Title: Models and Theories
1ModelsandTheories
2GOMS ExamplePhotocopying an article
- Create a GOMS description of the task of
photocopying an article from a journal. - Assume
- copy 1 page at a time
- cover must be over the imaging surface of the
copier before a copy can be made. - What can the pit fall of this design be?
3Goal PHOTOCOPY-PAPER Goal LOCATE-ARTICLE Goal
PHOTOCOPY-PAGE repeat until no more
pages Select Goal SELECT-PAGE --gt
CHOOSE-PAGE-TO-COPY Goal ORIENT-PAGE OPEN
-COVER POSITION-PAGE CLOSE-COVER PRESS-BUTTON
Goal VERIFY-COPY LOCATE-OUT-TRAY EXAMINE-COPY G
oal COLLECT-COPY LOCATE-OUT-TRAY REMOVE-COPY
(outer goal satisfied!) Goal RETRIEVE-JOURNAL OPE
N-COVER REMOVE-JOURNAL CLOSE-COVER
4GOMS and Selection Rules
Selection rules exist if a spoiled copy was
printed. Consider the following Rule 1
SELECT-PAGE if last page was copied successfully
or start of article. Note The goal SELECT-PAGE
is only valid if we are at the start of the
article or the last copy was successful. If the
last copy was spoiled the we must recopy the
current page, so only a re-orientation would be
required.
5Goal PHOTOCOPY-PAPER Goal LOCATE-ARTICLE Goal
PHOTOCOPY-PAGE repeat until no more
pages Select Goal SELECT-PAGE --gt
CHOOSE-PAGE-TO-COPY Goal ORIENT-PAGE OPEN
-COVER POSITION-PAGE CLOSE-COVER PRESS-BUTTON
Goal VERIFY-COPY LOCATE-OUT-TRAY EXAMINE-COPY G
oal RETRIEVE-JOURNAL OPEN-COVER REMOVE-JOURNAL CL
OSE-COVER Goal COLLECT-COPY LOCATE-OUT-TRAY
REMOVE-COPY (outer goal satisfied!)
Closure to Outer Goal, must force user to collect
copy last
6GOMS
- We need to force the outer goal.
- Possible Solution
- Do not eject the page unless the cover is up.
- Warning light if something is left in the copier.
- Copy cards can be withheld until cover is lifted.
7Table 4.1 Raskin Placing the Ms
- Rule 0 - Insert Ms in front of all Ks. In front
of Ps but not arguments of those commands. - E.g.
- Menu sub selection
- Rule 1 - Delete anticipated Ms
- If we point and click then the click takes no
mental preparation. It is an automated response. - Rule 2 - Delete Ms in cognitive units
- Cognitive units are chunks of actions which are
performed as a unit. Like typing a short
sentence. 1 M needed for whole string.
8Table 4.1 Raskin Placing the Ms Cont...
- Rule 3 - Deletion of Ms before consecutive
terminators. - E.g.
copy 3p94 marks.xls /backup
Consecutive delimiter and /
Consecutive delimiter space and
9Table 4.1 Raskin Placing the Ms Cont...
- Rule 4 - Deletion of Ms that are terminators.
- E.g.
- Terminators are habitual and thus become part of
the string. - No thought required to execute
rm . ?
Carriage return is delimiter
10Table 4.1 Raskin Placing the Ms Cont...
- Rule 5 - Deletion of Overlapped Ms.
- Delays R caused by the system which overlap a M
does not count. - User is waiting for the system.
- The fact that they are thinking about the next
task is masked by the system delay. - E.g. cp . .. ? logout ?
- MkkkMkkkMkk R Mkkkkkk
- The green M is removed since the user is waiting
for the system to respond R - The logout task has been cognitively processed
during the R. We do not count this M.
11GOMS example
- We wish to solve an age old debate. Is using just
keyboard commands to edit a string faster or
slower then using a mouse and a keyboard to edit
the same string. - We assume expert knowledge
- Users are trained and proficient.
12GOMS example cont...
- Given the following string
- Hi! how are you.
- Edit this string to read
- Hi! how do you do.
- Assume the cursor is at the beginning of the
string. - What is the relative time measure if we were to
edit this string in NotePad using. - Keyboard
- Keyboard Mouse
13GOMS example Cont...
- Keyboard
- We have knowledge of short cuts such as
- CTRL arrow skips to beginning of next word
- CTRL Shift right arrow highlights word.
- ? ? ?? do? ? ? ?do
- KK KK KK KKK
KK K KKK - MKMK MKMK MKMK MKMKMK MKMK MK MKMKMK
- Delete Ms in cognitive unit.
- MKK KK MKK MKKK MKK
MK MKKK - 6 Ms 61.35 8.1
- 15 Ks 15 0.2 3
- Total 11.1 seconds
14GOMS example Cont...
- Keyboard Mouse
- dcl double click of mouse
- scl single click of mouse
- dcl are do? scl ltlocate cursorgt
?do - MHPMKMK MHKMKMK HMPMK
MHKMKMK - Delete anticipated Ms
- MHPKK MHKMKMK HMPK
MHKMKMK - Delete Ms within cognitive units
- MHPKK MHKKK HMPK
MHKKK
15GOMS example Cont...
- 4 Ms 4 1.35 5.4
- 2 Ps 2 1.1 2.2
- 4 Hs 4 0.4 1.6
- 9 Ks 9 0.2 1.8
- Total 11 seconds.
- Note there is almost no difference between the
two methods. - Argument solve.
16Alternate
- B Represents mouse click 0.1 sec.
- BB Double mouse click 0.2 sec.
- Consider editing the string.
- dcl are do? scl ltlocate cursorgt
?do - MHPMBMB MHKMKMK HMPMB
MHKMKMK - 4 Ms 4 1.35 5.4
- 3 Bs 3 0.1 0.3
- 2 Ps 2 1.1 2.2
- 4 Hs 4 0.4 1.6
- 6 Ks 6 0.2 1.2
- Total 10.7 seconds.
- Note Basic times vary depending on skill of
user. - Eg. M can range from 0.6 to 1.35.
17NYNEX
- The US telephone company NYNEX was intending to
install a new computer system to support their
operators. Before installation a detailed GOMS
analysis was performed taking into account the
cognitive and physical processes involved in
dealing with a call. Because an operator performs
several activities in parallel a PERT-style GOMS
description was constructed. The PERT analysis
was used to determine the critical path, and
hence the time to complete a typical task. It was
discovered that rather than speeding up
operations, the new system would take longer to
process each call. The new system was abandoned
before installation, leading to the savings of
many millions of dollars.
18How to Evaluate using KLM
- Gives a relative time when all other things can
be considered equal. - Must ensure there is no variance in
- Knowledge w.r.t. system, all users have
functional knowledge of system. - Users, all are trained equally well on both
systems. - Same tasks are performed on each system.
- Other factors for the KPH Ms do exist. These
are relative to system.
19Goal Orientated Help System
- Outer Goal - Connect to the Internet
- Sub-Goals - listed as steps 1to7
- GOMS can be used to analyze how a user solves
problems and then create a help system which
takes advantage of this model. - The model will mirror the thought process of
solving the problem.