There are two reports available for this project - a detailed research report and a short account of subsequent trials.

Jungulator trials report

December 2005
Fleeta Siegel

This report is available to download in pdf format:

Jungulator trials report (145kb pdf)

Jungulator research report

October 2005
Teresa Dillon

The full version of this report is available to download in pdf format - see box below. On this page you'll find the report's executive summary.

Executive summary

Self-generative systems are rule-based structures that unfold in 'real-time', with varying levels of complexity. This paper describes the redesign and development of a new user interface for a self-generative audio-visual (AV) prototype called Jungulator. Modifying an existing version of this software, a new application was created for use with young people (14-17 years) within school settings, community arts education and as a professional artistic-performance tool. Within this paper our iterative, user-centred approach to re-designing this tool and the outcomes from our initial workshops are discussed, along with an outline of progress to date.

The main aim of the project was to extend and improve on the existing Jungulator software. In the original version, the audio and visual components of Jungulator were separate. The project supported the integration of both the audio and visual generators, creating an interoperable tool that allowed users (young people aged 14-17 years) to create their own unique arrangements. Integral to this phase was the development and evaluation of an appropriate user interface that facilitates young people's creativity when working with Jungulator.

We envisaged that Jungulator's two main functions would be as:

1. A composition tool: that allows users to input sound and visual sample files, carry out basic editing and manipulate the samples using various self-generative effects. We are also exploring the potential of enabling the tool to allow users to record and save their works.
2. A live performance tool: where ad hoc audio-visual samples can be triggered and manipulated in a responsive, spontaneous fashion.

Our key learning and research objectives were:

1. How can we best integrate the audio-visual aspects of the tool and realise the software's full potential as a self-generative tool?
2. What type of graphic interface would best support young people to work with ease when using the Jungulator?
3. What is Jungulator's potential as a learning tool and how can it support young people's creativity, composition and performance practices in both school and community centre settings?
4. What kind of interactions does Jungulator support?
5. What kind of support material would we need to consider developing for the use of Jungulator in school and community centre settings?

As with all Futurelab projects we were also interested in:

1. What this project tells us about the best ways of designing educational digital resources?
2. What this project tells us about how learning processes can be transformed through use of these tools?
3. How this project helps us understand the potential of next generation technologies to create intrinsically motivating and engaging learning experiences?

We have partially addressed the above research goals, successfully designing an attractive interface that retains the flexibility and responsiveness of the original system. We have made initial attempts at exploring how Jungulator may be used within school and community art settings and with a range of teachers from music, film and media. However further work needs to be completed as we have not touched on the kinds of support material necessary to develop for Jungulator. Additionally given that our current evaluations were more akin to in-depth usability tests, we need to examine the kinds of interactions that emerge when Jungulator is placed in a more authentic setting, over a prolonged period of time. To address this we are currently planning a more appropriate test session, embedding the software in a specially designed VJ course at a community arts centre in London.

To date we have shown that the new interface has the potential to engage young people, providing them with easier access to understanding the random generator functions of the software and allowing them to create real-time, novel audio and visual combinations. Feedback from the young people has indicated that Jungulator pushed their thinking in this field, as the random generator 'spews' out previously unthought-of AV combinations. However some work still remains to be done on the graphic interface. Both teachers and young people continually reiterated that its functions need to be made more explicit. Despite this, the young people found the interface enticing, that it warranted further exploration, and supported them in thinking about new ways of combining sound, image and text.

In terms of the kinds of interactions that Jungulator supports it was clear that participants' use of the software depended on their previous experiences and backgrounds. For example, young people from a traditional music background, who were studying music for their A-levels, expressed a preference for having greater control over the generative process. Such young people were more accustomed to having complete control over the compositional processes and found that they needed clearer definitions of when and where the random generator was applied to the material. Many others outside of the music tradition also expressed an interest in controlling the level of generation and also would have liked to have greater visual representations of how and when the randomisation was applied to the samples. For example, some asked for a timeline, which showed when the generator was applied, or a visual that showed when and how the generator was affecting the sample. One of the most responsive and enthusiastic participants was a young person who was already well accustomed to creating multimedia texts. This participant studied music, music technology, ICT and psychology and outside of school time created animated storyboards and comics. It was interesting to note how this participant worked with the software and he discussed how it could lead him to think of new ways to combine material, which he considered very useful and inspiring.

From the community centre evaluations participants also reiterated the need for working with samples that were your own 'type', ie samples which meant something to you or which you had created. This is important and unfortunately for the trial periods was not possible, which is why a second level of evaluations in community centre contexts, for a sustained period of time is necessary. Finally, from the observational notes gathered and from practical experience in demonstrating the software, it is clear that the kind of visual samples that best articulate the random generation possibilities of Jungulator were minimal, clear visuals. What is meant by this is that single-shot images, with only one central object, worked better than images with several objects in them. From the observations, participants who worked with single-shot images tended to get the most out of the visual generator.

Speaking with teachers and tutors from the areas of music, film and art, practitioners in this field found Jungulator an intriguing software, one which could potentially support young people's brainstorming skills as well as acting as an active storyboard or novel form of presentation. In a similar manner to the young people, teachers also considered the potential of Jungulator in live performance contexts, such as in making visuals for dramas, music groups and in community art settings. However despite teachers' initial positive indications, they were apprehensive about the uses of the tool in secondary school settings, particularly in post-GCSE years. Many expressed concerns over assessment, noting that it was unclear with Jungulator which aspects of the work were the young people's and which were the computer's. Some teachers also questioned whether Jungulator would undermine traditional media skills, desensitising young people and undermining the craft skills which practitioners in this field have fought hard to retain within secondary school contexts. This raised questions about evolving practices in this field and what happens when the computer can do the task better than if a person was to do it manually. This led to a consideration of how such changes will affect practices in this area. Such questions were not resolved in this phase of development but indicated that the software raised a provocative set of questions about the nature of digital media-making practices and the ways in which they are taught in schools. Extending this, teachers also highlighted how Jungulator legitimatised audio-visual practices and provided a novel way of articulating and facilitating multimodal literacy practices. Our next phase of research should extend our work in this area, by providing insight into how a tutor scaffolds and structures lessons around Jungulator and how the multiple forms of expression which the tool supports are combined in a performance situation.

Additionally our evaluation sessions complemented the ongoing quality and assessment tests at the lab, highlighting major bugs which we addressed during the latter period of the project. Although some bugs do remain, we have a working prototype which will be embedded within a community arts centre during November and December 2005. At the time of writing we have already begun to plan these sessions, which will take the form of weekly, three-hour sessions, designed in collaboration with the Futurelab team and facilitated by a VJ tutor. The core aims of this extended period of research will be to address unanswered research questions and gain authentic insights into the use of Jungulator as a composition and live performance tool. Our goal is to build on the findings from our initial period of development, exploring in more depth the strengths and weaknesses of Jungulator from both a young person's and tutor's perspective. We also aim to have better understanding of the kinds of support materials necessary to develop for Jungulator and the skills necessary to use it a teaching tool. Our key questions for this period of development are:

1. How can we best design an appropriate activity to embed Jungulator in a community setting?
2. In this setting, what kind of composition/performance practice and interaction does Jungulator support?
3. What kinds of support materials would we need to consider developing to use Jungulator in school and community centre setting?

In sum, Jungulator builds on previous work in the lab, which explores the use of generative computing systems. The work complements our ongoing research in exploring how best to implement novel uses of computing and digital technologies within learning, by bringing together multidisciplinary teams to co-develop the next generation of digital learning tools.

top of page