Dear All

In our previous post we introduced the Argument Visualisation and Tracking tool (AVT) and gave an overview of its main design concepts. In this post we take a closer look at one scenario of using the AVT.
The rationale for the AVT is grounded firstly in current e-participation research priorities, chief among them the need for technology that improves public participation in a range of democratic processes. Thus, ultimately, the aim of the AVT is to improve the efficiency, inclusiveness, openness, and accountability of democratic processes. This means that, in addition to developing the AVT tool and exploring how best to improve the readability of very large visualisations of arguments, we will investigate the mediating role that such large, Web-based debate maps can play in e-participation scenarios. In particular, we intend to investigate the method and practice of how relevant e-participation actors use the AVT tool in the policy-making process.
Here, we will focus on an e-participation scenario from the perspective of a Policy Analyst, who for the purpose of discussion we call Bernd Gröninger. We ask that you consider yourself to be Bernd, where you are responsible for running a consultation on a green paper. You have to facilitate the debate, make sense of the responses, make these responses generally available so that relevant stakeholders can understand how the debate is progressing and at the end of the process you need to report back to the relevant government agency on the responses received.
(To ensure authenticity, the consultation concerns an actual Green Paper consultation, “Copyright in the EU Knowledge Economy” published by the European Commission. The IMPACT project is using this consultation as legacy data to trial the whole IMPACT toolbox. Due to various resource restrictions we are using a selected set of questions in the Green Paper and selected set of responses.)
Currently we do not have a fully functional prototype IMPACT toolbox and AVT tool that can be used to demonstrate all of our design choices. To overcome this difficulty and also to ensure we are progressing in the right direction, we have put together a detailed description of how we expect you the policy analyst to use the tool.

Configuring the debate

Bernd Gröninger logs into the IMPACT toolbox and clicks on the AVT tool icon to configure the tool for this particular consultation.

• He creates a new “Debate Map” in the AVT called “Copyright in the EU Knowledge Economy”.
• He enters the specific Green paper issues into the system. Importantly, these then frame the subsequent debate, bounding what counts as a relevant response from stakeholders. 
• He creates a Group called “Copyright in EU Knowledge Economy debate contributors” and then creates accounts on behalf of each contributor to the debate (e.g. “Aston University [created by Bernd Gröninger]” and “BBC [created by Bernd Gröninger]”).

(Bernd then uses the IMPACT Argument Reconstruction Tool to help him extract the arguments from each contribution to the policy consultation and these are automatically available for the AVT.)

Publishing the debate map

Bernd then publishes the argument maps in the AVT so that the stakeholders are able to browse the debate complete with the contributions made so far during the consultation process. Below is the first Treemap showing the different sections of the Green Paper. The size of the rectangle indicates the aggregated arguments under each section. Indeed, Bernd can check on how the consultation is progressing by going to the main debate window in the AVT to see the density of arguments for each section of the green paper.

Clicking on a rectangle shows another Treemap of the specific questions asked under that section. Clicking on specific questions then shows a network diagram of all the arguments addressing that particular question.

Reporting on the consultation

The deadline for the consultation is reached and Bernd needs to write a report, the AVT tool will help him with this task by providing summary statistics of the responses to the Green Paper issues made by different stakeholders. Statistics might include identification of which where the most controversial and hotly contested issues in the debate.

We welcome any feedback as it relates to the steps outlined in the above scenario from the policy analyst’s perspective.

Regards
Ann

Ann Macintosh
Professor of Digital Governance
Co-Director Centre for Digital Citizenship
Institute of Communications Studies
The University of Leeds,
Leeds, LS2 9JT, UK

http://ics.leeds.ac.uk/staff/a.macintosh

Dear All

Neil and I would welcome your comments on this, our first posting about the AVT tool.

This post provides an overview of the tool. Later this week we will send a second post which will describe how a Policy Analyst might use the tool.

There are pictures in this email so you may have to adjust your html setting in your email account in order to see them.

The Argument Visualisation and Tracking tool (AVT) is the IMPACT tool being developed at the Centre for Digital Citizenship, Institute of Communications Studies, University of Leeds. The AVT supports the work of relevant actors by enabling them to navigate through arguments contained in relevant consultation and policy documents. Specifically, the AVT displays arguments about policies as browsable debate maps. Users can browse these debate maps about public policies and follow links from the visual summaries of the arguments back to the original policy documents. Thus, the AVT is designed to help users make sense of the range of publicly expressed opinions about public policies. Indeed, it is part of the class of tools often referred to as "sense-making" tools.

To achieve this goal, the AVT is being based on the state-of-the art methods and tools in the field of computer-supported argument visualisation (CSAV). With this end in mind, the AVT is powered by the Cohere argument visualisation tool developed at the Open University  (Buckingham Shum, 2008).

One of the major challenges we face in using argument visualisation tools is their current poor usability when displaying large-scale argument maps.  In order to address this challenge we are basing the AVT tool around two main concepts. The first concerns document-centricity. The main IMPACT usage scenario involves an organization publishing a policy-consultation document in order to solicit feedback from relevant stakeholders.  Thus, the argument maps generated by the AVT are anchored in this policy-consultation document, and all arguments generated by stakeholders are entered into the argument maps with links to the original policy-consultation document. In this way all visualized data in the AVT tool will have a connection to the original consultation document.  This document-centricity is important since the policy-consultation document is central to our underlying objectives of achieving transparency and understanding in the argument map.  Furthermore, this document-centricity promotes sense-making for users joining at any time during a lengthy consultation period as they can see how their arguments fit within the ongoing policy-deliberation process consultation.  Finally, this document-centricity gives confidence to the policy-makers that the contributions provided by stakeholders are on-topic and relevant.

The second concept concerns visualising the potentially large amounts of information contributed during a consultation. In order to allow users to get an overview of the vast amount of information and to be able to appreciate which issues are causing the largest debate we are adapting a special kind of visualization called the “treemap”, which has been pioneered by Ben Shneiderman (Shneiderman, 1992) in the field of Information Visualization.  Adapting this technique, we will create “issue maps” which will use colour-coded rectangular blocks to depict issues within the debate. The different sizes of the rectangles indicate the comparative number of arguments associated with each issue. In this way, one of the first screens/views of a consultation a user will get will be a treemap of the questions asked within the Green Paper – with size of the each rectangle dependent on the responses at that time to the question.

Clicking on a rectangle takes the user to the arguments addressing that issue, where the arguments are displayed using the more conventional argument visualization technique of argument-network diagramming.

We welcome any feedback as it relates to these two concepts of “document-centricity” and “issue-mapping”.

Any comments or questions?

References

Buckingham Shum, S. (2008). Cohere: Towards Web 2.0 Argumentation. 2nd International Conference on Computational Models of Argument (COMMA '08) (pp. 97–108). IOS Press.

Shneiderman, B. (1992). Tree visualization with tree-maps: 2-d space-filling approach. ACM Transactions on Graphics , 11 (1), 92-99.

Regards

Ann

Ann Macintosh

Professor of Digital Governance,

Co-Director Centre for Digital Citizenship,

Institute of Communications Studies,

University of Leeds

Leeds, LS2 9JT, UK

Email: A.Macintosh@leeds.ac.uk  

Web: http://ics.leeds.ac.uk/staff/a.macintosh