- 1 User Requirements
- 2 User Profile Management
- 3 Related Pages
A Cloud4all user workshop was carried out in Berlin at the end of August 2012 with the main aim to gather requirements for the Cloud4all profile initialisation and the profile management tool, as well as to get general information on user needs and preferences regarding user interface auto-configuration within the GPII.
Eight users with different disabilities attended the workshop. The user expressed their needs and requirements with focus in the following disabilities: Low Vision, Learning Difficulties, Blindness, Deafness, Hardness of Hearing, Physical limitations.
The following topics where discussed with the different users:
- Cloud4All main concept and ideas: Within this session the Cloud4All project and its main objectives were presented.
- Self-adapting systems: This session collected first experiences, attitudes and opinions on self-adapting systems. The users talked about their needs regarding user interface adaptation and interaction modalities. The users were asked which problems and obstacles they have to face in their everyday life when dealing with ICT and what kind of experience they have with self-adapting systems. Furthermore the users were asked how ICT solutions have to be adapted to fit their needs and preferences. User answers where collected and pinned to a board.
- Profile Management Tool: Within this session the profile management tool was introduced to the participants. Requirements regarding the tool and its functionalities were collected and discussed.
A second requirement analysis was conducted at the Fraunhofer laboratories in Stuttgart. Focus of the analysis was on requirements for profile management in more depth. In an early stage of the profile management concept work it was decided to focus on elderly persons as user group first, for that reason eight elderly participants with different limitations (low severity level) in seeing, hearing abilities took part in the focus group.
General requirements of user interface auto-configuration refer to both profile initialisation and management.
When talking about auto-configuration of interfaces and devices the following main areas should be adaptable to the user’s needs and preferences:
- UI parameter: e.g. font size, font type, colors, etc.
- Content: e.g. simple language
- Assistant technologie (AT): different AT software should be available within the GPII cloud, AT hardware connected to the end device should be automatically identified by the GPII system.
- Range of AT functions: users want to define the used range of functionality of AT.
- Context parameters: The user interface should be automatically adapted to changing context variables like light conditions. This adaptation should be independent from the actual user interface profile of the user.
In fact, as much parameter and variables as possible should be adaptable.
Some more relevant requirments regarding user interface profiles:
- Auto-configuration: self-acting should not be based on diagnostic data.
- Privacy settings: should be included in the management tool because users want to decide between high privacy/data security, accepting less comfort/easy access and less privacy/ data security and high comfort/easy access.
A major outcome of the requirement analysis is that users are willing to spend a lot of time on profile initialisation, if they have an added value afterwards, e.g high accessiblity, high usability. Furthermore all workshop participants want to do the profile initialisation by themself or in mutual exchange with friends as support. Help from a second person is conceivable in case of cognitive difficulties or low computer literacy.
The following requirements could be gathered regarding profile initialisation:
- Users would use apersonal device (e.g. personal computer, smartphone, tablet PC) for profile initialisation that is already adjusted to the user needs/ already equipped with the needed AT. On public devices only profile management is conceivable.
- There should betwo modes for profile initialisation – an expert mode for advanced users and a simple mode for users with cognitive difficulties and low computer literacy.
- Profile initialisation could be handled comparable to an application store: different alternatives for preference settings are offered and can be accepted or rejected.
- Instead of selecting predefined personas users would prefer astep by step approach for profile initialisation (e.g. dialogue wizard), where user interface parameters can be adjusted to the user’s needs by selecting an option out of an defined number of proposed options or by adjusting the value for an interface parameter (simple and advanced mode).
- Next to the step by step approach for profile initialization, the user should be able to take over the device settings of the device used for initialization as default profile.
- If the user want neither the wizard or the to take over the device settings, the profile should be left blank until the user changes the setting of a parameter.
- Users can imagine to generate a first „rough“ profile, that is extendable, as well as to compile a detailed profile in one session.
- Users would prefer to set main preferences by selecting between alternatives first and to adjust the preference setting afterwards in more detail.
- Simple language and pictures/videos should be used to support the user in profile initialisation.
The following requirements were gathered concerning profile management:
- The profile management tool should be clearly structured.
- The menu structure should be not too deep – three levels within the menu hierarchy are adequate.
- The WYSIWYG principle should be applied – when changing a preference setting a preview of the change should be given.
- Menu items should be clearly labelled. Simple language should be used.
- Preferences should be clearly categorised within the profile management tool. A categorisation in sensory channels, e.g. „visual“, can be ambiguous, because it is not clear if the category contains preferences regarding visual elements or settings for people with visual impairments.
- A filter function should be provided to select between a view of the main preferences and all preferences.
- The management tool is mostly used on personal computer so a browser based tool is suitable.
- The management tool should fulfilaccessibility standards
- Recommendations for preference settings could be a useful tool for selecting suitable AT software (e.g. screen readers). settings can indeed be suggested, but they should nevertheless be personally customizable (for example, some customizable schemes). Recommendations should be given not too intrusive.
- There should be an opportunity to maketemporary settings that will not be stored within the user interface profile
- A global profile should consist only of the main important preference settings (e.g. to assure that an interface is perceivable). Device and platform specific settings should be managed in sub-profiles
- Users would like to manage profiles and to change preference settings of public devicesvia remote, using a personal device (e.g. preferences settings for a ticket machine are assessed via a persona smartphone using near-field-communication).
Approach & Decisions
High impact variables
Previously User interface variables have been defined, which are the basis for the basic profile parameters. It is necessary to reduce all adaptable interface variables to a defined number of variables that are presented to the user as preference settings in a first step. This is also important for the profile initialization. Even though users are willing to invest a lot of time in profile initialization it is not useful to confront a first time user with all variables.
This reduced set of preference settings will be called high impact preferences. They are fundamental to make an interface accessible or rather perceivable. These variables/ preferences will be presented within the Discovery Aid Wizard and will be provided in the filtered view of the profile management editor.
There are different ways to identify these high impact preferences:
- User interviews could be used to identify the variables that are adjusted by the user when accessing an interface. This approach focusses mainly on the user preferences and would lead to a high user satisfaction.
- Common guidelines for accessibility and user interface design could be reviewed and thereof high impact variables could be deduced.
- Interface and Accessibility experts could reduce the list of adaptable variables to a set of high impact preferences relying on their expertise.
For the first conceptual stage of the profile management development the expert approach was used to identify the high impact variables focusing on elderly users. The user can set the high impact preference variables within the Discovery Aid Wizard by selecting a proposed alternative of this variable. Within the profile management editor the preference setting can be done by adjusting to a specific value.