MULTI-SKILL COOPERATION IN USER INTERFACE DESIGN

I. Lambert*, N. Portolan**

*CCETT
4 rue Clos Courtel
35512 Cesson Sevign�
+ 33 99 12 45 48
lambert@ccett.fr

**CCETT
4 rue du clos courtel
35512 Cesson Sevign�
+ 33 99 12 43 84
nportola@ccett.fr

ABSTRACT

The importance of pictures in today's interfaces makes a multi-skill approach between various people necessary: ergonomist, graphic designer, terminologist, psychosociologist. The question of the role of each partner and the integration of the different approaches is dealt with via two design projects.

KEYWORDS :

design process, ergonomist, graphic designer, terminologist, telecommunication product.

INTRODUCTION

The evolution of user interfaces is expressed by the increased importance of pictures shown in the form of graphics, pictograms, icons, metaphors, animations, use of 3D. These new interfaces raise specific questions of design. Pictures, more than text, evoke the aesthetics and imagination to the cultural universe of users. Pictures are also more polysemic than text.
Dealing with pictures issues requires a multi-skill approach involving various partners - ergonomist, graphic designer, terminologist, psychosociologist, software engineers - in order to reach the best solutions.

The purpose of this paper is to discuss the cooperation between these partners within a design team. We will illustrate this through two projects in the telecommunications sphere in which we have been involved as ergonomists:

the design of a metaphor for access to telematic services,
the design of a touch screen mobile telephone with a user interface based on icons.

It is not possible to present the process of each project in this communication. We have preferred to outline a few key points of the collaboration which seem of most importance to us.
The cooperation between partners from different disciplines raises questions at various levels:

what is the focus of each partner and from what point of view does he approach the picture issues?
at what stage of the project and with what level of involvement each partner must intervene ?
how are the partners approaches integrated in the course of the project?

DIFFERENT POINTS OF VIEW ON PICTURES

If certain elements of the interface are treated more specifically by a partner, pictures are nevertheless handled differently by all the partners with their differing points of view. Two examples allow us to illustrate this question: icons and metaphor.

The case of icons

For icons, the points of view are as follows:

The ergonomist approaches them from a functional and dialogue standpoint: the icon is seen as a dialogue element allowing implementation of a function. He will therefore watch if the user easily pinpoints the function with respect to the icon.
The terminologist takes the approach from a semantic standpoint : the icon is seen as a graphical representation of a definite wording. He will therefore watch if the user associates the graphics with the expected meaning.
The graphic designer approaches them from an aesthetics and graphical legibility standpoint. He will watch if the user finds the icon aesthetically pleasing and if he can easily see what it represents.

Therefore each partner approaches the same icon with different questions. This multiple-approach allows choice and compromise in the most reliable and effective way.

For example, an icon can easily be associated with a function by users on a drawing board, althrough it is not legible on the screen because it presents too many details. The demand made to the graphic designer is to simplify the representation, which may affect user comprehension.
A compromise is to look for a balance between legibility vs significancy for graphics. Iterativity between graphic creation and user tests will assess the validity of this compromise.
As another example, the terminologist with an out of context approach will throw light on usage difficulties noted by the ergonomist. This might also bring further suggestions for icons. In certain cases, on the contrary, problems of meaning which appear in the out of context terminology tests are settled in real situations, context then reduces polysemy. Here also, iteration between ergonomical and terminological approaches will allow to find the most suitable solution.

The case of a global metaphor

For the metaphor design project, another partner, a psychosociologist was involved. He focused on the very content of the metaphor. His intervention helped to better understand user representations and therefore to define the metaphor themes.

For the metaphor, multi-skill approach seems to be more interwoven and has been more difficult to identify. This is due to the stronger integration of pictures in the product concept itself because the functional architecture is built upon graphics.

This predominance of pictures in the interface had moreover the effect of giving the graphic designer too important a role at the beginning of the project in comparison with the other partners.

INTERVENTION IN THE DESIGN PROCESS

Competing methods to be set out Intervention methods of each partners must be linked together in the project. These methods are implemented in order to ratify the design choices of the users and to therefore progressively define the interface.

Ergonomical and terminological approaches are both based on heuristic evaluation and user tests. The graphic designer approach is essentially based on expertise and a knowledge grounded on graphical legibility and aesthetic rules. The psycho-sociologist is more specialised in extracting ideas from user groups.

In the touch screen phone project, heuristic evaluations were done individually, and results discussed during meetings. At the end of the heuristic evaluation, the elements of the interface to be validated and the tests were prepared together. For example, the "problematic" graphic elements were listed with possible alternatives grouped together, drawned by the graphic designer and put forward in the course of the test.
User tests were conducted concurrently by the ergonomist and the terminologist, according to the methodology mentioned by M. Blandel and N. Portolan (1) . The graphic designer assisted at some sessions in order to better understand the problems linked to graphics. Results were consequently discussed with software engineers.

In the metaphor project, the key phases have also been the object of cooperative work conducted together. For example, following an evaluation phase, it appeared that the chosen metaphor presented fuzzy areas for certain functions. A user group was realised by the psycho-sociologist to re-evaluate the metaphor at a more conceptual level.
On the other hand, maintaining the coherence of the group was not an easy task. In fact, little by little, for practical reasons, the design was supported by a nucleus consisting of the graphic designer, the ergonomist and the software engineer. The rest of the group found itself to be put back, as it was not quite following the concept evolution closely enough. To overcome this problem, we decided in the course of the project, to formalize the work phases with roughs. These roughs were then distributed to all partners in the group for individual heuristic evaluation. This use of drawings on paper allowed a greater interactivity in the group and improvements in collaboration. Plenary meetings of the group took place at more critical moments, e.g. to consolidate heuristic evaluations.

Partners elbowroom within projects

Partners operation latitude is different throughout the design process. The ergonomist has a more limited operation latitude, because certain elements of the interface are arranged sooner than others, such as the functional architecture, the linking of screens and the dialogue principles. The terminologist and the graphic designer can bring about modifications later, as it mostly influences presentation software, it is still possible to change the wording or the graphics of an icon at the end of a design. For the metaphor however, all partners must intervene from the very start of the project, as the operation latitude is limited in time for concept and graphical functional architecture.

CONCLUSION

Grouping different skills together for a project isn't enough to generate cooperation in the design. Key issues still remain to be settled, namely : everyone's tasks definition and a global project methodology that fits partners contributions all along the design process. The formalization of this work has to be carried on, especially with the integration of marketing skills to define the commercial visual identidy for our products.

REFERENCES

Blandel, Portolan, User interface design and terminology : how to succeed the integration ?, in proc. HFT'95 (Human Factors in telecommunication) Telecommunications), pp.123-129.