Tonight Human Information n Xerox Star Processing n History Xerox - PDF document

Tonight Human Information n Xerox Star Processing n History – Xerox Parc n Design – Desktop metaphor n Human Information Processing CSEP 510 n Memory Lecture 3, January 22, 2004 n Fitt’s Law - Movement Richard Anderson n GOMS/KLM – Human modeling Announcements Saigon Deli – U. District Xerox Parc Alto - Star (Palo Alto Research Center) n Parc invented more than its share of n Enabling technology successful computing technologies n High DPI screens n Not economically n Alto viable machines n Ethernet n Star price $16,500 in 1981 n Smalltalk n 384 KB RAM, 10 MB n Bravo (Simonyi -> Word) Hard disk, 8 inch floppy drive n Laser printing n Nor was the Apple Lisa at $9995 in n Press (Interpress -> Adobe) 1983 1

Xerox Star Document centered computing n Single user computer “Star, in contrast, assumes that the primary use of the system is to create and maintain documents. The n Document Centered Computing document editor is thus the primary application. All n Desktop Metaphor other applications exist mainly to provide or manipulate information whose ultimate destination is n Direct manipulation the document.” n Modeless n Other types of computing n Developer Centered Computing n Computation Centered Computing Desktop Metaphor Desktop Organization “Every user’s initial view of Star is the Desktop, which resembles the top of an office desk, together with the surrounding furniture and equipment.” n Documents and tools available on desktop n Waste basket, floppy drive, printer, calendar, clock, files, in basket, out basket n Document organization on desktop (grouping, piling) n Windows compromises on desktop metaphor n Task bar Metaphorically speaking Direct manipulation n Why use metaphors? n Physical / continuous actions n Drag file to move (or delete) n Resize windows by dragging n Why build UI around a metaphor? n Direct vs. Command not completely distinct n What are the pitfalls about metaphors? n Window resize by pointing to source / target 2

Direct manipulation Modes n Recognized as a key UI problem by Parc n What primitives are available for Researchers direction manipulation? n Modeless editor n When is direct manipulation superior? n Evil modes n Insert / Overwrite / Delete n When is command superior? n Copy vs. Move n Is direct manipulation easier to learn? n Good modes (?) n Color and other ink effects n Is command more powerful? n Text formatting n Is one form less risky than the other? n What about cruise control? Noun-Verb vs. Verb-Noun Human Information Processor n Noun-Verb n Model how a human work to n Choose object, understand how to design interface choose operation n Attempt to make HCI more rigorous n Predictive and explanatory n Verb-Noun n Choose operation, choose object Simple interaction model Basic operations n Vision n Memory n Physical movement n Mental processing 3

Memory Simple experiment n Working memory (short term) n Volunteer n small capacity (7 ± 2 “chunks”) n Start saying colors you see in the list of n 6174591765 vs. (617) 459-1765 words n DECIBMGMC vs. DEC IBM GMC n rapid access (~ 70ms) & decay (~200 ms) n When the slide comes up n pass to LTM after a few seconds n As fast as you can n Long-term memory n Say “done” when finished n huge (if not “unlimited”) n Everyone else time it n slower access time (~100 ms) w/ little decay Simple experiment n Do it again Paper n Say “done” when finished Home Back Schedule Page Change Memory n Interference Yellow n Two strong cues in working memory Green n Link to different chunks in long term Red memory White Orange Brown 4

Memory and application design Physical Input Devices n Novice vs. expert use n Difficulty for user in navigating application n Ability for expert users to thrive on obscure systems n Control navigation techniques n Grouping, Icons, Conventions, Shortcuts n Limit short term memory usage Physical Movement Modeling human action Target selection n Speed – key strokes per second n Fitts’ law n Precision – how large a target is needed ID = log 2 (2A / W) Where: n Task complexity ID is the index of difficulty n Difficulty of specific tasks A is distance moved (amplitude) n Trade offs (distance, speed, accuracy) W is the target width History Fitts’ Law n Information Theory n ID = log 2 (2A / W) (1940s) n MT = a + b ID n Shannon, Wiener n Basic predictions n Human Performance modeling (1950s) n Difficulty is the ratio distance and target n Miller, Hick, Hyman, size Fitts n Operation time increases logarithmically in n Application to HCI distance and precision n Card, English, Burr (1978) 5

Why do we believe this? Implications of Fitts’ Law n Substantial experimental support n Radial Menus n Uniform difficulty n Very high correlations observed n Standard Menus n Results for wide range of devices / n Increasing difficulty scenarios from current selection n Increase item size to keep difficulty constant Systems level modeling of Homework assignment humans n Write a program to test Fitts’ law n How should a computer think about the user? n Bring to class next week (?) n Suggested platform – Tablet PC n Development for Tablet PC can be done on a windows desktop machine Model Human The Model Human Processor Processor n Card, Moran, Long-term Memory Newel, 1983 Working Memory n 3 processors sensory Visual Image Auditory Image n 4 memories buffers Store Store n 19 parameters n 10 principles of operation Eyes Motor Cognitive Perceptual Processor Processor Processor Ears Fingers, etc. 6

MHP Basics Modeling human activity n Text editing by expert users n Based on empirical data n Users relied on repertoire of patterns n Three interacting systems n Search / problem solving behavior not observed n perceptual, motor, cognitive n Cognitive skill n Serial and Parallel n Key stroke model n Engineering level model to predict behavior on n Parameters specific task n processors have cycle time (T) ~ 100-200 ms n GOMS Model n memories have capacity, decay time, & type n Model behavior in a domain where users have a set of patterns to use Keystroke level model User study n Analyze task by summing individual n 28 users, 10 systems, 14 tasks operation times n 12 users on editors, 4 tasks n 4 on each of 3 editors Moving hand to mouse 360 ms n 12 users on drawing programs, 5 tasks Pointing to a new line with mouse 1500 ms n 4 on each of 3 drawing programs Clicking the mouse 230 ms Moving hand to keyboard 360 ms n 4 users on systems utilities, 5 tasks Total 2450 ms Editing systems Editing tasks n 12 users, 3 systems, 4 users per system n T1. Replace one 5-letter word with another n Users only worked on one system n Users given 10 instances each of 4 tasks (40 n T2. Add a 5 th character to a 4-letter total) in randomized order word n Data logged and user video taped n T3. Delete a line, all on one line n Training n T4. Move a 50-character sentence, n Typing test for calibration spread over two lines, to the end of its n Operations specified for tasks paragraph n Practiced on typical instances of the tasks 7

Methodology / Results Discussion n Unsuccessful tasks discarded (31 %) n Experiment n Compute / derive operation times n Participants n Predicted execution times within about n Methodology 20% n Analysis GOMS GOMS n Modeling behavior where users have n Goals patterns of use n Goals available for solving the task n Operators n Primitive operations n Methods n Compiled collection of sub-goals and operators n Selection rules n Rules to choose amongst methods GOMS Example Room Cleaning: Room cleaning Goals n Goal: Clean room n Goal: Put away item n Goal: Pick up toy set n Goal: Put set item in box n Goal: Make bed 8

Room Cleaning: Room Cleaning: Operators Methods n Method: Pickup dirty clothes n Pickup Object n While dirty clothes on floor n Carry Object n Pickup clothing item, place in laundry basket n Drop Object n Method: Push stuff under the bed n Method: Pickup multiple toy sets (A) n Push Object n While pieces on the floor n Throw Object n Put piece in the appropriate box n Place Object n Method: Pickup multiple to sets (B) n Make pile for each set n Open Drawer n Dump each set in appropriate box n Close Drawer Room Cleaning: Selection rules Class Exercise n Multiple Sets – greedy algorithm n Design a GOMS for the task of processing email n Multiple Sets – partition algorithm What is the value of GOMS? Short comings of GOMS/KLM n Skilled users n Does not address mental workload n Ignored learning n Ignores user fatigue n Errorless performance n Does not account for individual n Did not differentiate differences cognitive processes n Does not consider n Serial tasks broader issues of application 9

User variation Skilled vs. Unskilled users n Extent of knowledge of tasks n What is the difference between modeling skilled and unskilled users n Knowledge of other systems n Motor skills n Technical ability n Experience with system n Novice, Casual, Expert Modeling Errors Parallel vs. Serial execution n How would you model a KLM with n Instruction scheduling analogy errors? n Summing individual instruction times on a pipeline processor is a poor predictor n Does this analogy apply for KLM? n How does GOMS apply to email when user is working on many messages simultaneously? Lecture summary n Xerox Star n History - commercial realization of a radical vision n Design – introduced new computing metaphor n Human side n Understand basic human operations n Model humans to support rigorous analysis of applications 10