Graphically and spatially displayed text tool: Araucaria

Reed & Rowe (2004) explain that Araucaria has been developed with the learner, facilitator, and researcher in mind. It is easy enough to be useful to individuals learning how to restructure arguments, diagram them, and apply argumentation schemes. It is suitably flexible for facilitators to provide examples, sample analyses, as well as alternate sets of argumentation schemes. It is also powerful research tool, as it provides examples of argument analyses to support claims.

Araucaria is a software tool for analysing arguments. (Reed 2006 It allows an individual to graphically display an argument using a simple point-and-click interface. The software supports argumentation schemes, and provides a user-customisable set of schemes with which to analyse arguments.

Araucaria Main Window taken from Reed and Rowe (2006)

Constructing Standard Diagram

Araucaria is principally a tool for analysing existing arguments. (Reed & Rowe, 2006) Therefore individual begins with loading the text of an argument. Use the select-and-click technique to add nodes to the diagram so that all the propositions articulated in the text of an argument can be visualised in the diagram. To view parts of the diagram, select a node or arrow and click on it once. Upon selection of a node the text linked with the node will appear in the yellow message area at the bottom of the window. If an arrow between nodes is selected, the text corresponding to the nodes at either end of the arrow will be displayed.

To support node X with node Y, click on Y and hold the mouse button down while dragging the dotted line to X. Each node may support only one other node. Araucaria does not allow a single premise to be linked to multiple conclusions (known as divergent argumentation). If a link is removed, the diagram will rearrange itself automatically to accommodate the changes. Some arguments involve an assertion which supports a conclusion independently, known as convergent arguments. In contrast, other arguments involve premises which need to be linked together in supporting a conclusion and are known as linked arguments.

Some arguments called enthymemes involve ideas that are left implicit or tacit. Araucaria allows the individual to insert missing premises by creating of a node and connecting in the usual way. To delete a missing premise, select it and delete it as usual. The text of a missing premise node can be edited after it has been added to the diagram. Sometimes an analysis requires the notion of refutation which in Araucaria is close to the logical idea of negation. As with support, one node can have a maximum of one refutation. To build an analysis in which there are many ‘counterarguments’ or multiple rebuttals establish a refutation node and then drag supports to it.

Araucaria analysis can be visualised in three ways:

1.            Full Text - the full text of every node is displayed. Full Text mode is useful for understanding an argument; however it quickly fills the screen. To save space, one can collapse parts of the diagram. When a node has collapsed, the supporting argumentation is hidden beneath it. A small blue cross will be visible inside the node (in Full Text view) or just by the node (in the other views) to indicate hidden text. Double clicking of the node will re-expand the diagram. 

2.            Full Size - each node is represented by an ID label (e.g. A, B, C), but the image is guaranteed to remain clear with no overlapping components. In very large analyses the diagram will yet again not fit in the window.

3.            Scaled - each node is represented by an ID label (e.g. A, B, C), and the image is contracted to fit in the display window. Therefore, in very large analyses, some components may overlap a little.

Individual claims and the supports between them can be labelled within Araucaria to indicate:

•             Ownership to show which speaker is associated with them. If the text is a summary, argument for more than one point of view or it may be an argument which uses or attacks the arguments of others, Araucaria can show such arguments by marking each node as to who owns each claim after it has been linked to the diagram. Ownership can be attributed to nodes but not to arrows between nodes.

Ownership Window taken from Reed and Rowe (2006)

•             Evaluation - how the analyst has evaluated them. During the analysis of an argument individual may wish make judgements about the claims or their relationship; Araucaria can support this by marking evaluations on each node/arrow. These labels are user-defined to best meet own needs. Only a single evaluation can be given to a node in the diagram.

Evaluation Window taken from Reed and Rowe (2006)

Constructing a Toulmin Diagram

Creating nodes, selecting diagram components, and deleting diagram components works exactly the same as for Standard style diagrams as explained already. Connecting components are also mostly similar. Major difference is that Toulmin diagrams involve nodes with different roles: Datum, Claim, Warrant, Backing, Rebuttal, and Qualifier. These different roles are also recognised by different coloured boxes. In Full Size and Scaled views, they also are of different shapes. The different roles require slightly different rules for diagramming. For example, dragging the dotted line from one node to another will create a Toulmin argument with a Datum and a Claim. To add a Warrant, click a node and drag onto an arrow between an existing Datum and Claim. To add a Backing, click and drag onto an existing Warrant. To add a Qualifier, select the arrow between a Datum and a Claim, right-click, and from the popup menu, select ‘Add/Edit data qualifier’. This opens the same evaluation label window as in the Standard diagram. Select or type in an evaluation and then click OK. To link multiple arguments in the Toulmin style, simple drag and drop is needed. Every node in a Toulmin diagram (except Qualifiers) can act as a Claim in another argument. If you chain together many arguments, you may sometimes want to support a Warrant with a single Backing, and at other times, to support a Warrant with a whole other argument.

The facilitator will have to introduce the tool to the participants and allow them to familiarise themselves with it. Again the role of the facilitator appears to be minimal during the development of the map. The development of the argumentation map can be achieved collaboratively by the learners. Thus maybe here the facilitator is functioning as one of the community and contributes accordingly.

Graphically and spatially displayed text tool: Compendium

Compendium was developed by combining of an Issue-Based Information System (IBIS) and concept-mapping tool (Conklin and Burgess Yakemovic, 1991). This supported informal and exploratory conversation and facilitation, with a structured modelling approach (Selvin, 1999) with the underlying knowledge management aim. Conklin (2007) identifies this combination of a shared hypertext display, a trained facilitator, and a conversational grammar as Dialogue Mapping, a structural augmentation of group communication. As the conversation unfolds and the map grows, each participant can see a summary of the discussion thus far. The map serves as a "group memory," virtually removing the need for participants to repeat themselves in order to get their points across.

Other benefits of Dialogue Mapping include:

• Each participant's contribution is heard and acknowledged in the map.
• Each participant can see how their comments relate to others.
• The group sees where they are, where they've come from, and where they are going, and is thus self-correcting if they get "off-topic." (example Agenda map)
• The shared display map shifts the dynamic of the group into a collaborative mode ... "What can we think and learn together."
• The map focuses the group on a kind of "lightly logical" perspective as they work on the issues at hand.
• The map increases the group's shared understanding about the problem at hand, possible solutions, meaning issues, roles and responsibilities ... all of the key elements of a successful project.
• At the end of the meeting or during breaks, the group can view various printed snapshots of their discussion.
• Thanks to the IBIS grammar, the map summarizes the rationale behind any decisions that are made.
• Since the map captures the thinking process of the group, anyone who was not at the meeting can be quickly brought up to speed by reviewing the map with them.
• The map easily displays all of the open issues and action items at any point.

Benefits of Dialogue Mapping taken from Conklin 2007

‘Compendium is about sharing ideas, creating artefacts, making things together, and breaking down the boundaries between dialogue, artefact, knowledge, and data.’ (Selvin and Shum 2006, pp 6) They claim that this software provides a faster as well as a better way for groups and project teams to collaborate. Compendium offers a set of templates, methods, and tools that bring together people and their ideas. Compendium gives participants control over knowledge elements and allows the groups to use them in numerous ways such as, in meetings, in documents, working alone and working at a distance.

Compendium tool taken from Selvin & Shum (2002)

Ideas are expressed as icons or nodes which can be moved around, and connected to other ideas via links. These nodes are contained in Views. Views can also contain other Views. There are two kinds of Views:

•             a Map in which nodes can be placed anywhere in a 2D space, and

•             a List which organizes nodes in a sortable column.

There are different node types which can be used -

Question -         To ask a question or raise an issue related to a specific subject matter

Idea -                To provide a likely answer or alternative to a question

List View -       To generate a sortable list of nodes

 To place the results of a search, for example, to create a Catalogue of items

 To create a collection of nodes that do not need to be linked with each other

Map View -      To produce a representation of the relationships between ideas

 To group questions and ideas together in meaningful clusters

 To create associative links between nodes

Pro -                 To support an idea

Con -                To argue against an idea

Reference -       To bring in a link to an external file (e.g. a websites, spreadsheet, picture, or     document). A file can be dragged and dropped from anywhere in Windows into a Map  or List. A common file type is recognized by Compendium and labelled with the right icon automatically.

Note -               To provide extra or useful information about another node or the current view

Decision -         To resolve a question – link it to an Idea, or right-click on the Idea to change it directly to a Decision node

The links between the nodes can be of three types –

•             Associative links which connect ideas in the same context. Participants just draw the links between ideas and they are displayed as physically connected.

•             Transclusive links which connect the same idea in different contexts. Thus, transclusive relationships show the same idea in multiple contexts. For example, within a project, the same idea that originated in an email in context might resurface in another context such as an agenda item for a meeting, or as part of an analysis session, or as part of a formal requirements document. In each case, one idea is associated with different ideas, but the same idea appears in all the different cases. Compendium allows participants to track this by reusing the same node for a given idea in multiple maps/lists. Editing a node in one view (e.g. its label, detail, tags, or its icon) instantly updates in all other views. Transclusive hyperlinking gives you the means to track an idea in all its appearances.

•             Categorical links are connections between ideas because they are in the same category. Compendium, allows participants to classify an idea by giving it a tag / keyword. You can invent tags to mark anything that you may want to search for. Tags can be arbitrarily combined when searching and therefore provide yet another way to express connections between ideas, because they have something in common.

This seems to be a good option in the software as adding tags to nodes enables searches, cross-references, and the on-the-fly creation of specialized maps and lists without the manual work of assembling them node-by-node. To further aid in the coding process, Groups of tags can be created and maintained to support different tasks. New tags can be defined on-the-fly. The maps created using Compendium are complex and can be used to manage knowledge in projects both in business and in education.

Navigating in Compendium in a similar to navigating the Web, so having good view labels is akin to having good titles for web pages. View labelling is particularly essential, since nodes can be located in multiple views. When displaying the list of those views, it actually helps if one can understand from their names what they contain! For example, give View node labels suffixes that show what type of view it is.

Compendium also allows templates to be developed to facilitate the creation of maps that follow a known, structured format - a set of questions driven by the process to be followed. These questions can be exported as templates and imported into new maps. ‘Any standard process or methodology can be translated into intuitive Question templates to focus analysis and decision-making. The structured maps that result can then be used to generate documentation, or pass data to another system.’ (Selvin and Shum 2006, pp. 33) The tutorial for compendium is available online.

The facilitator here is reduced to the role of maintaining the information and developing the map in real time by inputting the information as the discussion is carried on, for the participants to view and develop. There is no suggestion that the facilitator contribute to the discussion. The development of the argumentation map is taking place using technology while the discussion is taking place face to face. This tool makes little use of facilitator and so can be considered for use by collaborative communities to maintain records and put the knowledge into perspective within different situations.

Graphically and spatially displayed text tool: Belvedere Inquiry Diagrams

Belvedere Inquiry Diagram is designed to help express graphically how ideas are connected. When an individual is trying to solve a problem or explain something, Belvedere can help keep track of own as well as other people's ideas by allowing them to be mapped out graphically. Belvedere can also aid in figuring out whether there is more information needed to strengthen or complete an idea. Making a diagram (concept map) of an argument can help in seeing the "big picture" out of the details. The system allows for 2 kinds of files to be made - 

• New Collaborative Investigation for diagrams stored on the server and available to others. It allows for collaborative building of the diagram.
• New Private Investigation for diagrams stored on your computer and available only to you or other users of your computer. This allows for development of personal concept maps.

Suthers (1998) believes that these inquiry diagrams are useful for summarizing the overall trends in a multifaceted debate. The visual depiction of ideas and relationships helps participants visualise the abstract ideas. The act of creating the diagrams leads to explanations and reflection on one's ideas.

Learning within Belvedere inquiry diagram

The task of constructing an inquiry diagram encourages individuals to work together, thereby allowing them to form a community of learners. The activity of constructing a shared diagram leads to valuable negotiations of meaning and collaboration plus scaffolding. Participants can switch between working on private and collaborative diagrams with little effort and without losing track of what they are doing, because the diagram keeps track. Since both the computer and the teacher can see what the participants are thinking, and provide appropriate feedback, the map also plays an evaluative role.  They can view their diagram as a graph, matrix or hierarchy as each view supports students' learning in different ways. (Suthers, 2001)

The file is opened with the research question identified with the title of problem statement just under the tool bar. The tool bar itself has all the basic commands in forms of shapes - data, hypothesis, for and against links, and the coach. The large white space is the part of the window where the diagram will be displayed. At the very bottom of the diagram window is another rectangular bar called the message bar where the software gives instructions and messages. Sometimes Belvedere will pop up a small box on the screen, called a dialogue box, to give a message. When Belvedere needs to ask for some information in order to complete an action, then the dialogue box will display some questions for the participant to answer or some areas to fill in.

When one clicks on a statement button such as Data or Hypothesis, in the tool bar, a dialogue box appears, where one types a Statement Summary which will appear in the diagram. By clicking on the Comments tab at the top, further comments (not displayed) can be added on the statement. When entering information in complete, click on the "Add to Investigation" button at the bottom of the box for it to appear on the map at the point you choose. Connections in the form of for or against links can also be drawn from a shape to or from another link, as well as from a link to another link. Editing for shapes and connection is also possible within the software.

While viewing or constructing an inquiry diagram, the participants can consult Belvedere's online Coach for ideas or suggestions on what to consider next. Frequently the Coach will highlight certain statements or relations in the diagram when presenting a suggestion, to point out the elements of the diagram the suggestion refers. When the Coach runs out of new suggestions to give, the button in the tool bar will turn off. Participants may still consult the coach at any time, but its ideas will not be new.

Belvedere Inquiry Diagram and Advice from Suthers et al (1997)

Graphically and spatially displayed text tools – The Concept

Alternative to a linear text display of synchronous communication graphic displays can be used. It is the way the space is created and used to display the interactions that is different. van Amelsvoort (2006) take the view that argumentation has a vital position is the process of negotiating meaning during learning. It is a form of thinking or reasoning. (Toulmin’s Argument Schema, Pragma Dialectics, Waltons Dialogue, Baker’s Learning Mechanisms, Knowledge Building and Progressive Inquiry) Chat is the representation of the communication is in the linear form. Van Amelsvoort (2006) refers to Suther’s (2003) research in relation to using different ways of representing interactions such as diagrams, matrices, text, containers and threaded discussions. The use of diagrams allows the learners to use concepts to construct knowledge. This relates to Buzan’s (2004) work on learning by developing Mind Maps as discussed in earlier blog. Similar to this are concept maps (Akinsanya & Williams 2004, Williams 2004) which are used in many disciplines as visual language. The map allows the individual to develop on prior knowledge and make links between concepts and have a visual pathway of individual understanding of meaning. ‘The concept is usually enclosed in a circle or square (and) a connecting line and linking words showing the relationship between two concepts.’ (Williams 2004, pp 34) van Amelsvoort (2006) refers to Lohse et al’s (1994) ‘network charts’ which show relationship between components via use of lines, arrows, proximity, similarity or containers.  The meaning can be understood from the spatial arrangement of the information. As part of the learning strategy this approach can be taken to allow for collaborative learning (Akinsanya & Williams 2004, van Amelsvoort 2006). van Amelsvoort (2006) goes on to explain that there are four benefits of graphical representation :

•             Diagrams are concrete and require less interpretation than a verbal representation. If we use the argumentation approach, the links would be positive or negative and so nuances can be missed. However, if we take the view that learners are building concept maps, then the wider links and even weaker links can be accounted for.

•             Diagrams show structure and relations and so it is easy to see the patterns. The discussion can be broadened and deepened when using diagram as it generates discussion on clearly visible links.

•             Diagrams give an instant overview and keep a complete record. However, when the diagram is made up of shapes with text in them it is difficult to get the overview specially if boxes contain a lot of text and the map gets complex.

•             Diagrams can stimulate and guide communication and reasoning. The students can use each other’s ideas to generate further discussion which is relevant. It forces the students to make clear arguments and negotiate to come to a shared understanding.

Van Amelsvoort (2004) goes on to say that these reasons have not been substantiated by empirical evidence. It does not give a clear idea about the underlying cognitive processes. The success of the representational tool is dependent on how it is used and not on the fact that the tool is available to use. The effect is also reliant on the kind of task as well as knowledge of the learner. Moreover, the research into tools is hard to generalise as there are many different tools and many different studies. The use of open task diagrams appears to be more congruent to the earlier described benefits. These are again similar to the concept mapping as it is a good way to organise information and bring together prior and new learning. It also makes the process of learning transparent to all involved. (Williams 2004) van Amelsvoort (2004) summarise that ‘diagrams may benefit both construction and communication of arguments for many reasons, such as clarifying relations (Suthers 2003), illustrating the structure of argumentation (Shwarz, Neuman, Gil & Ilya 2000), giving overview (Larkin & Simon 1987), helping to maintain focus (Veerman 2000), and promoting reflection of alternative perspectives, solutions, and critiques (Kolodner & Guzdial 1996).’ (pp 40)

The advantage that graphic tool has over the text based chat is that it is not linear and so can be seen in two dimensional space. ‘Argumentation theory aims to analyse, describe, and evaluate arguments that occur in the real world. It aims to teach students both to think critically about the arguments of others, and to create better, more measured arguments of their own. One of the key tools available to the discipline is diagramming. The claims and their associated reasons within a given argument are identified, and the relationships between them drawn up as trees. This diagram then serves as a basis for criticism and reflection’. (Reed & Rowe, 2004, pp 1) The first premise of this theory is in pedagogy. It uses diagrams to support the teaching of critical thinking skills. The need to teach critical thinking, and the increased use of technology, has led to the appearance of a number of software systems for argument diagramming that are intended for pedagogical use. Secondly diagrams allow for the construction and implementation of theories of argument evaluation within the research community. Lastly, argumentation itself has many applications within computing, especially for artificial intelligence.

Linear text display tool - InterLoc

InterLoc is a web technology specifically developed for computer supported collaborative learning (CSCL). It uses dialogue games to get learners thinking individually as well as together about topics, media or material that is relevant to them. This is achieved by setting up a context and interaction for synchronous reasoned dialogue, such as critical and creative discussion, and allowing the participants to generate reusable content from their group experiences. (Baur et al 2008) InterLoc is a dialogue game that allows group discussion to refine the knowledge already gained through readings. The discussion encourages thinking and collaborative approach to understanding issues. It promotes the development of general critical thinking, argumentation and discussion skills in learners via scaffolding and support from other participants.

The software can be downloaded into home computers for learners to participate in online discussion. However, it would raise the same issues for installing unfamiliar software and trying to learn its use. Also required is the additional Java software for InterLoc to work. It can be installed in the computers in an educational institute and allow for groups to carry on discussions. The learning manager sets up Discussion Activity which includes the discussion question, along with preparation materials. The learning manager also sets up access to learners and facilitators into particular rooms.

 Baur et al (2008) advise that learners should have some familiarisation time after a detailed introduction by the learning manager. ‘It is recommended that a short ‘practice’ session should be carried out, which covers a news item/hot topic that the majority of students will know something about and possibly have a view on.’ (Baur et al, 2008, pp 6) This basic understanding of the tool, as well as carefully chosen activity, is the key to holding learner interest and motivation to collaborate in this synchronous discussion. The wording should be clear and be able to generate argument and rebuttal for interest to be held.  Like all synchronous activities the date and time of session is published and adhered to.

The learners respond to the question / activity by using starter phrases that have been provided. So the learner can choose phrases like ‘I support because’ or ‘I do not agree because’ and so on. A fairly wide range of choice of phrases is provided to allow a good depth to develop in the discussion. Only one person at a time can input their message. This can make the discussion slow but at the same time allow learners to organise their thoughts and put forward considered responses. Thus, it appears to draw on the strength of the asynchronous discussion of having time to compose responses. By being a structured approach to discussion with fixed openers to messages a map of argumentation can be constructed albeit in a linear textual manner. There is a possibility of maintaining a coherent thread to the themes being discussed by responding within the thread just as in asynchronous discussions.

                                  Activity set up for Online Discussion in InterLoc

 

 Discussion in InterLoc

The facilitator is expected to greet the learners and create a collaborative environment just like in other linear tools. Often the facilitator may need to initiate the discussion with the first message. The facilitator’s contribution to the discussion is minimal and kept to sending out general messages to the group, keep order and focus the discussion should the participants stray from the topic (particularly when there are many issues that can be discussed).The facilitator may also be responsible for saving and archiving the discussion which can be accessed later for review as well to continue with the discussion. 

Each discussion room can ideally support small number (5 – 6) of participants at one time as suggested by the developers. With the generally large cohorts in educational courses the learning manager will need to open many discussions so all can participate. This also has an impact on the facilitator time, so now one can appreciate their minimal contribution to the discussion. In fact facilitator maybe supporting 4 to 5 groups simultaneously.  This is a big departure from the facilitator role in a face to face discussion, which can be maintained within the other linear text tools reviewed.