principles to establish system-spanning interoperability.
As this approach is rather young, the workshop seeks to attract both research results and work in progress in order to chart out the current state-of-the-art of MUPPLEs in TEL and to define main enablers and future challenges. Naturally, it will serve as a forum for establishing new collaborations.
Using the presentations as impulses and continuing post-talk debates, the workshop will conclude the day with an open discussion exchanging ideas, summing up, and defining a medium- to long-term research agenda.
SUBMISSIONS
Authors are invited to submit original unpublished research as full papers (8 pages), work-in-progress as short papers (max. 4 pages) or position statements (max. 2 pages). All submitted papers will be peer reviewed by at least three members of the program committee for originality, significance, clarity, and quality.
The workshop proceedings will be published online as part of the CEUR Workshop proceedings series. http://CEUR-WS.org is a recognized ISSN publication series, with ISSN 1613-0073.
Furthermore, the workshop serves as stage for presenting a snapshot of the work on contributions planned to be submitted to the upcoming special issue on mash-up personal learning environments in the International Journal of Technology Enhanced Learning (http://www.inderscience.com/browse/index.php?journalCODE =ijtel).
Authors should use the Springer LNCS format (http://www.springer.com/lncs). For camera-ready format instructions, please see “For Authors” instructions at http://www.springer.de/comp/lncs/authors.html.
All questions and submissions should be sent to: f.wild @ open.ac.uk
IMPORTANT DATES
- Paper Submission: July 31st, 2009
- Results Notification: August 15th, 2009
- Camera Ready Submission: September 15th, 2009
- Workshop Date: September 29th, 2009
PROGRAMME COMMITTEE
- Abelardo Pardo (University Carlos III de Madrid, Spain)
- Dai Griffith (University of Bolton, UK)
- Denis Gillet (EPFL, Switzerland)
- Effie Law (University of Leicester, United Kingdom)
- Felix Mödritscher (Vienna University of Economics and Business, Austria)
- Graham Atwell (Pontydysgu, United Kingdom)
- Gytis Cibulskis (Kaunas Technical University, Lithuania)
- Mart Laanpere (Tallinn University, Estonia)
- Martin Wolpers (Fraunhofer FIT, Germany)
- Mohamed Amine Chatti (RWTH Aachen, Germany)
- Nikos Karacapilidis (University of Patras, Greece)
- Scott Wilson (University of Bolton, United Kingdom)
- Stephane Sire (EPFL, Switzerland)
- Tony Hirst (Open University, UK)
- Bernd Simon (Knowledge Markets)
- Jad Najjar (Vienna University of Economics and Business)
- Kai Hoever (IMC AG)
- Zuzana Bizonova (INT Paris)
- Jan M. Pawlowski (University of Jyväskylä)
ORGANISERS
- Marco Kalz (Open University, The Netherlands)
- Daniel Müller (IMC AG, Germany)
- Matthias Palmer (University of Upsala, Sweden)
- Fridolin Wild (The Open University, UK)
ABOUT EC-TEL09
After three successful EC-TEL conferences in 2006, 2007, and 2008, the Fourth European Conference on Technology Enhanced Learning provides a unique forum for all research related to TEL, among them education, psychology, and computer science. The contributions will cover the design of innovative environments, the implementation of new technological solutions, results of empirical studies on socio-cognitive processes in learning, and field studies regarding the use of technologies in context.
EC-TEL is a competitive and broad forum for TEL research in Europe and beyond. In its specialised accompanying workshops and the highlighting main conference programme, EC-TEL09 provides unique networking possibilities for participating researchers throughout the week and includes project meetings and discussions for ongoing and new research activities supported by the European Commission. See http://www.ectel09.org/ for details.