Deliverable 1.2.2 First year study report Coordinator: Alexander Mikroyannidis With contributions from: John Domingue, Richard Power, Barry Norton, Maria Maleshkova Quality Assessor: Elena Simperl Editor: Alexander Mikroyannidis, The Open University Deliverable nature: Report (R) Dissemination level: (Confidentiality) Public (PU) Contractual delivery date: April 30, 2013 Actual delivery date: April 30, 2013 Version: 1.0 Total number of pages: 31 Keywords: Course production and delivery, training material, module, webinar, screencast, eBook, curriculum.
Transcript
Deliverable 1.2.2 First year study report
Coordinator: Alexander Mikroyannidis With contributions from: John Domingue, Richard Power,
Barry Norton, Maria Maleshkova Quality Assessor: Elena Simperl
Editor: Alexander Mikroyannidis, The Open University
Deliverable nature: Report (R)
Dissemination level: (Confidentiality)
Public (PU)
Contractual delivery date: April 30, 2013
Actual delivery date: April 30, 2013
Version: 1.0
Total number of pages: 31
Keywords: Course production and delivery, training material, module, webinar, screencast, eBook, curriculum.
EUCLID 296229
2
Executive summary This deliverable reports the working progress of WP1: “Course production and delivery”, complemented with the working progress in WP2: “Community building and outreach”. WP1 focuses on the production of educational and training content and its delivery over the most effective and relevant channels. WP2 ensures that the learning materials are presented, adjusted and improved through feedback and engagement with their intended audience, and the Linked Data community in general.
The production of EUCLID modules is based on the guidance of a curriculum plan, the first version of which has been finalised by the end of M3 and reported in D 1.1.1 “Curriculum plan”. The revised curriculum is composed of 6 modules to cover all the major aspects of the Linked Data consumption lifecycle organised in ascending learning levels, i.e. introductory, advanced and expertise.
The EUCLID modules consist of multi-format resources, i.e. presentations, webinars, screencasts, eBook chapters, and online courses. Since the start of the project, the production cycle of the EUCLID modules has been implemented, evaluated, and refined. An earlier version of the module production cycle and its evolution throughout the first 6 months of the project was reported in D1.2.1 “Half year study report” (M6). The production cycle has now matured to a point that it will be followed for the production of all remaining EUCLID modules.
Section 1 of this report presents the EUCLID curriculum and the course production process, along with their evolution during the first year of the project. Section 2 describes the implementation of the production process, as well as the content being produced. We present in detail all different types of training materials produced by the project and how these materials are connected with each other to deliver a comprehensive and concise training programme to the Linked Data community. Finally, Section 3 presents the lessons learned from the design and implementation of the production process, as well as our best practices for the design and delivery of learning materials for Linked Data.
EUCLID 296229
3
Document Information IST Project Number
FP7 -‐ 296229 Acronym EUCLID
Full Title Educational curriculum for the usage of Linked Data Project URL http://www.euclid-‐project.eu/ Document URL EU Project Officer Martina Eydner Deliverable Number 1.2.2 Title First year study report Work Package Number 1 Title Course production and delivery Date of Delivery Contractual M12 Actual M12 Status version 1.0 final X Nature prototype □ report X dissemination □ Dissemination level public X consortium □ Authors (Partner) Alexander Mikroyannidis (OU), John Domingue (OU), Richard Power (OU), Barry Norton
Partner The Open University Phone +44 1908 652073 Abstract (for dissemination)
This deliverable reports on the production cycle of the EUCLID training materials, as this has been formulated in the first year of the project. The EUCLID training materials consist of a series of modules, each containing multi-‐format resources, such as presentations, webinars, screencasts, eBook chapters, and online courses. In this report, the course production process is presented in Section 1, along with its evolution during the first year of the project. In Section 2, we describe the implementation of the production process, as well as the content being produced. We present in detail all different types of training materials produced by the project and how these materials are connected with each other to deliver a comprehensive and concise training programme to the community. Finally, Section 3 presents the lessons learned from the design and implementation of the production process, as well as our best practices for the design and delivery of learning materials for Linked Data.
Keywords Course production and delivery, training material, module, webinar, screencast, eBook, curriculum.
Version Log Issue Date Rev. No. Author Change 28.03.2013 0.1 Alexander Mikroyannidis (OU) Initial deliverable draft 05.04.2013 0.2 Alexander Mikroyannidis (OU) Version for internal review 23.04.2013 0.3 Alexander Mikroyannidis (OU) Version aligned to the comments
from the internal review 30.04.2013 1.0 Alexander Mikroyannidis (OU) Final version
DOCUMENT INFORMATION .................................................................................................................................................. 3
TABLE OF CONTENTS ............................................................................................................................................................ 4
LIST OF FIGURES AND/OR LIST OF TABLES ............................................................................................................................ 5
1 INTRODUCTION ............................................................................................................................................................. 7 1.1 THE EUCLID CURRICULUM .................................................................................................................................................... 7 1.2 THE COURSE PRODUCTION PROCESS ........................................................................................................................................ 8 1.3 COMMUNITY INVOLVEMENT ................................................................................................................................................. 10
2 IMPLEMENTATION OF THE COURSE PRODUCTION PROCESS ........................................................................................ 11 2.1 PRESENTATION SLIDES ......................................................................................................................................................... 11 2.2 WEBINARS ........................................................................................................................................................................ 11
2.2.1 Module 1 Webinar: Introduction and Application Scenarios ................................................................................. 13 2.2.2 Module 2 Webinar: Querying Linked Data ............................................................................................................. 13 2.2.3 Module 3 Webinar: Production of Linked Data ...................................................................................................... 13
3 LESSONS LEARNED AND BEST PRACTICES ..................................................................................................................... 25 3.1 BEST PRACTICES FOR THE DESIGN OF LEARNING MATERIALS ......................................................................................................... 25 3.2 BEST PRACTICES FOR THE DELIVERY OF LEARNING MATERIALS ....................................................................................................... 26
ANNEX A THE EUCLID CURRICULUM ............................................................................................................................... 28
EUCLID 296229
5
List of figures and/or list of tables Figure 1: The initial EUCLID curriculum ............................................................................................................................................ 7 Figure 2: The revised EUCLID curriculum ......................................................................................................................................... 8 Figure 3: The initial course production process .............................................................................................................................. 9 Figure 4: The revised course production process .......................................................................................................................... 9 Figure 5: The finalised course production process ..................................................................................................................... 10 Figure 6: The EUCLID presentation pack ........................................................................................................................................ 11 Figure 7: Still from a EUCLID webinar .............................................................................................................................................. 12 Figure 8: Geo-‐coverage of the 3rd EUCLID webinar .................................................................................................................... 12 Figure 9: The introduction of Chapter 1 in HTML format, featuring a webinar clip about the development of a
music portal (http://www.euclid-‐project.eu/modules/chapter1) ............................................................................ 14 Figure 10: A quiz in HTML format, consisting of 10 multiple-‐choice questions, testing the knowledge learners
have acquired in Chapter 1 (http://www.euclid-‐project.eu/node/30/take) ....................................................... 15 Figure 11: Chapter 1 presented as an iBook on an iPad, featuring a screencast about the MusicBrainz web
site (http://www.euclid-‐project.eu/sites/default/files/resources/Chapter1.ibooks) .................................... 16 Figure 12: Chapter 1 presented in the ePUB format on an iPad (http://www.euclid-‐
project.eu/sites/default/files/resources/Chapter1.epub) ........................................................................................... 17 Figure 13: A SPARQL exercise in Course 1, executed on the dedicated EUCLID SPARQL endpoint
(http://euclid.sti2.org/Exercises/Exercise1) ..................................................................................................................... 19 Figure 14: Course 1 presented on an iPad via iTunes U (https://itunesu.itunes.apple.com/enroll/DRM-‐CVV-‐
7Z6) ........................................................................................................................................................................................................ 20 Figure 15: Course 1 presented as a SocialLearn pathway
OWL-S – OWL for Services/ OWL-based Web Service Ontology (formerly DAML-S)
RDF/S – Resource Description Framework / Schema
SPARQL – SPARQL Protocol and RDF Query Language
URI – Uniform Resource Identifier
URL – Uniform Resource Locator
WP – Work Package
XML - Extensible Markup Language
EUCLID 296229
7
1 Introduction EUCLID is developing a series of training materials, including presentations, webinars, screencasts, eBook chapters, and online courses for Linked Data practitioners. The structure and the content of the training materials are guided by a curriculum plan. Initially, seven modules have been planned in the curriculum plan, as reported in D1.1.1 “Curriculum plan” (M3), to cover all the major aspects of the Linked Data consumption lifecycle. The primary target audiences of these training materials are data practitioners who, as part of their daily jobs, are interested in using Linked Data technologies for facilitating integration and easy access, technology enthusiasts who plan to broader their expertise in using Linked Data technologies for Big Data management and analytics, as well as Computer Science researchers who intend to gain a basic understanding of proven and tested Linked Data usage concepts.
Additionally, EUCLID has a strong focus on the community and encourages community engagement in the course production process through, for example, collecting user feedback via Twitter, webinars, and more. It combines online and real-world presence, and attempts to integrate with on-going activities in each sphere such as mailing lists, wikis. The project engages with the Linked Data community, developers and academics, through the community engagement activities described in WP2, which collects user requirements as well as provides feedback to the materials so that the course can be tailored to what the learner really needs and the course production process can be improved for effectiveness.
This deliverable is an update of D1.2.1 “Half year study report” (M6). In particular, it reports on the progress and lessons learned during the first year of the project regarding the production and delivery of the EUCLID training materials. The work carried out during this period includes the creation of the curriculum plan, and then, guided by the curriculum, the production of training materials in forms of presentations, webinars, screencasts, eBooks, as well as online courses.
1.1 The EUCLID Curriculum
The EUCLID curriculum has been designed to gradually build up trainee’s knowledge. It enables trainees with previous knowledge on a specific area of interest to only briefly go over the introductory materials and directly dig into one of the more advanced modules. The initial version of the curriculum was composed of seven modules to cover all the major aspects of the Linked Data consumption lifecycle, as reported in D1.1.1. Figure 1 shows the individual modules and their grouping into 3 levels of expertise (top: introductory, middle: advanced, bottom: expertise).
Figure 1: The initial EUCLID curriculum
In an effort to provide high-quality training, suitable for the data practitioner’s needs, the EUCLID curriculum has been through several revisions on structure, arrangement and content after presenting it to a number of experts and gathering their feedback. Details about these revisions are provided in D1.2.1. As a result of these revisions, the curriculum was refined and developed in more detail in order to include a number of expected outcome competencies, as well as more exercises and examples. These revisions also changed the structure of the curriculum
EUCLID 296229
8
to contain six modules. The current version of the curriculum plan is based on a series of modules, each of which is targeting a different crucial task related to Linked Data. The content of the updated modules has been redesigned to be better aligned and support a smoother process of skills built-up and development. While having an individual objective, each module contributes to further developing the skills and knowledge gained by the previous one thus aiding to acquiring an overall understanding and expertise in the field.
The revised curriculum is shown in Figure 2, organized in the order of 3 expertise levels (top: introductory, middle: advanced, bottom: expertise).
Figure 2: The revised EUCLID curriculum
The curriculum is constantly updated based on feedback from the community. Its latest version is available on the EUCLID web site1 and is also listed in Annex A of this deliverable.
1.2 The Course Production Process
The course production process defines the sequence of steps for the production of the EUCLID training materials. In the DoW, 3 basic steps were planned in order to create each module and its exercises (see Figure 3). Firstly, following the curriculum, the draft of the training material would be created, which includes slides for a webinar, as well as HTML content for online distribution. Secondly, feedback on the drafts would be gathered and analysed. Finally, based on the comments and feedback, each module would be refined before delivering an eBook encompassing all the training materials, which include written documents, examples, presentation slides, as well as the video recording of the webinar.
During the production of the first EUCLID module in the first six months of the project, this process was refined and elaborated to include some intermediate steps (see Figure 4). As documented in D1.2.1, the training materials in the form of slides and HTML content were revised through collecting comments and feedback from the broadcasting of the first webinar and the publication of the first version of HTML content. The webinar was recorded again after finalising the slides. Finally, the eBook chapter was produced from all the finalised content.
Figure 4: The revised course production process
Based on the lessons learned from the production of the first EUCLID module, this production process was further revised. One thing that became obvious was the instrumental role of the preparation and delivery of the webinar in the production process. It was therefore decided that the first recoding of the webinar would precede the production of the initial version of the HTML content. It was also decided that additional material in the form of an online course would accompany the final eBook chapter and would be part of the training programme offered to the
EUCLID 296229
10
community. More information about the EUCLID online courses is given in Section 2.5 of this deliverable. Figure 5 shows the finalised course production process according to the aforementioned revisions.
Ini$al'Version'of''eBook'Chapter'
(OU)'
Slides'–'First'Version'(KIT)'
Slides'–'Final'Version'(KIT'+'ONTO)'
Final'eBook'Chapter'and'online'course'
(OU)''
Webinar'–''Final'Recording'(ONTO+OU)'
Webinar'–''First'Recording'(ONTO+OU)'
Collec$on'of'Raw'Chapter'Materials'
(KIT)'
PreIfinal'Version'of''eBook'Chapter'
(OU)'
Figure 5: The finalised course production process
1.3 Community Involvement
Centred around the learning materials produced by the project, EUCLID is creating a learning community in order to acquire feedback on the learning materials. In this way, comments and feedback can be collected from potential trainees even before the design of the curriculum plan and can be consolidated in the form of requirements for updates to the learning materials, once the course production process has started. In order to have a better understanding of potential trainees’ learning focuses, needs and requirements, various social media channels are actively monitored, including mailing lists, discussion forums, as well as relevant blogs. Additionally, EUCLID maintains dedicated channels in Twitter, LinkedIn, Vimeo and SlideShare. More details about these activities and the current progress in engaging the community online and in the real world are provided in D2.1.4 “Interim online community engagement report” and D2.3.2 “Interim real-world community engagement report” respectively, both due on M12.
EUCLID 296229
11
2 Implementation of the Course Production Process Based on the curriculum and following the course production process, the EUCLID training materials are produced in various forms, in accordance with the targeted means of delivery. All EUCLID training materials are made available under a Creative Commons Attribution 3.0 Unported License2. This means that they can be shared, remixed, as well as used for commercial purposes.
The following sections provide more details about the various forms of EUCLID training materials, their scope, and how they are being produced.
2.1 Presentation Slides
The EUCLID presentations slides are the first training materials produced for each module. They are based on the collection of “raw” study materials (see Figure 5) and provide an overview of the main concepts covered in each module. They also contain an extensive set of examples, so that the concepts of the module are explained to practitioners more effectively.
The EUCLID project has created a SlideShare channel3 where the slides for each module are uploaded and made available to the community for feedback via comments. The presentations are also available in the EUCLID web site. Additionally, the EUCLID SlideShare channel contains the presentations delivered in the various EUCLID training events. Figure 6 shows the EUCLID presentation pack in SlideShare.
Figure 6: The EUCLID presentation pack
2.2 Webinars
The webinars are conducted based on the slides for each module. The webinars are broadcasted live from the Podium facility of the Open University4, after which a recording is made available both from the same facility and also through the EUCLID channel in Vimeo5. Figure 7 shows a still from one of the recorded webinars, where Dr Barry Norton is talking about the integration of Linked Data.
At the time of writing this deliverable, the public webinars for modules 1, 2 and 3 have been broadcasted and recorded. The live broadcast of the first 3 webinars has been watched by an average of 90 participants per webinar. Figure 8 shows a map with the locations of the attendees of the 3rd live webinar.
Figure 8: Geo-‐coverage of the 3rd EUCLID webinar
The lessons learned so far from the delivery of this series of webinars, as well as the feedback received by stakeholders are documented in D2.2.2 “Interim webinar report” (M12). The structure and content of each webinar are briefly outlined in the following sections.
EUCLID 296229
13
2.2.1 Module 1 Webinar: Introduction and Application Scenarios
This module aims to provide a general overview of the main topics related to using Linked Data. It is only an introduction and some of the topics are only mentioned and then discussed in greater detail in one of the following modules. The main goal of this module is to describe the overall motivating scenario and to teach the fundamental Linked Data principles, while briefly describing the context of the technologies and possible application solutions.
Part I of this webinar is available at https://vimeo.com/61612182 and Part II at https://vimeo.com/61612378.
2.2.2 Module 2 Webinar: Querying Linked Data
This module provides further details on making queries over Linked Data with SPARQL, and also covers the update language and protocol. Insight is given into the interaction of inference and querying and on the formation of a query plan via the SPARQL algebra.
Part I of this webinar is available at https://vimeo.com/61618438 and Part II at https://vimeo.com/61618437.
2.2.3 Module 3 Webinar: Production of Linked Data
The third EUCLID webinar will cover the whole spectrum of Linked Data production and exposure. After a grounding in the Linked Data principles and best practices, with special emphasis on the VoID vocabulary, we will cover R2RML, operating on relational databases, Open Refine, operating on spreadsheets, and GATECloud, operating on natural language. Finally we will cover means to increase interlinkage between datasets, especially the use of tools like Silk.
Part I of this webinar is available at https://vimeo.com/64709409 and Part II at https://vimeo.com/64709408.
2.3 Screencasts
The EUCLID screencasts consist of short clips (2-3 minutes) that provide a quick overview and a walkthrough of a representative set of tools and platforms related with the EUCLID modules. The screencasts are made available in the EUCLID Vimeo channel and are also included in the eBook chapters in order to better explain the relevant sections of the chapters. Currently, screencasts have been produced for the following tools and platforms:
• Sig.ma (https://vimeo.com/57931687)
• Data.gov.uk (https://vimeo.com/57931688)
• BBC Music (https://vimeo.com/57931689)
• Seevl (https://vimeo.com/57931690)
• MusicBrainz (https://vimeo.com/57935375)
• Sesame (https://vimeo.com/61612180)
• OpenRefine (https://vimeo.com/62430786)
2.4 eBook
The EUCLID eBook encompasses all the content for each module in a structured and interactive way. The eBook serves as the basis for self-learning, as well as for revisiting certain topics after a training is completed, e.g. as part of a EUCLID training event. For each module, the feedback gathered after the delivery of the presentation slides and the webinar is used to restructure the module content for final delivery as an eBook chapter. The eBook, therefore, represents the final outcome of the training materials revising process.
Each eBook chapter contains several multimedia and interactive elements, such as clips from the webinars, screencasts, as well as quizzes based on multiple-choice questions. In order to maximise the impact of the training materials on the community and bring them closer to as many people as possible, the eBook chapters are made available for a variety of platforms and formats:
EUCLID 296229
14
• Web browsers (HTML format)
• Apple iPad (iBook format)
• eReaders (ePUB format)
• Amazon Kindle devices (AZW3 format)
At the time of writing this deliverable, Chapters 1 and 2 of the EUCLID eBook have been released in all the above formats. Chapter 3 is scheduled to be released in May of this year. The following figures showcase various sections from Chapter 1 of the eBook in different platforms and formats, featuring some multimedia and interactive elements.
Figure 9: The introduction of Chapter 1 in HTML format, featuring a webinar clip about the development of a music portal (http://www.euclid-‐project.eu/modules/chapter1)
EUCLID 296229
15
Figure 10: A quiz in HTML format, consisting of 10 multiple-‐choice questions, testing the knowledge learners have acquired in Chapter 1 (http://www.euclid-‐project.eu/node/30/take)
EUCLID 296229
16
Figure 11: Chapter 1 presented as an iBook on an iPad, featuring a screencast about the MusicBrainz web site (http://www.euclid-‐project.eu/sites/default/files/resources/Chapter1.ibooks)
EUCLID 296229
17
Figure 12: Chapter 1 presented in the ePUB format on an iPad (http://www.euclid-‐project.eu/sites/default/files/resources/Chapter1.epub)
2.4.1 eBook structure
The chapters of the EUCLID eBook have been structured to cover the following range of topics:
Chapter 1: Introduction and Application Scenarios
This chapter introduces the main principles of Linked Data, the underlying technologies and background standards. It provides basic knowledge for how data can be published over the Web, how it can be queried, and what are the possible use cases and benefits. As an example, we use the development of a music portal (based on the MusicBrainz dataset), which facilitates access to a wide range of information and multimedia resources relating to music. The chapter also includes some multiple-choice questions in the form of a quiz, screencasts of popular tools and embedded videos.
Chapter 2: Querying Linked Data
This chapter looks in detail at SPARQL (SPARQL Protocol and RDF Query Language) and introduces approaches for querying and updating semantic data. It covers the SPARQL algebra, the SPARQL protocol, and provides examples for reasoning over Linked Data. The chapter uses examples from the music domain, which can be directly tried out and ran over the MusicBrainz dataset. This includes gaining some familiarity with the RDFS and
EUCLID 296229
18
OWL languages, which allow developers to formulate generic and conceptual knowledge that can be exploited by automatic reasoning services in order to enhance the power of querying.
Chapter 3: Providing Linked Data
This chapter covers the whole spectrum of Linked Data production and exposure. After a grounding in the Linked Data principles and best practices, with special emphasis on the VoID vocabulary, we cover R2RML, operating on relational databases, Open Refine, operating on spreadsheets, and GATECloud, operating on natural language. Finally we describe the means to increase interlinkage between datasets, especially the use of tools like Silk.
Chapter 4: Interaction with Linked Data
This chapter focuses on providing means for exploring Linked Data. In particular, it gives an overview of current visualization tools and techniques, looking at semantic browsers and applications for presenting the data to the end used. We also describe existing search options, including faceted search, concept-based search and hybrid search, based on a mix of using semantic information and text processing. Finally, we conclude with approaches for Linked Data analysis, describing how available data can be synthesized and processed in order to draw conclusions. The chapter includes a number of practical examples with available tools as well as an extensive demo based on analysing, visualizing and searching data from the music domain.
Chapter 5: Creating Linked Data Applications
This chapter gives details on technologies and approaches towards exploiting Linked Data by building LD applications. In particular, it gives an overview of popular existing applications and introduces the main technologies that support implementation and development. Furthermore, it illustrates how data exposed through common Web APIs can be integrated with Linked Data in order to create mashups.
Chapter 6: Scaling up
This chapter addresses the main issues of Linked Data and scalability. In particular, it provides gives details on approaches and technologies for clustering, distributing, sharing, and caching data. Furthermore, it addresses the means for publishing data trough could deployment and the relationship between Big Data and Linked Data, exploring how some of the solutions can be transferred in the context of Linked Data.
2.5 Online Courses
The EUCLID online courses are learning pathways (or syllabi) that combine the EUCLID eBook chapters with further multi-format interactive learning materials. These materials include the full webinars, the presentation slides, as well as a set of exercises that allow learners to test their knowledge of Linked Data queries by using dedicated EUCLID SPARQL endpoints. Learners can study these courses at their own pace, as there is no predetermined start or end date. The online courses have a focus on learning outcomes, which drive the organisation of the content. This means that the ultimate purpose of the learning material is to help the learner achieve the specified learning outcomes.
The EUCLID courses are available for the following platforms:
• Web browsers (HTML format)
• iPad, iPhone and iPod touch (iTunes U)
The release of each course follows the release of an eBook chapter. At the time of writing this deliverable, the first 2 courses have been released in the above formats. Course 3 is scheduled to be released in May, together with Chapter 3 of the eBook. The following figures show sections of Course 1 in a web browser and in an iPad.
EUCLID 296229
19
Figure 13: A SPARQL exercise in Course 1, executed on the dedicated EUCLID SPARQL endpoint (http://euclid.sti2.org/Exercises/Exercise1)
EUCLID 296229
20
Figure 14: Course 1 presented on an iPad via iTunes U (https://itunesu.itunes.apple.com/enroll/DRM-‐CVV-‐7Z6)
In order to engage the learning community through an additional channel, the HTML version of the online courses is also made available via the SocialLearn platform6. SocialLearn provides a set of tools that are familiar from other social media platforms and that can be used to support learning conversations and interactions. These tools support a standard set of social network functions, such as user profile creation, personalised views of peer activity, following, friending, status updates, messaging, media sharing, tagging and group formation. Groups provide learners with the possibility of moving from an open learning space to a private area where necessary, for example for private mentoring, to carry on a conversation without interruption, or to share resources that are not freely available. The user profile and personal area can be used to supply personalized recommendations, tailored to the user’s interests and learning goals.
A EUCLID group7 has been setup in SocialLearn in order to host our online courses as SocialLearn learning pathways. Learners are able to engage with these pathways by joining the discussion associated with each step, suggesting changes, asking for or offering help, adding a rating or review or collaborating in a related group. Figure 15 shows a step of the learning pathway belonging to Course 1. 6 http://sociallearn.open.ac.uk 7 http://sociallearn.open.ac.uk/community/view/id/211177
EUCLID 296229
21
Figure 15: Course 1 presented as a SocialLearn pathway (http://sociallearn.open.ac.uk/path/view/id/211179/collectionId/211178)
EUCLID 296229
22
The learning pathway for the first course has been structured as follows. Subsequent courses for the rest of the EUCLID modules will follow a similar structure.
1. Learning Outcomes
By the end of this learning pathway you should have an understanding of:
• the enabling technologies for Linked Data, in particular the RDF data model
• the current status of the Linked Data technology
• how Linked Data can be applied in a particular scenario related to a music portal
2. Semantic Technologies and Linked Data Foundations
Learn about the foundations of Linked Data, in terms of the related technologies and possible application solutions.
The covered topics are:
• Motivation scenario
• Linked Data foundations (introduction to the underlying technologies - HTTP, URIs, XML, RDF, SPARQL)
o Evolution of the Web
o Web technology basics (HTTP, URIs)
o Describing and exchanging data (XML)
o Semantics on the Web
o Querying semantic data (SPARQL)
Watch the webinar 'Semantic Technologies and Linked Data Foundations' (73 minutes)
How much have you learned from this learning pathway? Test your knowledge by completing the following quiz and exercise.
Take the Quiz (10 multiple-choice questions).
Study the following RDF statements, expressed in the Turtle syntax, and then attempt the exercises that follow. For question 3 you should use the Euclid SPARQL endpoint. @base <http://www.euclid-project.eu/examples/module1#> .
<euclid> rdfs:label "The Euclid Project"@en, "Das Projekt Euclid"@de ;
vocab:consortiumMember <barry> .
1. Re-express the statements in NTriples (i.e. remove all prefixes and abbreviations to give full triples in absolute URIs).
2. Add a resource representing yourself, attaching your name using the FOAF properties.
3. Execute the following SPARQL query and consider why the class "Agent" has members, even though none are explicitly asserted in the data: SELECT ?agent WHERE {?agent a foaf:Agent}
4. Add a property to "consortiumMember" to assert that all subjects should be research projects (members of the "ResearchProject" class).
5. Adapt the query from (3) to ensure that "euclid" is now a research project
6. Create a new property to relate training participants to research projects and use it to relate yourself with "euclid".
Use the Exercise 1 form to try out your RDF and SPARQL.
5. Further Reading
If you are interested in more learning materials and resources about Linked Data, here are some suggestions that are relevant to this particular pathway:
• V. Bush and J. Wang (1945) "As we may think". Atlantic Monthly vol. 176, pp 101-108. Available on-line at http://dl.acm.org/citation.cfm?id=227186.
• T. Berners-Lee and R. Cailliau (1990) "WorldWideWeb: Proposal for a HyperText Project". Published on-line at http://www.w3.org/Proposal.
EUCLID 296229
24
• T. Berners-Lee, J. Hendler and O. Lassila (2001) "The Semantic Web". Scientific American vol. 284 number 5, pp 34-43. Available on-line at http://www.scientificamerican.com/article.cfm?id=the-semantic-web.
• T. Berners-Lee, R. Fielding and L. Masinter (2005) "Uniform Resource Identifier (URI): Generic Syntax". Published on-line at http://tools.ietf.org/html/rfc3986.
• O. Lassila and R. Swick (1999) "Resource Description Framework (RDF) Model and Syntax Specification". Published on-line at http://www.w3.org/TR/1999/REC-rdf-syntax-19990222/.
• E. Prud'hommeaux and A. Seaborne (2008) "SPARQL Query Language for RDF". Published on-line at http://www.w3.org/TR/rdf-sparql-query/.
EUCLID 296229
25
3 Lessons Learned and Best Practices The adopted production process involved a rigorous, iterated review procedure. This review procedure improved the produced training materials in two ways. First, it served the purpose of copyediting. These were corrections that in principle the author could have made, although they might be more apparent to another reader. Second, it revealed errors of content that the author would not have discovered alone.
Although some content errors were simply mistakes, most were subtler, arising from statements that were not exactly false but misleading, or oddly worded. This kind of error arises when the author has some familiarity with the topic, but lacks the deep understanding that comes from years of direct experience. Typical errors of this kind were as follows:
• Correctly describing a technology or standard, but explaining its significance in a misleading way (missing the point)
• Using terminology that is not exactly wrong, but non-standard, and hence confusing because the expert wonders why the usual word was not used
• Using examples that might work, but are implausible because an expert would have achieved the same result in a better way
Obviously, such errors can be detected only by a reviewer with deep knowledge of the area, and the review process has shown clearly the value of having such a review.
One issue that arises is whether it would be better to entrust the writing of chapters to someone who has deep knowledge of the area. There is probably a trade-off here between correctness and accessibility: it might be easier for someone like the author to address the points that perplex the non-expert reader.
The design and implementation of the production process has provided us with a valuable insight into the various challenges associated with the design and delivery of learning materials specifically for Linked Data. We have thus distilled our experiences and lessons learned into a set of best practices, which is outlined in the next 2 sections.
3.1 Best practices for the design of learning materials
1. Industrial Relevance – our curriculum takes into account the needs of industry related to Open Data and Linked Data. Future work aims to automatically mine and analyse relevant job adverts to gain desired competencies for the sector. This is supported by the 2nd best practice below.
2. Team Curriculum Design – where the team is composed of a number of roles to fully capture industrial, academic and pedagogical requirements. Our team comprises industry (Ontotext, FluidOps) who have extensive experience with professional training, industrial requirements and scalable tools, academia (KIT, STI International) who have research expertise in Linked Data and pedagogical experts (The Open University).
3. External Collaboration – to gain world-class curriculum expertise where necessary and to facilitate course delivery and dissemination.
4. Explicit learning goals – to which all content (slides, webinars, eBooks) are developed. Learners are guided through the learning goals by learning pathways – a sequence of learning resources to achieve a learning goal.
5. Show realistic solutions – rather than toy examples we utilize systems, which are deployed and used for real.
6. Use real data – we use a number of large datasets including for example, the MusicBrainz dataset, which contains 100Ms of triples.
7. Use real tools – our collection of tools are used for real including for example: Seevl, Sesame, Open Refine and GateCloud.
EUCLID 296229
26
8. Show scalable solutions – based upon industrial-strength repositories and automatic translations – for example using the W3C standard R2RML for generating RDF from large data contained in standard databases.
9. Eating our own dog food – we monitor communication and engagement with the Linked Data community through W3C email lists, in the social network channels LinkedIn and Twitter, as well as content dissemination channels such as Vimeo and SlideShare. We transform the monitoring results into RDF and make these available at a SPARQL endpoint. In this respect, we use Linked Data to support Learning Analytics.
3.2 Best practices for the delivery of learning materials
1. Open to Format – our content is available in a variety of formats including: HTML and as an eBook, as an Apple iBook and on Amazon Kindles.
2. Addressibility – every concept in our curriculum is URI-identified so that HTML and RDF(a) machine-readable content is available.
3. Integrated – to ease navigation for learners the main textual content, relevant webinar clips, screencasts and interactive components are placed into one coherent space.
4. High Quality – we have a formalised process where all materials go through several iterations to ensure quality. For example, for each chapter we run both a practice and a full webinars facilitating critique and commentary.
5. Self-testing and reflection – in every chapter we include inline quizzes formulated against learning goals enabling students to self-monitor their progress.
EUCLID 296229
27
4 Conclusions This deliverable reports on the progress during the first year of the project regarding the course production and delivery. The work carried out during this period includes the creation of the curriculum plan, and then, guided by the curriculum, the production of training materials in forms of presentations, webinars, screencasts, eBooks, as well as online courses. It is important to mention that all these materials went through at least one iteration of revision with the collection of comments and feedback after the delivery of the first versions of the produced materials. Based on our experiences and lessons learned in this first year of the project, we have also established a set of best practices for the design and delivery of learning materials specifically for Linked Data.
Most comments and feedback being received so far, are related to the quality of the training content, such as technical correctness and material ordering, which is no doubt of the uttermost importance to ensure delivering the correct knowledge in the most effective manner to learners. For example, the material should be organized in a way such that no too complex components should be introduced too early for learners at beginner level. Also, in terms of technical correctness, not all the proposed revisions are due to incorrectness, but rather to prevent potential misinterpretations or ambiguities among learners, which could hamper their learning effectiveness.
One of our main goals in the delivery of the EUCLID training materials is to reach out to the community in as many ways as possible, in order to engage it and acquire its feedback. For this purpose, considerable effort has been put to delivering the training materials in a variety of formats and for different purposes. The EUCLID training materials can be accessed from a wide range of platforms, both from desktop/laptop computers, as well as from different mobile devices. With the training materials reaching a constantly growing community, it is expected that there will be more comments and feedback received, which we will continuously monitor in order to improve the quality of the EUCLID training programme.
EUCLID 296229
28
Annex A The EUCLID Curriculum The goal of this curriculum is to offer learning support to data practitioners, covering different levels of knowledge and experience. In particular, the individual modules are structured in such a way that they can be used as building blocks for acquiring skills and knowledge but also be viewed individually, depending on the area of interest.
The latest version of the EUCLID curriculum is also available online at: http://www.euclid-project.eu/resources/curriculum.
