Web Collage is a tool developed to support non-expert designers in the creation of CSCL scenarios. Web Collage is built on top of the notion of design pattern, understood as a general solution to a recurrent design problem (Alexander et al., 1979). The first version of the tool, Collage (Hernández-Leo et al., 2006), integrated a set of six Collaborative Learning Flow Patterns (or CLFPs), which described validated collaborative techniques such as Jigsaw, or Think Pair Share. The integration of CLFPs in Collage supported designers in two ways: first, CLFPs serve as ideas of suitable collaborative learning structures, and second, the tool automatically creates the needed components of the design to implement the CL techniques selected by the designer.
The current version, Web Collage, has been improved by the integration of assessment design patterns, in order to foster designers to think and design assessment plans (Villasclaras-Fernández et al., 2009). Furthermore, Web Collage can retrieve the list of participants from a LMS (such as Moodle), to later let the designer create groups and assign users to groups. Once this step has been completed, Web Collage, working alongside with GluePS! (Prieto et al., 2011), can deploy the learning design into the same LMS, creating the needed activities, groups, etc.
The scenario proposed in this paper can take advantage of some of the features of Web Collage. The left part of Figure 1 shows the “learning flow” which model the scenario learning and staff activities in Web Collage. Some of the activities consist of individual tasks, and most of the activities do not follow any of Web Collage’s CLFPs; therefore, we used generic phases to create a sequence of activities. However, some of the activities can be represented using CLFPs. This is the case of the activities within Lesson 1 in which students have to propose how to have a healthy diet, and to come up with a list of items of food. In Web Collage, this is modeled as a Brainstorming. Similarly, in Lesson 2, at some points students, working in groups, have to select a number of photos related with the inquiry question. Later, the teacher shows to the classroom the photos chosen by each group, and the whole classroom discusses on the photo selection. This couple of activities correspond to the pattern Pyramid. By using the graphical representation of the Pyramid and Brainstorming, the designer’s intentions may be easily understood, including the type of collaboration and outcomes expected for those phases. The left part of Figure 1 shows only the coarse-grain phases; specific learning and support activities can be described inside each phase (see, for instance, Figure 1-A).
In the Pyramid, as well as in other phases, the tool models the number of groups (one “Class” in some phases, seven “Student groups” in other phases). Furthermore, in the close-up picture Figure 1-B, the students’ flow is shown. For instance, we can see that the composition of the seven four-people groups remain the same in different phases of the design.
Another aspect of the design that can be modeled explicitly is the assessment plan. For instance, sections A and C of Figure 1 show the assessment that happens in Lesson 1 and Lesson 5. In Lesson 1, students take a quiz, and this is later reviewed by the teacher to provide feedback to the students in the following activity. In Lesson 5, on the other hand, students first become assessors of the teacher’s oral presentation; later, the teacher assesses the students’ presentations, in order to produce feedback again for the students. These three “assessment processes” are shown by the tool.
Figure 1: Four screenshots of Web Collage’s model of the proposed "Healthy Eating" learning scenario.
On the other hand, the representation of Web Collage has limitations. The most relevant ones are the lack of a mechanism to specify the duration of the activities, and limitations of resource modeling. Web Collage lets the designer add resources, including the URL of the Fuel Box game, or the usage of a whiteboard software system. However, further work would be needed to specify how the user-created resources are going to be used by the participants, or the access to the whiteboard system. While future work with Web Collage will include support to model the usage of resources and software tools in Web Collage, this functionality is yet in its early stages. Finally, another problem is that the specific activities do not model the type of tasks to be carried out in them (expect for assessment activities); the description of each activity is done only in natural language.