Applying Reach in Direct Manipulation User Interfaces

1
Applying Reach in Direct Manipulation User
Interfaces

• Aaron Toney, Bruce H. Thomas
• School of Computer Information Science
• University of South Australia

2
Common Sense and Math

• If you can not convert your common sense into a
  quantifiably testable hypothesis then at best you
  dont understand what you are talking about.
• In order to be useful for use for human authored
  code design rules, principals, and practices need
  to be expressible in qualitative terms.
• The paper uses set notation to encode and test
  what are otherwise common sense relationships.
  This formal notation was used as it allows
  programmers not only gain intuition about their
  users environment but immediate insight into how
  to implement the software algorithms behind
  applications dynamically based on user reach.

3
What we show in the paper

• For individuals, depth of reach predicts the
  segmentation of the working plane into working
  and storage areas.
• For collaborating users regions of overlapping
  reach enable prediction the formation, size, and
  shape of shared group spaces.
• Applied models of reach predicts the rough impact
  of such variables as working surface size,
  height, shape, and number of users on the working
  surfaces utilization.

4
The Reach Envelope

• Direct manipulation user interfaces are
  constrained to areas reachable by the user.
• The On Table Reach Envelope (OTRE) represents the
  boundary of the set of all reachable points (SR)
  .

5
An Individuals Reachable Space

• Reachable space (SR) is the total space reachable
  by either hand (AL U AR)
• Bimanual reach (SRB) is (AL n AR)
• For individuals working space (SW) is equivalent
  to SRB and the storage space (SS) is the
  relative complement of the total reachable space
  and the working space (SR - SRB).

6
A Groups Reachable Space

• Under the hypothesize that group spaces are
  formed in preference to personal spaces group
  space
• Group space (SRG) can be described as the
  intersection of all users reachable spaces (nSR)
• Co-located working space (SW) can then be
  described as the reachable area that is not a
  group space (SRB - nSR ) or (SRB - SG). These
  definitions agree with both the observations in
  the literature and the results of the user study
  presented in this paper.

7
Converting Common-Sense about Reach into
testable Hypothesis

• Working space (SW) is primarily contained within
  the most easily reachable areas.
• Highest usage will occur in the area of bimanual
  reach (SRB), while lowest usage will occur in
  areas reachable only by one hand (SR -SRB).
• In general group spaces will form in preference
  to personal spaces.
• For each user and task there is an ideal working
  distance (RIDEAL), where reach working surface
  usage probability is attenuated for distances
  other than RIDEAL.

8
What do Simulations Tell us about user Reach?

• (All simulated subjects are anthropometrically
  representative of the 50 females seated 17.5 cm
  from the working surface)

9
Simulated reach at 80, 90, and 100cm square
tables.
Workspace utilization is sensitively dependent
on table size An Increase in table size of only
10cm on a side is predicted to loose the
comfortable communal reach region
10
Predicted on table reach probability for
adjacently seated users

• (90 and 120 centimetre tables shown)

11
Impact of Inter Subject Angle
(45 and 90 degree tables shown)

• Decrease in inter subject angle causes
• Overall decrease in both private and working
  spaces
• Increase in available group space on the table

12
What we learned from initial user studies

• (Sixteen subjects performing a LEGO sort and
  assembly task. Subjects given 997 LEGO pieces
  mixed in with 150 model pieces. There was a
• 45 minute time limit and Hand position tracked
  with Polhemus sensor. Results presented in
  thresholded spatial histogram.)

13
Study Observations

• In general results confirmed the authors working
  hypothesis
• Working space (SW) can be seen as the reachable
  area that is not a group space (SRB - nSR ) or
  (SRB - SG).
• Highest usage will observed in the area of
  bimanual reach (SRB)
• Lowest usage occurred in areas reachable only by
  one hand (SR -SRB).
• Group spaces tended to formed in preference to
  personal spaces.
• Regions of overlapping reach were shown to help
  explain territoriality and the size and shape of
  collaborative areas observed in Scotts (2003,
  2005) example from that work shown right.

14
Applying Reach
15
Questions?
16
(No Transcript)
17
(No Transcript)
18
(No Transcript)