Exhibit 1015 – Part 4
`
`Exhibit 1015 — Part 4
`
`

`
`Ch. 4 Command Languages
`
`163
`
`X#
`S#
`R#
`M
`RF‘
`
`detailed information on a listed flight
`schedule information for the listed fare
`return flight information for this route
`main menu
`return fares
`
`Experienced users come to know the commands and do not need to
`read the prompt or the help screens.
`Intermittent users know the concepts
`and refer to the prompt
`to jog their memory and help them retain the
`syntax for future uses. Novice users do not benefit as much from the
`prompt and must take a training course or consult the online help.
`The prompting approach emphasizes syntax and serves more frequent
`users.
`It is closer to but more compact than a standard numbered menu
`and preserves screen space for task—related information. WORDSTAR
`offers the novice and intermittent user help menus containing commands
`with one or two word descriptions (Figure 4.7). Frequent users can turn
`off the display of help menus, thereby gaining screen space for additional
`text.
`
`“.
`
`R:EETTVS
`
`INSERT DN
`PQGE 1 LINE 9 COL ea
`>
`>
`>
`M E N U
`<
`(
`<
`M R I N
`-Other Menus-
`—Miscel1anenue— I
`——Cur5or Movement-7
`-Delete— 1
`1
`(from Main only)
`char
`left ‘D char right
`1’G
`char
`3 “I Tab
`"8 Reform 1
`ward left “F word right
`IDEL Chr 17}
`“V XNEERF DN/DFF
`l“J Help
`"K Block
`line
`up
`“X line down
`1’T wurd rtl“L Find/Replce again(‘Q Quick “P Print
`~~Scrol1ing—<
`t”V
`line
`lPETURN End paraqraph!‘D Dnsrreen
`line down ‘N line uD
`:
`:
`‘N Insert a RETURN !
`“C screen up ‘R screen down1
`i
`“U Stop a command
`.
`L~———I~———l———»l————l—— —[44-_[____[___V|___«g___—g—»W—t —————— —~R
`Fouvscure
`and seven years ago our fathers brought forth on
`this continent a new nation Conceiven in liberty and dedizated to
`the propaeition
`that all men are created equat.
`Now we
`are
`engaged in a great Civil war
`testing whether
`that nation,
`or any
`nation so conceived and so dedirated, can long endure.
`
`we have come
`We are met on a great battlefield of that war.
`LO dedicate a portion of
`that field as a final FeSt1flg‘DlaCE for
`those who here gave their llVES that
`that nation might
`live.
`
` ,i
`
`Figure 4.7: WordStar offers the user the option of bringing a help menu to a
`poition of the screen while the task is
`in process.
`
`

`
`164
`
`Designing the User Interface
`
`interactive systems on personal computers have another still
`Several
`more attractive form of prompts called command menus. Users are
`shown a list of descriptive words and make a selection by pressing the
`left and right arrow keys to_ move a light bar. When the desired
`command word is highlighted, the user presses the return key to carry out
`the command- Often, the command menu is a hierarchical structure that
`branches to a second— or third-level menu.
`Even though arrow key movement is relatively slow and less preferred
`by frequent users, command menu items can be selected by singleyletter
`keypresses. This strategy becomes a hierarchical command language, but
`it
`is identical
`to the typeahead (BLT) approach of menu selection.
`Novice users can use the arrow keys to highlight their choice or type
`single letter choices, but frequent users don’t even look at the menus as
`they type 2, 3, 4, or longer sequences of single letters that come to be
`thought of as a command (see FinalWord commands in Figure 4.5).
`The Lotus 1-2-3 (Figure 4.8) implementation is especially fast and
`elegant. As command words are selected, a brief description appears on
`the line below, providing further assistance for novice users without
`distracting experts from their concentration on the task. Experienced
`users appear to work as fast as
`touch typists, making three to six
`keystrokes per second.
`
`___:_%
`Hcrksheat Hangs Copy Muve File Print Graph Data Quit
`L.
`uuary Distribute
`sart
`Fill Tabla
`x A E c n E r Reset View Save uptians Name Quit
`
`Type
`Print File
`
`netreivr. save. Camblney Ktvacl. Eraee. Lust. 1mport- Direcln-Y
`(Have a cell ur range of (2115) Enter Pange FHDN:
`(Copy a call or range of calls) Enter range FROM:
`Farmat. Label—Prefxx, grass. Name; Justify» Precast. unprotect.
`Blflbaly Insert. Delete; Culumn-Width; Erase. Titles, Window. Status
`
`Figure 4.8: The first two levels of command menus from LOTUS 1~2—3 reveal
`the rich function available to users. At the third level, users may receive another
`menu or enter values.
`
`Input
`
`

`
`Ch. 4 Command Languages
`
`165
`
`Pop-up or pull-down menus that use mouse selection are another form
`of command menu. Frequent users can be extremely fast, and novices
`can take the time to read the choices before selecting a command. With
`a fast display, command menus blur the boundaries between what
`is
`thought of as commands and menus.
`
`4.7 NATURAL LANGUAGE INTERACTION
`
`Even before there were computers, people dreamed about creating
`machines that would accept natural language.
`It is a wonderful fantasy,
`and the success of word manipulation devices such as tape recorders,
`word
`processors,
`printing
`presses,
`and
`telephones may
`give
`encouragement to some people. A recurring hope is that computers will
`respond to commands issued by typing or speaking in natural language.
`Natural language interaction (NLI) might be defined as the operation of
`computers by "people using a familiar natural langauge (such as English)
`to give instructions. They do not have to learn a ‘command syntax nor
`select from menus.
`
`The problem with NLI is not only implementation on the computer, but
`also desirability for large numbers of users for a wide variety of tasks.
`People are different from computers, and human—human interaction is not
`necessarily an appropriate model
`for human operation of computers.
`Since computers can display information ’1,000 times faster than people
`can enter commands,
`it seems advantageous to use the computer to
`display large amounts of information and allow novice and intermittent
`users simply to choose among the items. Selection helps guide the user
`by making clear what functions are available. For knowledgeable and
`frequent users, who are thoroughly aware of the available functions, a
`concise command language is usually preferred.
`The syntactic/semantic model helps sort out the issues. NLI does not
`provide information about actions and objects in the task domain; users
`are usually presented with a simple prompt that invites a natural language
`query. But assume that the user is knowledgeable about the task domain;
`for example,
`the meaning of database objects and permissible actions.
`
`

`
`166
`
`Designing the User Interface
`
`the computer
`Neither does NLI necessarily convey knowledge of
`concepts; for example,
`tree-structuring of information,
`implications of a
`deletion, boolean operations, or query strategies. NLI does relieve the
`user of learning new syntactic rules, since it presumably will accept
`familiar English language requests. Therefore, NLI can be effective for
`the user who is knowledgeable about some task domain and computer
`concepts but who is an intermittent user who cannot retain the syntactic
`details.
`
`NLI might apply to checkbook maintenance (Shneiderman, 1980)
`where the users recognize that there is an ascending sequence of integer
`numbered checks, and that each check has a ‘single payee field, single
`amount, single date, and one or more signatures. Checks can be issued,
`voided, searched, and printed.
`In fact, following this suggestion, Ford
`( 1981) created and tested an NLI system for this purpose. Subjects were
`paid to maintain their checkbook registers by computer using an
`APL—based program that was
`incrementally refined to account
`for
`unanticipated entries. The final system successfully handled 91 percent
`of users’ requests, such as:
`
`Pay to Safeway on 3/24/86 $29.77.
`June 10 $33.00 to Cindy Lauper.
`
`Show me all the checks paid to Ronald Reagan.
`Which checks were written on October 29?
`
`Users reported satisfaction with the system and were eager to use the
`system even when the several months of experimentation were completed.
`This can be seen as a success for NLI, but alternatives might be even
`more attractive. _ Showing a full screen of checkbook entries with a blank
`line for new entries might accomplish most tasks without any commands
`and minimal typing. Searches could be accomplished by entering partial
`information (for example, Ronald Reagan in the payee field) and then
`pressing a query key.
`
`There have been numerous informal tests of NLI systems, but only a
`few have been experimental comparisons against some other design. A
`simulated query system was used to compare a subset of the structured
`SQL database facility to a natural
`language system (Small & Weldon,
`1983).
`The SQL simulation resulted in faster performance on a
`
`

`
`Ch. 4 Command Languages
`
`167
`
`benchmark set of tasks. Similarly, a field trial with a real system, users,
`and queries pointed to the advantages of SQL over the natural language
`alternative (Jarke .et al., 1985). Researchers seeking to demonstrate the
`advantage of NLI over command language and menu approaches for
`creating business
`graphics were
`surprised to
`find no
`significant
`differencesfor time, errors, or attitude (Hauptmann & Green, 1983).
`
`system
`research and
`in NLl may claim that more
`Believers
`is needed before excluding NLI, but
`improvements
`in
`development
`menus, command languages, and direct manipulation seem equally likely.
`Supporters of NLI can point with some pride at the modest success of the
`commercially available INTELLECT system that has approximately 300
`installations on large mainframe computers (Figure 4.9).
`Business executives,
`salespeople, and others use INTELLECT to
`search databases on a regular basis. Several innovative implementation
`ideas help to make INTELLECT successful. The parser uses the contents
`of the database to parse queries; for example,
`the parser can determine
`that a query containing Cleveland refers
`to _city locations because
`Cleveland is an instance in the database. Next, the system administrator
`can conveniently include guidance for handling domain—specific requests,
`by indicating fields related to who, what, where, when, how, etc.
`queries. Third, INTELLECT rephrases the user’s query and displays a
`response such as: PRINT THE CHECK NUMBERS WITH PAYEE =
`RONALD REAGAN. This structured response serves as ‘an educational
`aid,
`and users gravitate toward expressions
`that mimic the style.
`Eventually, as users become more frequent,
`they often use concise
`commandlike expressions that they believe will be successfully parsed.
`Even the promoters of INTELLECT recognize that novice users who are
`unfamiliar with the task domain will have a difficult time, and that the
`ideal user is a knowledgeable intermittent user.
`There is some portion of the user spectrum that can benefit from NLI,
`but it may not be as large as promoters believe. Computer users usually
`seek predictable responses and are discouraged if they must frequently
`engage in clarification dialog. Since NLI has such varied forms,
`the
`users must constantly be aware of what the computer’s response is to
`verify that
`the users’
`intentions were recognized.
`Finally, visually
`oriented interactions, embracing "the notions of direct manipulation (see
`Chapter 5), make more effective use of the computer’s capacity for rapid
`
`

`
`Designing the User Interface
`
`.
`PLEASE ENTER VUUR FIRST REQUEST
`
`KBEIf§,lN_IH§_EEI6§6§EZ _
`FIELDS IN THE FILE OF EHPLDVEES
`NAME
`JOE
`AGE
`FhMxLv
`EMPLDVEE NUMBER
`
`SALARY
`CITV
`
`.
`
`!EQ_E9BE§.£N-NEE_!QEE_ElIX
`PRINT THE JOB AND NAME or ALL EMPLDVEES WITH CITY = NEW vunx
`
`PROFESSIONAL
`OCCUPATION
`HACHINIST
`PHYSICIAN
`
`NAHE
`HNGELJN
`ANGUS
`
`EH61_i§-IuE,NHflEEB_QEE6B§u§
`couwr THE EMPLOYEES NITH NAME = ANGUS
`ANSNER:
`1
`
`,EEEEQXEE,EQflE§E;9E_6EE_EUELQ §§§_fllIE_NfiUE_E_fiN§Q§
`
`5591 L
`EMPLDV
`NUMBER
`0034
`
`.
`_
`WULEH_HQEEN_!QBK_lN_BEH-!QBE_ElI!
`FRINT THE JOB or ALL EMPLOYEES uTTH sEx = FEMALE L CITY = NEN YORK
`bnnFEs§roMAL
`UCEUFATIUN
`MACHINIST
`ADMINISTRATOR
`ELECTRICIAN
`EEUe£§_§EELQ!EE§-!H_N§!_!QBE_§£I!
`PRINT THE NAME OF ALL EMFLDVEES NITH SEX = FEMALE 5 CITY = NEW VDRK
`NAME
`ANGELIN
`CANTEEURV
`EDUARD5
`
`H
`H9H_!ed!_flE!-2985-911!-u§B_e5§_Q!§B_§9_!§eE§,DLD
`COUNT THE EMPLDVEE5 HITH CITY = NEW VURK 5 sEx = MALE L AGE>3O
`ANSWER:
`9
`EBLNI-IB§E
`PRINT THE NAME AND AGE OF ALL EMPLOYEES NITH crrv = NEH VURK a
`SEX = MALE K AGE>30
`YEARS
`0:
`AGE
`ea
`69
`HA
`
`.
`
`»
`
`NAME
`
`nweus
`ELEV
`HILTON
`
`1
`
`.
`
`Intelligence
`from Artificial
`Figure 4.9: Demonstration session with Intellect
`Corp., Cambridge, MA. User
`input
`is underscored.
`Intellect rephrases user
`input into a structured query language, which users often mimic as they become
`more frequent users.
`
`is often‘ more
`ln short, pointing and selecting in context
`display.
`attractive than typing or even speaking an English sentence.
`It
`is surprising that designers of expert systems have attempted to
`embed NLI. Expert systems already tax the user with complexity,
`lack
`
`

`
`Ch. 4 Command Languages
`
`169
`
`of visibility of the underlying processes, and confusion about what
`functions the system ca_nand_ cannot handle, A precise, concise notation
`or selection in context from a list of alternatives seems far more suitable
`in providing users with predictable
`and comprehensible behavior
`(Hayes-Roth, 1984) (Figure 4110).
`'
`‘
`
`An innovative blend of NLI and menus was developed under the name
`NLMENU (Tennant et a1., 1983) and is now distributed by Texas
`
`SAMPLE SESSION FROM AN EXPERT SYSTEM FOR OIL DRTLLING ADVISOR
`
`What is the name of WEL_L—159?
`AGFT—93E
`
`What is the profile of AG_FT—93E?
`DEVIATED
`
`Please enter information about FQRMATION—l:
`I_1pper—1imit
`lower-limit
`main-rock—
`meters
`meters
`type
`747
`806
`SHALE
`
`homogeneous/
`interbedded
`HOMOGENEOUS
`
`Please enter infonnation on PROBLEM—1:
`problem-type
`prior—action
`tota1—depth
`STICKING
`REAMING
`1111 METERS
`
`casing-shoe depth
`747 METERS
`
`Please enter the composition of the drill-string starting from
`the bit (type ? for assistance):
`BIT 9"5/8 STAPB”5/8 S1-_1ORTDC7"_3/4STAB9”5/S...NDP5
`
`What was the drilling methodlemployed when the problem occurred:
`ROTARY
`'
`
`What is the depth of the freepoint?
`UNKNOWN
`'
`
`Figure 4.10: This extract demonstrates one designers attempt at an expert system
`dialog. User input is shown in all upper case letters. Users must type in values
`even when selection from a menu would be 4 more meaningful,
`rapid, and
`error-free. Furthermore,
`there does not appear to be any way to go back and
`Change values, view values, or
`reuse values
`from previous Sessions
`(F-
`Hayes.-Roth, The knowledge-based expert system: A tutorial, IEEE Computer 17,
`9 (Sept. 1984), 11-28. © 1984 IEEE)
`
`

`
`170
`
`Designing the User Interface
`
`Find the
`
`.
`
`:
`
`sauna;
`tea.5:
`that are listed on
`birthaate
`that have
`data filled
`that list
`.
`datehired
`‘-sad the average that request
`maximum that were performed on
`minimum that were requested for
`that were turned in for
`who have
`
`between
`?
`x
`
`.
`[-lTTR_S:_
`(specific Jubcard .101)
`datts)
`__ -
`Enter jobcard job date : march 1515:]
`
`Jcbca
`apgra
`or er
`PIECE
`uerke
`
`F3-Pubout
`
`F8-Restart
`
`RET-Select
`
`ENT-Proceed
`
`langauge
`(TM) allows users to specify natural
`Figure 4.11: The NaturalLinlt
`English queries against a database by choosing phrases from a set of menus.
`The content of the menus is determined by the contents of the database.
`(Counesy of Texas Instruments, Dallas, TX)
`
`Instruments under the name Natura1Link (Figure 4.11). Natural language
`phrases are shown as a series of menus. As phrases (for example, FIND
`/ COLOR / AND / NAME OF‘ PARTS / WHOSE‘ COLOR IS) are
`chosen by a pointing strategy, a query is formed in a command window.
`Users receive information from the menus, obviating the need for a
`query. For example,
`if the parts and suppliers database contains only
`red, green, and blue parts, only these choices appear in the window
`containing the PART COLOR menu. Users can see the full range of
`possible queries ‘ and thereby avoid the frustration of probing the
`boundaries of
`system functionality. With this
`strategy,
`typing is
`eliminated and the user is guaranteed a semantically and syntactically
`correct query.
`
`A notable and widespread success of NLI techniques is in the variety
`of adventure games (Figure 4.12). Users may indicate directions of
`
`

`
`Ch. 4 Command Languages
`
`171
`
`Heleage to Oz, Dovathg. You uill_Jfii"
`the Tln Hoodsnan,
`the Cowardly L
`HE,
`the Scarecrow as we travel
`through
`Egg Marvelous L§n& of 92 in sea¥eh_?§eP
`we £5i'§§a§h3"i14i'£$33t3i'1Jiz§§fl' bgnailg‘ ta
`return to your Hunt EM and Uncle HenP9
`In Kansas.
`
`la NIZflRD-
`Press G to
`Press C for H zn D cpedlts.
`Press P for TREQSHRE PPeU19U-
`
`epen doom
`‘ii:
`‘
`1 the door.
`
`5 E
`
`gg iugk an? over a rollin hillside.
`r1ch_w1€$
`}%shes,
`trees an
`flouerg
`swaying 1n the earn breeze. There 15
`senethxnfl strange and Marvelous about
`this Ian
`Q stung ath leads south from
`your doorstep to a uhhling hrgok.
`.
`Pa;r pf feet shod in s1luer sl1PP8F5 15
`stlckxng out from underneath the house.
`take s11ppers_
`
`Figure 4.12: This adventure game is modelled on the Wizard of 0; story. The
`user types phrases such as “open the door” or “take slippers" or abbreviations
`such as “S” to move south. More complex phrases such as "put the hat on the
`scarecrow” are possible.
`(Courtesy of Spinnaker Software, Cambridge, MA)
`
`

`
`Designing the User Interface
`
`movement or type commands, such as TAKE ALL OF‘ THE KEYS,
`OPEN THE GATE, or DROP THE CAGE AND PICK UP THE
`SWORD. Part of the attraction of NLI in this situation is that the system
`is unpredictable and some exploration is necessary to discover the proper
`incantation .
`
`So much research has been invested in NLI systems that undoubtedly
`some successes will emerge, but widespread use may not develop because
`the alternatives may be more appealing. More rapid progress can be
`made if carefully controlled experimental tests are used to discover the
`designs, users, and tasks for which NLI is most beneficial.
`
`4.8 PRACTITIONER’S SUMMARY
`
`Command languages can be attractive when frequent use of a system is
`anticipated, users are knowledgeable about the task domain and computer
`concepts, screen space is at a premium, response time and display rates
`are slow, and numerous functions that can be combined in many ways are
`supported. Users will have to learn the semantics and syntax, but they
`can initiate rather
`than respond,
`rapidly specifying actions involving
`several objects and options. Finally, complex sequences of commands
`can be easily specified and stored for future use as a macro.
`
`Designers should begin with a careful task analysis to determine what
`functions should be provided. Hierarchical strategies and congruent
`structures facilitate learning, problem solving, and human retention over
`time. Laying out the full set of commands on a single sheet of paper
`helps show the structure to the designer and to the learner. Meaningful
`specific names
`aid learning and retention.
`Compact abbreviations
`constructed according to a consistent rule facilitate retention and rapid
`performance for frequent users.
`
`Innovative strategies, such as command menus, can be effective if
`rapid response to screen actions can be provided. Natural
`language
`interaction can be
`implemented, but
`its
`advantage for widespread
`application is yet to be demonstrated.
`
`

`
`Ch. 4 Command Languages
`
`4.9 RESEARCHER’S AGENDA
`
`task
`for
`Designers could be helped by development of strategies
`analysis, taxonomies of command language designs, and criteria for using
`commands or other techniques. The benefits of structuring such concepts
`as hierarchicalness, congruence, consistency, and mnemonicity have been
`demonstrated in specific cases, but
`replication in varied situations is
`important. Experimental
`testing should lead to a more comprehensive
`cognitive model of command language learning and use.
`(See Table
`4.3.)
`
`tool for
`A command language system generator would be a useful
`research and development of new command languages. The designer
`
`COMMAND LANGUAGE GUIDELINES
`
`Create explicit model of objects and actions
`
`Choose meaningful, specific, distinctive names
`
`Try for hierarichical structure
`
`Provide consistent structure
`
`(hierarchy, argument order, action-object)
`
`Support consistent abbreviation rules
`(prefer truncation to one letter)
`
`Offer frequent users the capability to create macros
`
`Consider command menus on high-speed displays
`
`Limit number of commands and ways of accomplishing a task
`
`Table 4.3: High-level design guidelines based on empirical studies and practical
`experience.
`
`

`
`Designing the User Interface
`
`could provide a formal specification of the command language, and the
`system would generate an interpreter. With experience in using such a
`tool, design analyzers might be built
`to critique the design, detect
`ambiguity, check for consistency, verify completeness, predict error rates,
`or suggest
`improvements. Even a simple but
`thorough checklist
`for
`command language designers would be a useful contribution.
`Novel input devices and high—speed, high-resolution displays offer new
`opportunities, such as command and pop-up menus, for breaking free
`from the traditional syntax of command languages. Natural
`language
`interaction still holds promise in certain applications, and empirical tests
`offer a good chance to identify rapidly the appropriate niches and design
`strategies.
`
`REFERENCES
`
`Barnard, P. J ., and Hammond, N. V., Cognitive contexts and interactive
`communication,
`IBM Hursley (U.K.) Human Factors Laboratory
`Report HF070, (December 1982), 18 pages.
`Barnard, P., Hammond, N ., MacLean, A., and Morton, J ., Learning and
`remembering interactive commands, Proc. Conference on Human
`Factors in Computer Systems, Available from ACM DC., (1982), 2—7.
`
`Barnard, P. J., Hammond, N. V., Morton, J., Long, J. B., and Clark, 1.
`A., Consistency and compatibility in human-computer dialogue,
`International Journal of Man-Machine Studies 15 , (1981), 87-134.
`
`Izak, and Wand, Yair, Command abbreviation behavior in
`Benbasat,
`human-computer
`interaction, Communications of the ACM 27, 4,
`(April 1984), 376-383.
`
`Black, J ., and Moran, T., Learning and remembering command names,
`Proc. Conference on Human Factors in Computer Systems, Available
`from ACM DC., (1982), 8-11.
`
`Carroll, John M., Learning, using and designing command paradigms’,
`Human Learning 1, 1, (1982), 31-62.
`
`

`
`Ch. 4 Command Languages
`
`cognitive
`the
`and
`J., Metaphor
`and Thomas,
`J. M.,
`Carroll,
`representation of compnting systems, IEEE Transactions on Systems,
`Man, and Cybernetics, SM(.-‘-172, 2, (March/April 1982)‘, 107-115.
`Ehrenreich, S. L., and Porcu, Theodora, Abbreivations for automated
`systems: Teaching operators and rules, In Badre, Al, and Shneiderman,
`Ben’,
`(Editors), Directions in Iiluman-Computer Interaction, Ablex
`Publishers, Norwood, NJ, (1982), 111-136.
`Ford, W. Randolph, Natural Language Processing by Computer —A
`New Approach,-Ph. D. Dissertation, The Johns Hopkins University
`Department of Psychology, Baltimore, MD, (1981), 88 pages.
`'
`Green, T. R. G., and Payne, S. J., Organization and learnability in
`computer‘ languages, International Journal of Man-Machine Studies 21,
`(1984), 7418.
`
`Grudin, Jonathan, and Barnard, Phil, When does an abbreviation become
`a word and related questions, Proc. CHI ’85 Conference on Human
`Factors in Computer Systems, Available from ACM'Order Dept., P.
`Q. Box 64145, Baltimore, MD 21264, (1985), 121-126.
`'
`Hanson, Stephen 1., Kraut, Robert E., and Farber, James M., Interface
`design and‘ multivariate ‘analysis of UNIX command use, ACM
`Transactions on Oflice Information Systems 2,
`1,
`(January 1984),
`42-57.
`’
`'
`'
`
`and Green, Bert F., A comparison of
`Hauptmann, Alexander G.,
`command, menu-selection and natural
`language computerlprogramis,
`Behaviottr and Information Technology 2, 2, (1983), 163-178.
`Hayes—Roth, Frederick, The knowledge-based expert system: a tutorial,
`IEEE Computer 17; 9, (September 1934), 11.—23.
`.
`'
`Jarke, Matthias, Turner, Jon(A., Stohr, Edward A., Vassiliou, Yannis,
`White, Nonnan H., and Michielsen, Ken, A field evaluation of -natural
`language
`for
`data
`retrieval,
`IEEE Transactions
`on Software
`Engineering SE-I1, 1, (January 1985), 97—113.
`Kraut, Robert E., Hanson, Stephen J ., and Farber, James, M., Command
`use and interface design, Proc. CHI '83 Conference on Human Factors
`in Computing‘ Systems, Available from ACM Order Dept., P. O.‘ Box
`64145, Baltimore, MD 21264, (1983), 120-123.
`Landauer, T. K., Calotti, K. M.-, and Hartyvell, 8., Natural command
`
`

`
`176
`
`Designing the User Interface
`
`names and initial learning, Communications of the ACM 26, 7, (July
`1983), 495-503.
`.
`
`Ledgard, H., Whiteside, J. A., Singer, A., and Seymour, W., The
`natural language of interactive systems, Communications of the ACM
`23, (1980), 556-563.
`'
`
`Norman, Donald, The trouble with UNIX, Datamation 27,
`556-563.
`’
`'
`
`(1981),
`
`Roberts, Terry, Evaluation of computer text editors, Ph. D. dissertation,
`Stanford University. Available from University Microfilms, Ann
`Arbor, MI, order number AAD 80-11699, (1980).
`Rosenberg, Jarrett, Evaluating the suggestiveness of command names,
`Behaviour and Information Technology I , (19822), 371-400.
`
`Rosson, Mary Beth, Patterns of experience in text editing, Proc. CHI
`’83 Conference on Human Factors in Computing Systems, Available
`from ACM Order Dept., F. O. Box 64145, Baltimore, MD 21264,
`(1983), 171-175.
`
`Scapin, Dominique L., Computer commands labelled by users versus
`imposed commands and the effect of structuring rules on recall, Proc.
`Conference on Human Factors in Computer Systems, Available from
`ACM DC, (1982), 17-19.
`
`Schneider, M. L., Ergonomic considerations in the design of text editors,
`In Vassiliou, Y.
`(Editor), Human Factors and Interactive Computer
`
`Systems, Ablex Publishers, Norwood, NJ, (1984), 141-161.
`Schneider, M. L., Hirsh—Pasek, K., and Nudelman, S., An experimental
`
`evaluation of delimiters in a command language syntax, International
`Journal of Man-.Machine Studies 20, 6, (June 1984), 521-536.
`
`Shneiderman, Ben, Software Psychology: Human Factors in Computer
`and Information Systems, Little, Brown and Co., Boston, MA, (1980),
`320 pages.
`'
`A
`
`Small, Duane, and Weldon, Linda, An experimental comparison of
`natural and structured query languages, Human Factors 25, (1983),
`253-263.
`
`Tennant, Harry R., Ross, Kenneth M., and Thompson, Craig W., Usable
`natural
`language interfaces
`through menu-based natural
`language
`
`

`
`Ch. 4 Command Languages
`
`’83 Conference on Human Factors’ in
`understanding, Proc. CHI
`Computing Systems, available from ACM Order Dept., F. O. Box
`64145, Baltimore, MD 21264 (1983), 154-160.
`
`

`
`CHAPTER 5
`
`DIRECT MANIPULATION
`
`“In
`to make the form of a symbol reflect its content.
`Leibniz sought
`signs,” he wrote, “one sees an advantage for discovery that is greatest
`when they express the exact nature of a thing briefly and, as it were,
`picture it; then, indeed, the labor of thought is wonderfully diminished.”
`
`Frederick Kreiling, “Leibniz,” Scientific American, May 1968.
`
`

`
`CHAPTER 5
`
`DIRECT MANIPULATION
`
`“In
`its content.
`to make the form of a symbol reflect
`Leibniz sought
`signs,” he wrote, “one sees an advantage for discovery that is greatest
`when they express the exact nature of a thing briefly and, as it were,
`picture it; then, indeed, the labor of thought is wonderfully diminished.”
`
`Frederick Kreiling, “Leibniz,” Scientific American, May 1968.
`
`

`
`180
`
`Designing the User Interface
`
`5.1.
`
`INTRODUCTION
`
`Certain interactive systems generate a glowing enthusiasm among users
`that is in marked contrast with the more common reaction of grudging
`
`acceptance or outright hostility. The enthusiastic users’ reports are filled
`with the positive feelings of:
`
`mastery of the system
`competence in “performance of their task
`ease in learning the system originally and in assimilating
`advanced features
`
`confidence in their capacity to retain mastery over time
`
`enjoyment in using the system
`
`-
`-
`
`eagerness to show it off to novices
`desire to explore more powerful aspects of the system
`
`These feelings are not universal, but this amalgam is meant to convey
`an image of the truly pleased user. The central
`ideas seem to be
`visibility of
`the "objects
`and actions of
`interest,
`rapid reversible
`incremental actions, and replacement of complex command language
`syntax by direct manipulation of the object of interest—hence,
`the term
`direct manipulation.
`
`5.2 EXAMPLES OF DIRECT MANIPULATION
`
`SYSTEMS
`
`No single system has all the admirable attributes or design features—
`that may be impossible; but each of the following examples has enough
`to win the enthusiastic support of many users.
`
`My favorite example of direct manipulation is driving an- automobile.
`The scene is directly visible through the front window, and actions such
`as braking or steering have become common knowledge in our culture.
`To turn left, the driver simply rotates the steering wheel to the left. The
`response is immmediate and the scene changes,. providing feedback to
`
`

`
`Ch. 5 Direct Mampulatton
`
`181
`
`Imagine trying to turn by issuing a command LEFT 30
`refine the turn.
`DEGREES and then having to issue another command to see the new
`scene; but this is the level of operation of many office automation tools
`of today.
`
`5.2.1 Display editors
`
`Users of full—page display editors are great advocates of their systems
`as compared with line—oriented text editors (Figure 5.1). A typical
`comment was, “Once you’ve used a display editor you will never want to
`go back to a line editor—you’ll be spoiled.” Similar comments came
`from users of stand—alone word processors such as the WANG system,
`personal
`computer word processors
`such
`as WORDSTAR 2000,
`
`(Courtesy of © Lotus
`from Symphony.
`processor
`5.1: Word
`Figure
`Development Corporation 1985. Used with permission.)
`
`

`
`182
`
`Designing the User lntertace
`
`FINALWORD, XYWRITE, and Microsoft WORD, or display editors
`such as EMACS on the MIT/Honeywell MULTICS system or “vi” (for
`visual editor) on the UNIX system. A beaming advocate called EMACS
`“the one true editor.”
`
`found overall performance times with line—oriented
`Roberts (1980)
`editors were twice as long as with display editors. Training time with
`display editors is also reduced,
`so there is evidence to support
`the
`enthusiasm of display editor devotees. Furthermore, office automation
`evaluations consistently favor fu11—page display editors for secretarial and
`executive use.
`
`The advantages of display editors include:
`
`,
`display of a full 24 to 66 lines of text.
`This gives the reader a clearer sense of context for each
`sentence while permitting simpler reading and scanning of
`the document. By contrast,
`the one—line—at—a—time view
`offered by some line editors is like seeing the world through
`a narrow cardboard tube.
`
`display of the document in the form that it will appear when
`the final printing is done.
`also
`formatting commands
`Eliminating the
`clutter of
`simplifies reading and scanning of the document. Tables,
`lists, page breaks, skipped lines, section headings, centered
`
`V
`
`text, and figures can be viewed in their final form. This
`style has come to be known as WYSIWYG (what you see
`is what you get). The annoyance and delay of debugging
`the format commands are eliminated because the errors are
`
`immediately apparent.
`cursor action that is visible to the user.
`
`Seeing an arrow, underscore, or blinking box on the screen
`gives the operator a clear sense of where to focus attention
`and apply action.
`
`cursor motion through physically obvious and intuitively
`natural means.
`
`Arrow keys or cursor motion devices such as a mouse,
`joystick,
`or
`graphic
`tablet
`provide
`natural
`physical
`
`

`
`Ch. 5 Direct Manipulation
`
`183
`
`in marked
`mechanisms for moving the cursor. This is
`require an
`contrast
`to commands
`such as UP
`6_
`that
`operator
`to convert
`the physical action into a" correct
`syntactic form that may be difficult to learn, hard to recall,
`and a source of frustrating errors.
`
`labeled buttons for actions.
`Many workstations designed for use with display editors
`have buttons with actions etched onto them,
`such as
`INSERT,
`' DELETE,
`CENTER,
`UNDERLINE,
`SUPERSCRIPT, BOLD, or LOCATE. These buttons act
`
`a permanent menu selection display to remind the
`as
`operator of the features and to avoid the need to memorize
`a complex command language syntax. On some editors,
`only ten or
`fifteen labeled buttons provide the basic
`functionality. A specially marked button may be
`the
`gateway to the world of advanced or
`infrequently used
`features that are offered on the screen in menu form.
`
`immediate display of the results of an action.
`When a button is pressed to move the cursor or center text,
`the results are shown immediately on the screen. Deletions
`are immediately apparent since the character, word, or line
`is erased and the remaining text is rearranged. Similarly,
`insertions or text movements are shown after each keystroke
`or function button press. This is in contrast to line editors
`in which print or display commands must be issued to see
`the results of changes.
`
`rapid action and display.
`Most display edito