[Date Prev] [Date Next] [Thread Prev] [Thread Next] Indexes: Main | Date | Thread | Author

[ba-unrev-talk] HyperScope Prototype: "WIMPy" to "SILKy" GUI?

Exploring the human side of computing & To Err is Human    (01)

Exploring the human side of computing
"Prof. Raj Reddy at Carnegie Mellon University, refers to his vision of
moving users from WIMPy to SILKy, from Windows, Icons, Mouse, Pointing
(WIMPy) to Speech, Image, Language and Knowledge (SILKy) — transforming
users from information technologists to true knowledge workers, focusing
not on software programs and raw data but on tasks, information,
outcomes and results.    (02)

"The most precious resource in a computer system is no longer its
processor, memory, disk or network. Rather, it is a resource not subject
to Moore's law: User Attention. Today's systems distract a user in many
explicit and implicit ways, thereby reducing his effectiveness."
< http://www.usatoday.com/life/cyber/ccarch/ccmak004.htm >    (03)

I extracted the text below from the following MS-powerpoint
presentation, since it may be of value for developers of HyperScope to
discuss and consider in the design criteria.    (04)

To Err is Human
Computational Limits to Human Thinking : Implications for the Design of
Human Centered Interfaces
(Raj Reddy, Carnegie Mellon University, Pittsburgh, PA  15213)
March 16, 2000
<  http://www.rr.cs.cmu.edu/toerr.ppt >    (05)

Human Strengths and Human Limitations    (06)

Strengths: People    (07)

   * communicate using speech and natural language
   * tolerate errorful, ambiguous and imprecise input
   * exploit vast amounts of knowledge
   * learn from the environment    (08)

Limitations: People    (09)

   * make errors
   * tend to forget
   * become impatient
   * get confused
   * need to collaborate
   * tend to be lazy    (010)

 WIMPy Interfaces have largely Ignored Human Strengths and Human
Limitations    (011)

   * Need to move towards SILKy
   * Interfaces: Speech, Image, Language and Knowledge based human
     centered interfaces    (012)

(Video of Carnegie Mellon Communicator illustrating interactive spoken
language dialog)    (013)

To Err is Human    (014)

Problem:    (015)

   * Catastrophic loss of data:  “I didn’t mean to do that ?”
   * Unanticipated side-effects: “how did that happen?”    (016)

Causes:    (017)

   * Sensory, Cognitive and Motor overload
   * Information Overload: “Like being in a traffic jam”
   * Timing Errors: Simplify the task    (018)

Present Solutions:    (019)

   * Partial Undo
   * Dialog Boxes    (020)

 To Err is Human (Cont.)    (021)

Future Opportunities:    (022)

   * Unlimited Undo
   * Anytime Anywhere Abort
   * DWIM (Do What I Mean)    (023)

System Issues:    (024)

   * Require redesign at the OS level, network level and the application
     level    (025)

To Forget is Human    (026)

Problem:    (027)

   * A non-expert occasional user can’t be expected to remember the
     details    (028)

Causes:    (029)

   * Forgetting is the loss of indexing structure
   * Redundancy in the indexing structure is the key    (030)

Present Solutions:    (031)

   * Recognition vs. recall: GUIs and Menus
   * On-line manuals    (032)

To Forget is Human (Cont.)    (033)

Future Opportunities:    (034)

   * Use of color, fonts, voice responses for focusing attention    (035)

Intelligent Help    (036)

   * “How do I” and “What if” MultiMedia documentation
   * Learning by Doing and Learning by Example
   * “Reference librarian” agent
   * Chemical abstracts metaphor    (037)

System Issues:    (038)

   * On-line help must change: enormous investment    (039)

 To be Impatient is Human    (040)

Problem:    (041)

   * Time to get the answer in interactive problem solving
   * Reduce the response time
   * Network operations:   Unpredictable retrieval and browsing times    (042)

Automatic MSN and AOL updates at login    (043)

Present Solutions:    (044)

   * Hour glasses and wheels
   * No idea how long it will take
   * Progress bars    (045)

To be Impatient is Human (Cont.)    (046)

Future Opportunities:    (047)

   * Updates in the background
   * Learn from experience: self-aware systems
   * Look ahead retrieval and computation
   * Hurry-up algorithms
   * Keystroke model    (048)

Systems issues:    (049)

   * Introduction of monitors in OS and applications
   * Background multitasking: intermixed packets    (050)

To be Confused is Human    (051)

Problem:    (052)

   * Unable to deal with information clutter    (053)

Causes:    (054)

   * Information overload
   * Computational constraints on human thinking
   * Incomplete and ambiguous information    (055)

Present Solutions:    (056)

   * Hide and/or re-arrange windows
   * Illegal syntax
   * Restrict use of Natural Language    (057)

To be Confused is Human (Cont.)    (058)

Future Opportunities:    (059)

   * Infoglut: Filters and Agents
   * Human Attention: Use multiple sensory modalities
   * Incomplete and Ambiguous Information: clarification dialog    (060)

System Issues:    (061)

   * Agent architecture and integration into Operating Systems.    (062)

To Collaborate is Human    (063)

Problem:    (064)

   * Many tasks which cannot be done by a single person    (065)

Causes:    (066)

   * A single person may not have the skills to solve a given problem
   * Independent / Loosely Coupled / Closely Coupled Interactions    (067)

Present Solutions:    (068)

   * Instant Messaging and  Chat Rooms
   * Serial sequential problem solving    (069)

Groupware    (070)

To Collaborate is Human (Cont.)    (071)

Future Opportunities:    (072)

   * Concurrent Engineering
   * Parallel Asynchronous Problem Solving
   * Common Language and Conventions
   * Building a bridge or plane
   * Distributed in Space
   * Distributed in Time and Space    (073)

Systems Issues:    (074)

   * Systems need new representations and architectures
   * Collaborative Writing: Transaction Files
   * Collaborative Design: Structured Dialog Trees
   * Collaborative Planning: Abstraction of Interactive Dialogs    (075)

To be Lazy is Human    (076)

Problem:    (077)

   * Most people use a minimal subset of functionality in Word,
     PowerPoint, etc.
   * Most people avoid tasks requiring too much cognitive effort
   * PGP - too much work
   * FTP  - too complex    (078)

Causes:    (079)

   * Principle of least effort    (080)

Present Solutions:    (081)

   * Tip of the Day in Word    (082)

To be Lazy is Human (Cont.)    (083)

Future opportunities:    (084)

   * Advice giving agents that look over your shoulder
   * Just-in-time learning
   * Gentle slope systems
   * Agents (wizards!) that know about PGP, FTP, or whatever    (085)

Systems issues:    (086)

   * Applications that know about their own functionality
   * End user agent creation technology
   * Intelligent tutoring tools    (087)

Sign Up for NetZero Platinum Today
Only $9.95 per month!
http://my.netzero.net/s/signup?r=platinum&refcd=PT97    (088)