Navigational Blocks

The first exercice was to study a tangible tangible user interface, by analysing the design and implementation and the behavior. I studies the Navigational Blocks: Navigating Information Space with Tangible Media, created by the Design Machine Group at the University of Washington.

Analysis of Design and Implementation

Design Concepts

• Kind of information represented physically: each block contains one general question to ask (Who, What, When, Where). The different faces of the block represents subquestion (Who – the women, Who – the founding fathers, etc.)
• Kind of information represented digitally: virtual galleries displaying historical information about the Pioneer Square
• Material & Shapes: wood blocks put on an active space + screen to display information. Cubes are used because it’s easy to understand (not a cognitive challenge to users).

Scenarios

• Users : visitors of the Seattle’s Pioneer Square, so basically anyone
• Scenarios: Someone walks up to a kiosk, takes a block, choose a face (i.e. a question) and put it down on the active space. Then, the information related to the query is displayed on the screen. If two blocks are places in the active space, then the query is “block 1” and “block 2”.

Interaction techniques

• How user interacts with system to perform tasks:
• Selects the exhibits to visits by rotating a block and puting it on the active space to display information.
• Navigate through the rooms by sliding the block to display information
• Haptic feedback about relationships between topics: either attract or repel on another

Implementation

• Main physical components:
• set of four identical wood cubes, each containing a microprocessor, electromagnets (to provide tactile feedback), ball bearing orientation sensor (to determine which face is up), communication device
• Active Space
• Computer with Screen (receives and displays the information related to the query)
• Connections and interactions:
• Blocks send information about orientation through infrared signal to the computer.
• Electromagnets in block attract or repel or do nothing depending on the two blocks put together.
• Sensing technology (what and how): circuits used with bearing ball to determine the orientation (different resistance value for each face)

Interface related skills and knowledge

• Skills and knowledge required: except the ability to read and move blocks, no particular skill or knowledge is required. The system is made with idea that there will be not training required or necessary.

Analysis of Behavior

TAC Palette

Representation			Manipulation
Variable	Token	Constraint	Action	Response
block	Query Data	Surface Other blocks	Add	Displays information according to the face on top.
			Remove	Removes related info from display
			Rotate	Change the query of the information (updates display according to the face chosen)
			Slide	Move in and through the information space

Observability and predictability:

• Does the physical state of the system indicate to the user the internal state of the system? Is there information in the internal state that is not presented in the physical state?

Yes, the physical state indicates the internal state: the face on top of the block represents the information displayed. By changing the physical state (sliding, rotating), the internal state (information) is changed as well. There doesn’t seem to be information in the internal state that doesn’t have repercussions on the physical state: if two blocks are not related, there will be a haptic feedback of repulsion that will indicate to the user the internal state of the system (incompatible query).

• Does the physical state of the system contain enough information to enable the user to determine what tasks are available for him to perform? Can the user determine what tasks are not available?

The constraints of the system (the active space as a surface; the block itself) are intuitive enough that within seconds, users understand how to retrieve information, to initiate a database query. Since each face of the blocks contain words indicating the title of the information that related to them, the physical state of the system contains enough information to enable the user to determine what major tasks are available to him. However, no indication about using multiple blocks at once is provided: it’s a question of whether the user is comfortable enough to test the possibilities.

There is no real information related to the tasks not available.

• Assuming multiple users interact with the system at the same time. Does the physical state of the system indicate to users what tasks they can or cannot perform in parallel? If possible bring an example of two tasks that different users cannot perform in parallel. What in the physical state of the system indicates to users that they cannot perform these tasks in parallel?

There are three tasks (rotating, sliding and combining). Rotating and combining can be done in parallel, and the query will reflect their state. Al though no indication is made about sliding when there are two cubes selected, I assume that sliding one block when the other remains still have the same effect on sliding a block that is alone on the surface. However, sliding two blocks at a time (in parallel) might be problematic, because the system will have a hard time determining which direction (or speed) to display information. Apart from that, any of those actions could be made in parallel. The information displayed should be updated without problems.

Modes

• Certain actions are meaningful only in certain contexts. Bring an example of a user action that is meaningless in one context and has a meaning in another. How can the user determine which actions are meaningful in each context?

Apart from the fact that blocks (tokens) manipulated outside the active space (surface constraint) have no effect on the information display, there doesn’t seem to be any actions specific to contexts. There are only a small number of actions possible, and they are not possible without the required number of blocks.

Physical Syntax

• Does the physical state of the system indicate to users which actions are legal\illegal? Does the physical state of the system indicate to users how actions can be combined into sequences in order to perform a task?

The haptic feedback provided when two blocks are placed side by side indicated to users which actions are legal or illegal. A legal action (resulting in an accepted query) will provide an attraction and an illegal on will make the two blocks repulse each other. However, there is no indication of how the users should sequences actions to perform a particular task.