This preliminary report was published by the W3C SWIG Community.It is not a W3C Standard nor is it on the W3C Standards Track. Please note that under the W3C Community Contributor License Agreement (CLA) there is a limited opt-out and other conditions apply. Learn more aboutW3C Community and Business Groups
Preliminary Report of Community Activities 2013/2014
Publication Date 7th April 2014
© 2014 the Contributors to the Prelminary Report, published by the SWIG Community under. W3C Community Contributor License Agreement (CLA) A human-readable summary is available.
TABLE OF CONTENTS
To follow up on related activities and discussions of previous years, Paola Di Maio initiated this community beginngin 2013, and suggested that members join forces to organise various inter related activities, such as a workshop (or series of workshops) and publications. Roberto Garcis and Heiko Paulheim promptly answered the call, and a first workshop was organized in Madrid, with Roberto as the local lead (HSWI ‘ @WIMS 13) and a call for papers published for a special issue of the SWJ (status: pending). A call was made for co-chairs, which was answered by Martin Voigt, who was nominated co-chair in August 13. As participation grew to 35 members, a Community Charter was agreed upon establishing that members must fill out a stakeholder survey and make minimal documented contributions to any group activity of their choice.
2. Introduction and Goals
The end goal of the SWISIG is to produce a summary of good practices and guidelines to guide the development of user interfaces for semantic web technologies. Ideally, the group would produce a high level specification to help developers design and implement user interfaces to support and facilitate the use of semantic web technologies
As we begin thinking toward this goal, we are aware that members come from different backgrounds and have different motivations, with no single coherent point of view. The purpose of this preliminary report is therefore to
1. Summarize stakeholder analysis. Active members of this community are profiled, with a view to identify competences, requirements, and contributions each member of the community is likely to make to the group
2. Assess the feasibility of a high level specification (such as a functional specification) See an example here or whether we should aim to produce only broad guidelines, and either way, what process we should follow.
3. Assess the interest in carrying out Semantic Web usability evaluations. To be discussed.
4. Finally, to provide a summary of related activities being carried out by members, with an analysis of lessons learned and pointers to good practices, if any.
5. The report concludes with a statement of intent and a tentative plan as to whether a specification can be produced, and a timeline for the same, or whether the community should adjourn its agenda and goals for the next period of activity.
The community charter, as of Feb 2014, consists of two main rules, devised to ensure participation and involvement of members in governance, vision and mission of the group. Stakeholders input is taken by means of questionnaires.
A first questionnaire (June 2013) is designed to identify chairs, vicechairs and other community roles (two people responded in the first round, one of which has since left the group due to other commitments (Benjamin) and the other of which has been nominated co-chair in Aug 2013 (Martin). The questionnaire remains open for future candidate pitches. A second questionnaire is designed to allow stakeholders to shape the vision and mission for the group. Anonymized responses here (partial, to be completed)
At the request of chairs and community members, the charter and other governance tools, such as member surveys, can be amended, to make any governance suggestions, active members (members who have filled out the stakeholer survey) can post their suggestions to the list for discussion.Only the opinions of those who have filled out the stakeholder survey are taken into account to make governance decisions.
4. Stakeholder Analysis
A questionnaire has been circulated to community members to find out what they have in mind when joining this community, and to make sure the opportunity of shaping the vision and mission is documented. The questionnaire is attached in ANNEX 1. of this preliminary report.
Answers (anonymized) are documented in ANNEX 2 . Analysis of the stakeholder survey serves as input on governance, and as a guide to plan the future activities of the group
- Respondents plan to contribute to the community in a variety of ways
- 6 Respondents say they would like to start a high level spec for semantic web interfaces, a shared page has been started to start drafting
- 12 Respondents say they would like to start with broad Guidelines, a shared page is started here to start drafting
Nine members say they are available/interested in participating in usability evaluations (an additional nine say maybe). We should now establish in what capacity and what modality this can take place, Members who responded yes or maybe will be contacted via email. Any other suggestion to be discussed on list, please share your ideas.
5. Summary of activities
Activities can be events in the physical dimension, such as workshops and talks, but also articles, blog posts, random shared thoughts on the issues at hand (ie, any indication of brain activity), as well as discussions and visions of future activities one would like to see happen. Below, short descriptions with URL of member activities to date
HSWI Workshop c/o WIMS 2013 Madrid, Spain (Roberto Garcia)
Workshop/IFLA Satellite Meeting on User Interaction based on Linked Library Data (UILLD) in Singapore=August 2013. (Lars G. Svensson)
Keynote by Paola Di Maio (slides)
Papers and abstracts:
Semantically Guided, Situation-Aware Literature Research. T. Heuss, et al.
Using Linked Library Data in Working Research Notes. R. Shaw, et al.
Building interfaces on a networked graph. M. Malmsten and N. Lindström.
Library linked data and its relationship to knowledge organization systems. A. Slavic, et al.
Proceedings to be published by DeGruyter
- EWSC Workshop 2014 (Martin Voigt et al)
2nd International Workshop on Human Semantic Web Interaction (HSWI 2014)
in conjunction with the 11th Extended Semantic Web Conference (ESWC 2014)
26th May 2014, Crete, Greece
-Within the VOICES project, we have been looking at voice interfaces to in local languages for rural african farmers. One example is RadioMarche, a Market Information System allowing farmers to enter product offerings and local radio station hosts to access this data via the Web or through a phone interface.
- As a SW researcher, I am interested in innovative interfaces to Linked Data. Here is an abstract from a paper currently under review:
“Linked Data movement has facilitated efficient data sharing in many domains. However, people in rural developing areas are mostly left out. Lack of relevant content and suitable interfaces prohibit potential users in rural communities to produce and consume Linked Data. In this paper, we present a case study exposing locally produced market data as Linked Data, which shows that Linked Data can be meaningful in a rural, development context. We present a way of enriching the market data with voice labels, allowing for the development of applications that (re-)use the data in voice-based applications. Finally, we present a prototype demonstrator that provides access to this linked market data through a voice interface, accessible to first generation mobile phones.”
Semantic Data Publishing and Exploration Tool
The proliferation of Linked Open Data on the Web has increased the amount of data available for analysis and reuse. However, casual users find it difficult to explore and use Semantic Web Data due to the prevalence of specialised browsers that require complex queries to be formed and intimate knowledge on the structure of datasets. We address this problem in the Rhizomer tool by applying the data analysis mantra of overview, zoom and filter. These interaction patterns are implemented using information architecture components users are already familiar with but that are automatically generated from data and ontologies. This approach makes it possible to obtain an overview of the dataset being explored using techniques, such as navigation menus, treemaps or sitemaps, which are usually not available in text-based semantic web browsers. From there, users can interactively explore the data using facets. Moreover, facets also feature a pivoting operation, motivated during tests with lay users, that removes the main constraint of most faceted browsers, i.e. the inability to combine filters for differently faceted views to build complex queries.
It is possible to use the structure of semantic data to automatically generate user interfaces that help users, even semantic web experts, better understand the structure and the reuse options of a particular dataset
Puting user in the semantic web applications development loop helps a lot making the resulting tool more usable and even facilitates exploring alternative interaction patterns
Semantic Web Explorations Tools Quality in Use Model
In order to make Semantic Web tools more appealing to lay-users, a key factor is their Quality in Use, the quality of the user experience when interacting with them. To assess and motivate the improvement of the quality in use, it is necessary to have a quality model that guides its evaluation and facilitates comparability. SWET-QUM is based on the international standard ISO/IEC 25010:2011 and focuses on Semantic Web exploration tools, those that make it possible for lay-users to browse and visualise it. The model has been applied to compare the three main Semantic Web exploration tools that feature facets and the pivoting operation (Virtuoso FCT, SParallax and Rhizomer). The results of this analysis have guided the development of Rhizomer following a User-Centred Design approach, with user evaluations between development iterations.
In order to study and improve the user experience while interacting with semantic web tools it is necessary to have well defined ways to measure effectiveness, efficiency, user satisfaction,… but also to explore other ways of measuring the interaction that take into account the potential of semantic web, for instance flexibility to explore richly interconnected data.
InfoVis of SemWeb Data for end-users
Understanding and interpreting Semantic Web data is almost impossible for novices as skills in Semantic Web technologies are required. Thus, Information Visualization (InfoVis) of this data has become a key enabler to address this problem. However, convenient solutions are missing as existing tools either do not support Semantic Web data or require users to have programming and visualization skills. We propose a novel approach towards a generic InfoVis workbench called VizBoard, which enables users to visualize arbitrary Semantic Web data without expert skills in Semantic Web technologies, programming, and visualization. More precisely, we define a semantics-based, user-centered InfoVis workflow and present a corresponding workbench architecture based on the mashup paradigm, which actively supports novices in gaining insights from Semantic Web data, thus proving the practicability and validity of our approach.
VISO – Visualization Ontology (Jan, Martin)
The Visualisation Ontology (VISO) defines concepts and relations from the field of graphics and visualisation. The ontology is split into seven main modules, each being concerned with a specific field of visualisation. These modules (see figure below) include the graphic module, the data module and the facts module and other modules such as activities and system context modules. We tried to encourage and simplify the discussion VISO: At the bottom of each resource in the documentation, we provide links for discussing VISO terms in a forum. Further, we integrated quotations and references into the documentation in order to justify our modelling decisions and provide collaborators with the underlying literature.
Currently VISO is used and developed in the context of the projects VizBoard and RVL, which are described in this report as well. Since there is general interest on further collaboration on multiple modules of VISO, we are currently figuring out how this collaboration could happen and how and where to maintain the ontology in future. One option could be to open access and version the ontology on GitHub. More information on VISO and the current activities can be found on the VISO Blog.
RVL – RDFS/OWL Visualization Language (Jan)
RVL is a declarative language that allows to specify a mapping from RDF(S)/OWL concepts to visual means that are defined in the viso/graphic module of Visualisation Ontology VISO (see above).
Initial tooling for RVL, such as an OntoWiki-based editor, has already been developed in student projects at TU Dresden. We are currently working on both improving the RVL specification and building a prototypical interpreter for the language that allows for generating an abstract visual model from the mapping definitions and the data.
As RVL does not directly map to SVG, HTML or X3D, for example, a further transformation step is necessary. We are currently experimenting with a transformation of RVL definition to D3.js graphics. While the additional transformation step may be seen as a burden, it allows for reusing visual mappings on multiple platforms and could enable interoperability between (Semantic web) visualization tools. Future work will also cover the composition of mappings, which works for many cases already, but needs to be extended to arbitrary graphic relations.
Feedback on RVL is very much appreciated. As soon as the prototype reaches a certain degree of stability, we will be happy to open it for other developers. More information on RVL can be found at http://purl.org/rvl/ and in a paper on RVL.
Wikidata Reasonator (Andreas)
Visualization of Wikidata entities in different languages:
SWOWS – Semantic Web Open datatafloW System (Miguel)
SWOWS (swows.org) is a project that aims at making a platform for declarative and modular development of interactive applications consuming linked data. Applications are defined as pipelines of RDF transformations. The output are HTML + SVG views.
To deal declaratively with interaction:
the stream of user interface events and the user interface state are pipeline inputs;
there are stateful operators, incrementally updated through SPARQL Update queries.
A prototype, including a visual pipeline editor and a pipeline engine (both web based), is used to experiment the system “in house”. The code is public and the developer is in the process of preparing a public release aiming to wider experimentation and feedback.
Some general takeaways from the work carried on so far:
defining the interaction declaratively leads to more tricky code for simple cases but allows to more easily support multi-user/multi-device/multi-modal applications.
Publications (in progress)
SWJ Special Issue (Paola, Roberto, Heiko)
IFLA WORKSHOP PROCEEDINGS (Lars)
Proceedings to be published by DeGruyter
- Extract TAKEAWAYS (useful points, lessons learned etc) rom activities
- Invite group members to do a recorded webcast or organize a group call on a topic related to their interest
- Exploring Synergy with other W3C communities, for example the swsensors community (can we help the sensor web design interfaces for the semantic web), emergency information community (can we help design interfaces for emergency information) etc
- Create a Business Group to offer commercial consulting services on SW Interface Development (design, research, testing, evaluation etc)
- Possible Research – http://www.isprs.org/proceedings/xxxviii/part8/pdf/w08p02_20100216134415.pdf
6. SUMMARY AND ACTION POINTS
Over 50% of group members took the stakeholder surve (20 out of 35 members responded). We aim to bring stakeholder participation up to 100% as soon as the survey can be linked to the membership application. (request sent to W3C Ian on 27feb)
An amazing thanks to group members who have carried out related activities and documented them.
- Takeways/lessons learned
- Respondents should start working on their respective drafts
- A bit more drilling of members interested in doing usability evaluations (understand what roles they want to have)
Thanks to all Contributors who have shared in this report
ANNEX 1. STAKEHOLDER ANALYSIS -QUESTIONNAIRE
Live form (still accepting answers)
ANNEX 2 – STAKEHOLDER ANALYSIS RESPONSES AS OF 28 FEBRUARY 2014
BELOW, some SUMMARIES visualizations
<end preliminary report >