not - みる会図書館


検索対象: Interface Age 1980年10月号
80件見つかりました。

1. Interface Age 1980年10月号

Learnmg with M ー 05 By Louis E. Frenzel, 」 r. The Home Computer and Education The idea Of a true home consumer computer must be at least five years 0 旧 . The original machines like the Altair,lmsai and the SWTP 6800 were hObby computers, and about as far from the home computer as the early DEC minis. The second generation computers came close 「 tO home. These were the Radio Shack T 日 S -80 , Commodore PET, and the AppIe ル These are enormously popular, but are not really consumer computers. However, manufacturers are getting closer. The Atari 400 and 800 , Tl's 99 / 4 , Mattel's lntellivision and the new APF are approaching what many believe tO be a real home computer. Where are the buyers? But while the hardware is here, consumers are simply not buying. The 引 99 / 4 is not selling well because Of its high price, despite discounting and rebates. Atari is just ramping up production, and Mattelis still experiencing introduction delays Of its keyboard unit. Numerous tests have been made by lnteract and OhiO Sci- entific in department stores with mixed results. WhiIe such tests have not been tOtal failures, neither have they been truly successful 0 「 conclusive. Technology has made avail- able a general PlffPOSe home computer, but the public either does not fully understand 0 「 has no real need or interest in it. Why? Price is a major factor. MOSt home computers still sell in the $ 600 tO $ 1 , 500 price range—too much for the average consumer. は may a 0 be that the things one can dO with a computer are things many people don't care tO dO. Game playing is a main application. But most Of the games are pretty trivial and, after the novelty wears Off, can get pretty boring. Keeping your checkbook, Christmas card list, family budget and calendar on a computer is certainly practical. But it takes time tO accumulate, record and enter all Of the data required. The average person doesn't have the time, pa- tience or perseverence When it iS a lOt faster and easier tO jOt notes in a checkbook, and use a $ 10 pocket calculator tO d0 banking calculations. Ordinary 3X5 cards 0 「 a spiral notebook are adequate fO 「 much recordkeeping. What it bOils down tO is people still need a stronger reason tO buy a home computer. A possible opportunity Perhaps that reason is education. MOSt manufacturers Of personal computers tout the educational value 0f their machines. But the potential goes far beyond teaching people hOW tO use computers. Using computer aided instruction (CAI), all types Of education can be presented. は may be that education is the one practial application that will entice con- sumers tO buy a home system. MOSt home computer manufacturers offer educational soft- ware. But most are extremely short and trivial programs fO 「 children: basic arithmetic, spelling, subjects that most kids 38 INTERFACE AGE learn in school with little difficulty. They certainly don't need a home computer. Some programs ー have seen are even t00 simple for children. Kids don't want tO come home from school and take a learning program in math 0 「 spelling.l think the home computer manufacturers are on the wrong track with their educational software. Forget the kid stuff and get on tO developing something that adults can use. There is a mini revolution going on in education these days. Adults are going back tO schoolin droves. They are seeking all kindS Of educational materials, sources and experiences. Colleges are opening up more evening programs and continu- ing education classes. MOSt Of this renewed interest is jOb or career oriented. PeopIe want tO dO their jObs more competently, prepare for advancement, 0 「 change careers. Doctors, lawyers, pharma- CiStS and engineers work hard at continuing education tO keep up with rapidly changing technology and society. AduIt education appeals tO the generalinterest, leisure time and hObby interests Of the public.ln fact,learning has become a life-long activity, not one that just terminates at the end 0f high school 0 「 college. The home computer should be a part of this. Manufacturers or software houses should develop CAI on subjects 0f inter- est tO adults. HOW about a learning program on the stock market? What about jOb イ elated knowledge and skills such as TechnoIogy has made available a ge れ e 「 purpose home computer, but the public either does 00t fully understand 0 「 has 00 real need 0 「 interest in it. management, salesmanship, and finance? What about prac- tical subjects such as lawn care, sailing, running, probability and statistics or business letter writing? These subjects and many more could be readily taught on the home computer. げ such a series Of teaching modules were available, the home computer would become one Of the hottest products around. The so ⅳ on SO what are my recommendations for solving this problem? Someone is going tO have tO Offer adult learning programs on practical subjects Of immediate interest and value. は is as simple as that. ー do not believe the manufacturers will be the ones tO dO this. MOSt are hardware design and manufacturing ori- ented. As a result, they don't know hOW t0 develop educa- tional material. Developing learning programs requires experts, instructional designers, as well as programmers. There are some individuals and small software houses willing and capable 0f tackling edu- cational softwa 「 e. One 0 「 more will eventually emerge as major developers. These should be traditional publishers. WhY don't the big textbook and AN publishers such as McGraw-Hill Wiley, Prentice-HaII get involved? They have the author re- sources and the expertise tO draw on. They 引 SO have the dis- tribution systems tO make such programs widely available. NO doubt some kind Of relationship between a manufacturer and publisher could result in some extremely interesting adult learning programs for home computers. This type Of educa- tion is what the home computer needs tO be successful. ロ OCTOBER 1980

2. Interface Age 1980年10月号

しこ凵こユ Court case legalizes stealing? ー was not surprised at the findings reported on in your JuIy Editor's Notebook. Having been tO Apple and T 日 S -80 computer clubs where the main attrac- tion was duplicating the latest version of Star Trek, and hearing that the Micro- soft Basic compiler would soon be avail- able from so-and-so at a reasonable five finger discount, nothing much surprises me, no 「 should most adults be surprised at finding out that immorality exists in the wo 月 d. When dealing with humans, two basic axioms should be observed: Paying nothing always costs less than paying something. lnnocence is equal tO not being caught. generall consider ethics codes a response tO the fact that immorality ex- ists; but codes without enforcement have little real meaning. Laws must be enforced tO be laws. は is easy tO forget that wars are won by attrition, not by good intentions and never by talk. 旧 trying tO enforce any law, you have tO first catch someone. And since a lot Of 、、 stealing" is done at computer club meetings and by groups Of friends and groups Of employees, the cost 0f find- ing the villain can be more than the product is worth. げー understand the recent Data Cash vs. J. S. &A. ruling, the courts have held that computer programming on magnetic media does not have the copyright pro- tection given tO printed material. SO we are getting close 「 tO a victimless SOCi- ety. You can't steal a program in the pub- lic domain because you already own it. 日 ona 旧 C. Wagener NorfoIk, VA Hearing things There are whispers around that someone is about tO hit the market with a cassette recorder for under $ 250 that has 8-track parallelinput.l assume that it is much like a tape reader, only much, much faster. Best Of 訓 it will offer approximately eight times the storage capacity per tape as a standard cassette recorder. は seems tO me that the electronics for parallell/O is a lot simpler than serial I/O, and therefore cheaper tO build. But whO makes a cheap 8-track head? Do any of your readers know whO is about tO put this recorder on the market? け so, would they share this infO with me? S. B. Wahlberg P. 0. Box 502 SiIverado, CA 92676 プ 0 INTERFACE AGE ScheduIing software needed WhO has T 日 S -80 software for PERT/ CPM scheduling? l'm to 旧 by experts that it's possible tO use a microcompu- ter for this purpose. lt's time that some- one came up with a low-cost computer package for use in project management. When lfind one,l'll give it exposure to the persons whO come tO my project management seminars. Make the contact for me, and 聞 put the show on the road. S. F. Love Advanced Professional DevelopmentInstitute 5519 Carpenter Ave. NO. HO 収 wood , CA 91607 Review provokes strong rebuke ー was quite upset tO find a favorable review Of Software lnterpreters fO 「 Microcomputers by Thomas C. Mclntire ()A May 1980 ). This is one of the worst books Of its kind thatl have read. The author is not a good programmer. He espouses using kinds 0f little tricks tO "save small amounts Of memory and tO "increase" execution speed at the expense Of producing incomprehens- ible code. He seems completely un- aware that execution speed iS far more sensitive tO the overall approach tO the problem than whether subroutines take a little more 0 「 a little less time. For ex- ample, in chapter 14 in the discussion Of execution speed, he ignores the fact that hash tables can give enormous in- creases in execution speed. He es- pouses packing tWO 4-bit counters in a byte Of memory, eight 1 -bit flags in a byte, and using one memory location fo 「 different things at different times, allin an attempt tO save memory space. Not on ツ does this trickery lead tO code that is extremely difficult t0 understand and modify, but in addition it creates the possibility that obscure bugs may cause occasional system crashes, and that these bugs will be virtually impossible tO find, as anyone whO has worked on pro- grams like this will testify. . . And Mcln- tire acknowledges that what he does is considered tO be bad programming practice, but he says, ・、は is considered tO be fundamentally correct here... that anything that works correctly is legiti- mate discipline. The programming practices espoused by Mclntire have been tried and proven tO be bad. は does not matter hOW fast a program is and hOW little space it occupies in the memory if it is not correct.l think that any programmer will agree that there are enough unreliable programs. We don't need bOOks encouraging people tO write mo 「 e. Some added notes on Zilog ー am grateful fO 「 the mention given my PDS by Alan Miller in his August article. However, a few points regarding the un- announced Z -80 instructions should be made: First, we should thank M 「 . D. 日 . Lunsford for making these 94 additional operations known tO us. Second, ー agree with Dr. Miller that software for distributjon should avoid the un- documented instructions. The individual user may favor these instructions if hiS processor has the capability.l have not found a Z -80 processor—Zilog or sec- ond source—which does not handle the extended instructions. Third, while it is regrettable that ZSID cannot handle these instructions, that fact cannot pre- clude their use when it may be advan- tageous. The 、 debug' component of PDS fully recognizes the enhanced in- struction set. AlIen AshIey Pasadena, CA Documentation criticized As lrecently received a Z -89 , the arti- cle by Tom FOX in August was of great interest.l fully agree that the hallmark Of the Z -89 is the terminal, and was in general agreement until he praised the documentation. 旧 the weeks thatl have worked with the system ー have found endless frustra- tions with the software documentation and sizable hOles in the hardware manual. Section 0 Of the HDOS manual is excellent for the beginner whether business man 0 「 hobbyist. After learn- ing tO use one-copy, the reader is forced intO section 1 which iS written as a reference for experienced program- mers. ー found sections 1 through 5 useless until the local Heath store gave me the name Of a systems analyst whO was gracious enough tO spend a Satur- day helping me. け Zenith is serious about marketing the Z -89 to businessmen, it must cookbook the manual more, supply much more interesting information and improve itS service. Robert C. Thompson Houston, TX OCTOBER 1980 Another annoyance is that the bOOk is set in a sans serif font that is impossible t0 read quickly or easily: the reader's eyes get lost on the page. But perhaps this was a good bOOk in which to make such an exception. Richard L. Miller Cambridge, MA

3. Interface Age 1980年10月号

謝旧 C れ『 L こ当こ Box 2654 CIinton,Iowa 52735 謝旧 C れ一しこ当こ Box 2654 CIinton,Iowa 52735 PLACE STAMP HERE The Post Office will not deliver mail without postage Magazine Magazine mail without postage will not deliver The Post Office HERE STAMP PLACE

4. Interface Age 1980年10月号

integrate hiS particular COdes intO SUbrOUtine and then recompile and integrate the subroutine. By these means, most programmers can employ the screen-manipulating codes that will permit application Of console input principles for their programs. Alter- native 4 has been coded in Basic for ADM, Hazeltine and Adds terminals. The expected response tO a program input prompt is Often ambiguous because Of unclear prompts or because Of user experience With inconsistent tives allowed by the computer program. MOSt Basics will hang up following an 'input' state- ment if い ) the entered material is not Of the type (numeric or string) specified by the computer program; ( 2 ) if the number Of separate entries does not corres- pond tO the number expected. An entry Of 'yes' when a numeric variable was specified by the program, 0 「 an entry Of only tWO out Of three required variables will in- G00d programs are written in a way that promotes user confidence voke the Basic monitor and inform the user that some input value seemed tO be illegal. GOOd programs are written in a way that promote user confidence. They dO not try imagination and patience during the interactions when jOb parameters are selected. The principle that corresponds tO this aspect Of person- computer interaction iS.• standardize user reponses. Request 加 p 砒 responses ″ om users by asking exp/icit questions and when possible by using standardized terms and formats. Ask fo 「 natural- /anguage (not code) responses and check them consistently. lnforming the user Of optionalresponses tO an input query such as '0kay tO proceed (y/n)?' is an easy part Of this practice. Other requirements for operating con- sistently with the principle are: 1 . always use (y/n) or some Other standard signal in the prompt; 2. never take a shortcut; and 3. always check the response tO the same extent each time it is input. Unfortunately, upper versus lower case and spelled versus abbreviated forms will be input unpredictably and must be anticipated by the programmer. lJsers that sometimes get away with non-standard responses will develop a mistrust Of the program. The programmer should therefore be thorough and absolutely consis- tent in checking input. Subroutines are frequently the best way tO handle these input check requirements, and tO provide specific feedback tO the user on de- tected input errors. Before discussing some examples Of user input checks, a word about user nervousness iS timely. は iS easy tO destroy a file 0 「 an entire disk 0 「 tape by dOing 、 'the wrong thing. '' This involves user input blunders, OCTOBER 1980 such as updating on a new file rather than a history file, or backing up a disk with the drives switched, thereby destroying the original disk. After all, computers work as efficiently against us as they dO fO 「 us. This principle tells us tO be explicit in questioning users tO help pre- vent these blunders. For example, where source and destination files are asked fO 「 the same input line, there should be a following statement and query such as, 'the disk on drive B will be copied ontO the drive A disk destroying A'S contents. Okay t0 proceed (y/n)?' Alternative check method Being explicit about tape or disk is usefulto the ex- tent that it relates tO the tape 0 「 disk Of interest and not tO the drives. We copy the magnetic medium, not the drive with a copy program, SO ask the question for which an answer is sought. These guideposts help users double-check their input and help tO somewhat reduce user nervousness. Checking numeric input can best be done by not re- questing a numeric variable at all, but by SUPPlYing a string variable tO the Basic input statement. This permits a tolerance check before the input is converted tO a nu- meric quantity. Another point is the standard use Of a cer- tain precision in decimal places. You might always omit decimals 0 「 always include a certain number Of places, just tO make their presence a routine matter for users. Break-out values are stopping points in a program and are always welcome when long lists are being in- put. They must be standardized and explicit rather than a matter Of guesswork. Popular choices for these values are 'bye,' 'stop,' 'finish,' ・ end,' 'subtotal.' Your choice should be clarified, Of course, and unerringly used in all interactions. 旧 the name Of consistence, never allow a 、 stop' tO be input in place Of 'bye' if the user has been informed that 'bye' is expected. TOler- ance for lower versus upper case should be programmed, because text input can lead tO an upper/lower pattern that starts each line in upper case. Care must be taken that the break-out word is not part Of the text, SO require that the length Of the input string be consistent with the length Of the break-out word before checking for a word-wise match against the upper/lower case break-out options. Programming will be more efficient if the check itself is staged, such as BYE, Bye and bye, 訓 having a string length Of three. 旧 cases when most input does match, but in some way the input is not acceptable, simply ask fO 「 clarification: 'was that bye (y/n)?' Finding input requests a 24 x 80 screen, there are over 1 900 possible places tO hunt fO 「 input requests. Admittedly, 0 可 Y 24 lines are used fO 「 the input request except in rather ex- Otic cases, but the standard screen principal further limits input tO the uppermost tWO lines and moves status 0 「 option information tO the screen area below these lines. This lets the user anticipate where the input prompt will happen tO appear and reduces clutter. A by-product Of this principle is that, because a reserved screen area iS set aside fO 「 options and status information, the programmer will tend tO display these usefulfacts more Often. Each request for user input, if not a series Of figures orlengthy text, should be displayed at the home loca- INTERFACE AGE 10 プ

5. Interface Age 1980年10月号

gation, they allowed it under the lnternal Revenue Code Sec- tion permitting "ordinary and necessary' business expense deduction. Among Other factors, ・ 'ordinary and necessary has been interpreted tO mean the acquisition Of an asset hav- ing a usefullife not exceeding one year. The ldaho taxpayer, concluded the tW0 courts, brought efficiency by purchasing the computer programs and held efficiency tO be currently deductible. Another recent case has suggested thät, where a Board of Directors elected not tO purchase a computer system tO make their business more COSt efficient Where such a system was available, it may be liable t0 its shareholder-owners for improper business judgement. For many corporate directors and officers, this glimpse Of futureshock may become a pre- sent reality. The pointl want tO make is that a company may be deemed negligent in not taking advantage Of a more cost efficient means tO operate. TO return tO the ldaho case,l have thus far presented the extreme opposing positions: amortization VS. current deducti- bility. there a middle ground? Yes, argued one judge; he argued that the 旧 S already allows software write-offs more quickly than over their usefullife by virtue oflnternal Revenue Service Revenue Procedure 69-21 which permits a software acquirer tO write-off his purchase over its usefullife, 60 months, or shorter if the taxpayer can show that the software will have a usefullife 0f less than 60 months in the taxpayer's trade or business. Strange, but why didn't the 旧 S choose tO attack taxpayer's current deductibility argument by this revenue pro- cedure? The same judge felt Other judges had erred when they sided in favor 0f allowing current deductibility. He compared the purchase Of the software by the taxpayer as the same as the turn-of-the-century purchase Of a player piano.l fail tO comprehend this line Of reasoning because a player piano made the tavern owner's business more profit- able by eliminating the piano player's salary in addition tO its novelty. NevertheIess, this judge failed tO strike a responsive chord with his colleagues because he was the minority dissenting opinion. More importantly, this opinion points out the vast technological gap between present business reality VS. archaic judicial analogies. TO summarize, where a business purchases programs and associated operating manuals tO establish or enhance its busi- ness services or products where immediate COSt efficiency iS not the taxpayer's primary objective, the chances Of obtain- ing current deductibility for its purchase decrease. TO date, there exists at least three major cases where software was permitted tO be currently deducted. AII fought and won against the positions taken by the 旧 S. The 旧 S will,l believe, con- tinue tO forum shop for another court which it forecasts will be more receptive tO its capitalization argument. ロ ] uRl,sppuöent computepl st¯ By E Ⅲ 0 廿 MacLennan Attorney at Law 質・ Tax DeductibiIity Of Computer Software A software purchaser iS sometimes a creature tO be pitied. Where he purchases software only, the federalinvestment tax credit is inapplicable because the 旧 S maintains that soft- ware is intangible property, and only tangible personal prop- erty qualifies. Contrariwise, many states have classified software, and not just its media, as tangible property trigger- ing the imposition Of sales tax, use tax, value-added taxes, personal property tax or some Other species concocted in the brain Of a creative tax administrator. AIthough an investment credit is unavailable, a deduction for software in the year purchased is, SO stated the court in a recent federal appellate case originating from an IdahO tax court case. Here the court held, and the federal appeals court agreed, that software is deductible in the year purchased as opposed tO being written Off over its usefullife, the latter position be- ing espoused by the 旧 S. The tWO positions, usefullife amor- tization vs. current deductibility clashed head-on with the tax- payer winning a clear victory. 旧 the ldaho case, the 旧 S argued that the purchase Of com- puter programs and accompanying operating manuals provided 0 可 y future economic benefits Of an unknown duration and, therefore, must be capitalized and ratably deducted or amor- tized over their usefullife. The 旧 S a 0 argued that by pur- chasing the software, the taxpayer corporation created a new and separate asset which must be capitalized, not currently deducted. The tax court and federal appellate court decided that the purchasers Of the software did not create a new and separ- ately identifiable asset but merely acquired a means t0 oper- ate its business in a more cost efficient way.ln addition tO cost efficiency, the tWO courts felt that software expenses would be a recurring charge tO taxpayer's business. ー believe that the decided opinions Of the tWO courts were correct because, by way Of example, the purchase by a busi: ness Of an application program designed tO centralize and CO- ordinate inventory iS not the creation Of a new and separate asset. Rather, it is the replacement and upgrading Of a human inventory control system. Will software charges fO 「 a com- puterized control system be recurring? That depends on a host Of severalfactors unique tO a given business application. it not reasonable tO assume that modification Of software is mandated when the manner Of physicalinventory storage and shipment is altered due tO market conditions, warehouse space availability and the like? Perhaps it would be simpler tO generalize and say that the better acquainted a business gets with its computerization, the more customized it will seek tO make its software. Software iS routine expense When the two courts allowed the software to be currently deducted in the year Of its acquisition by the taxpayer in liti- OCTOBER 1980 'NOW, about this Debugging Kit you sent me. INTERFACE AGE プ /

6. Interface Age 1980年10月号

THE M ℃ RO ・ 40 INTERFACE AGE an axis or plane Of symmetry Of the bOdy, as in (c). only pass through the center Of mass, but 引 SO coincide with the imposed torque, we mustlet the axis Of the torque not desired that the rotation take place about the same axis as rotation will not coincide with the axis Of the torque.lf it is the rotational motions mentioned previously, the axis Of the translational motion; however, due tO the coupling between the torque through the center Of mass. This eliminates the rotation, but 引 SO translation.ln (b) we have located the axis Of through the center Of mass. Such a torque produces not oniy ⅲ figure 2(a) a torque is imposed whose axis does not pass duces a pure translational motion. lational force that passes through the center Of mass pro- mass produces rotation, as opposed tO (b), in which a trans- translational force that does not pass through the center Of choice Of axes. Referring t0 figure 1 , we see in (a) that a translational or rotational motions shed some light on the Positions Of the forces and moments which produce the end up with six independent degrees Of freedom. motions are decoupled from the translational motions, and we pass through the center Of mass Of the b0dy, the rotational rotational motions tO each Other. Finally, if these coordinates tion drop 0 リ t. These are terms which relate 0 「 couple the fied, and a number Of complex terms in the equations Of mo- axes, the mathematical description Of the geometry is simpli- げ these same lines are chosen as the initial coordinate Uncoupling the motions fO 「 the same reasons. tion are customarily chosen coincident With these same lines motions are 引 SO mutually perpendicular, and the axes Of rota- tages in discussing the geometry Of the motion. The rotational and move with it. This choice presents a number Of advan- are referred tO as bOdy axes; that is they are fixed in the bOdy cular tO, lines or planes 0f symmetry Of the b0dy. Such axes tomarily these lines are defined coincident with, 0 「 perpendi- cular, occur along three mutually perpendicular lines. Cus- The three translations, since they are mutually perpendi- pendicular tO each Other and therefore mutually independent. boats, as 「 0 , pitch, and yaw. These motions similarly are per- There are a 0 three rotations, referred tO in aircraft and they occur at right angles 0 「 perpendicular tO each Other. tions, called translations, are independent Of each Other since move fore and aft, sideways, and up and down. These mo- they have six degrees Of freedom. Thus a rigid b0dy can different ways, each independent Of the Other. That is tO say, the three dimensional world, rigid bodies can move in six tO one another. tween the particles making up the bOdy are fixed with respect rigid. A rigid b0dy is one in which the specialrelationships be- in the form Of objects 0 「 bodies which can either be elastic or among Other things, with the motion Of matter. This matter is Mechanics is that branch Of physics which concerns itself, Rotation Of Coordinate Axes by Dr. Alfred AdIer MATHEMATICIAN Force Force 十 (a) 十 (b) Trans1ation plus Rotation Trans1ation Figure 1 . TransIational force (a) れ Ot passing through center Of mass, (b) passing through center Of mass. 十 (a) ( b ) ( c ) Trans1ation plus Rotation ト - Rotation about new axis Rotation about Torque axi S Figure 2. Torque (a) not passing through center Of mass, (b) passing through center Of mass but れ 0t aligned with principle axis, ( 0 ) passing h 「 ough center Of mass and aligned with principal axis. OCTOBER 1980

7. Interface Age 1980年10月号

P 日ロじ 6 Y 血日 by Hugh Poynor This is the first 0f a 2-part article, t0 be conc/uded next month, on わ 0 Ⅳ tO write programs 加 Basic that 0 ″ e 「 improved screen formatting, effective user prompting and convenient user input. Sitting down at a computer video console tO run most programs requires that the user select options tO meet the unique requirements Of his jOb. These options, commonly referred tO as jOb parameters, are the values given tO important variables used by the program. Para- meters specify which Of many operations are tO be per- formed: the account tO be displayed, the quarterly period tO be summarized, the data files containing the records Of interest and SO forth. What is needed is tO show how programmers can fully utilize the ability Of computer programs tO present stored variables as jOb parameters and ask for a mini- mum amount Of new information from the program user. TO achieve this, procedures will be covered that can best be described as reflecting good 、℃ omputerist psychology. ' ' This refers tO an understanding 0f person-computer interactions that simultaneously takes intO account the abilities Of users, as people with varying degrees Of urgency and sophistication, and the power Of stored program computers, particularly those With video screens. Screen printing Many programmers use the video screen as a line printer, scrolling line after line Of output from the bOt- tom Of the screen. They are using the computer's power in reverse, at least insofar as good computerist psychology is concerned. While bottom-printing may be a primitive example, it is one Of the most popular bad habits used by program- mers. Unfortunately, it is rife among leaders in the in- dustry. Popular operating systems written by the best software houses routinely incorporate bottom-printing. AIthough this article aims at source programmers work- ing at home, hopefully it will find its way intO the various software houses. What follows are a few principles tO take intO account. The first concerns screen printing. Use top-down p 「 / 厩加 g that stops when the screen is ル″ . Move only one line ontO the screen at the bottom, thus giving a natural /eft tO right and top tO わ 0 せ om page effect fO 「 easy reading. HOW can the top-down screen printing principle be programmed? 1 . Clear the screen and home the cursor 2. Print only the next 23 lines (a 24 x 80 screen) 7 00 INTERFACE AGE 3. Ask for input in order tO proceed or quit 4. LOOP on ( 1 ) The programming in this case operates tO the disad- vantage Of the machine's efficiency. Several micro- processor operating systems (). g. , CP/M, Cromemco, MITS, MicropoIis) have disk directory and file dump routines (). g. , 'dir, 、 type') that race the information past ou 「 eyes at a frightening clip. Often, therefore, users can 0 可 y glimpse the top parts Of larger disk directories. Sometimes smaller disk directories will fit a screen, and in those cases the directory display pro- grams are fully useful. CIearing the video screen Besides the inefficiency Of keeping a machine idle while someone reads and reacts, a more serious prob- lem stems from the codes that manufacturers use tO clear video screens. These are ordinarily documented in the user manuals, but sophisticated codes may be lacking in the case Of Older terminals with skimpy video controllers. Generally, tWO operations are performed: clearing the screen and homing the cursor. ThiS involves using tWO or more special codes which must be printed. Dif- ferent manufacturers select different codes for some non-character printing simply because the Ascii stand- ard does not establish the codes forthrightly. This is the case fO 「 homing and clearing the screen, for which there are no standards. は is suprising that codes have been established by Ascii standards tO position the cursor (one position at a time) and ring the bell, while the codes that are re- quired for utilizing the fundamental screen printing principle have not been formalized. The screen-manipulating codes are less well stand- ardized than the Ascii character codes. Bottom-printing may arise out Of this standardization problem, insofar as homing, screen clearing and Other screen-manipulating codes are concerned. Realistically, there is small vari- ance in the codes—possibly no more than three are employed by the video screen manufacturers. There- fore, the problem certainly does not stem from an over- whelming variety Of codes. Programmers can SOlve the unknown COde problem in one Of several ways: ( 1 ) determine the codes for the video screen on which the program will be operating, and then supply the program t0 the user appropriately coded; ( 2 ) use only the most popular codes, and 0f course, supply a caveat emptor for Other screens; ( 3 ) use all three codes where on ツ one is really required (this will not have undesired side effects, but it should be tested); ( 4 ) provide programs with a screen output subroutine arrangement wherein a user iS required tO OCTOBER プ 980

8. Interface Age 1980年10月号

By Roger Garrett Building a Simulation System 22 INTERFACE AGE quest fO 「 a challenging mission.l can't see what's happening We finally arrive at the 0 ⅱ well and ー begin to regret my re- Disaster strikes "Just fine, thanks. realistic as ー would like butl can change that next time around. sea. This plastic ocean apparently contains no life. Not as TO my left and rightl can see the midnight blackness of the above shows the ocean surface, choppy but not dangerous. The additional viewports appear as requested. The one 'How's this?" dows. lt's a little claustrophobic with this one little viewport. station, tO give me views Of the ocean. Something like win- head display, and perhaps one on either side Of my control ・ Everything is working fine but lreally could use an over- "Yes?" ℃ omputer!" knots towards the crippled 0 ⅱ well. east. FuII forward on joystick one and we are cruising at 60 begins turning, slowly, tO the correct heading; north-north- With a slight twisting motion of my left hand, the minisub ー push forward on my right joystick. The engines speed up. 、 'Then let's got 'Just fine here. nominal. "Everything Ok with you?"l call to Jack. ー make a visual check Of the control panel. Everything challenge. NOt quite that difficult,l think tO myself. Butl accept the own. Three men are missing. Seas rough. TWO kiiometers north-northeast an 0 ⅱ well has 'Your position is the GuIf Of Mexico. Depth, 30 meters. As you wish. gets the better Of reason. Perhaps ー shouldn't be so bO 旧 . Confidence occasionally "Select a task. A difficult one. perimental minisub. portantly,l am confident Of my own skills in operating this ex- satisfied that all systems are in full functional 0 「 de 「 . More im- ー have already run 1 3 checkout runs this month and am task tO you?" 、 DO you wish a standard checkout run 0 「 shalll present a ing around at hiS console preparing fO 「 ou 「 mission. craft's engines and the sounds Of my assistant, Jack, putter- me, where nothing existed before, ー hear the whir Of the and speed control, appear at the ends Of my armrests. Behind display directly in front 0f me. TWO joysticks, for directional array Of dials and gauges surround a color radar/status control and a minisub control console materializes in itS place. An My standard keyboard and display dissolve into blackness 、℃ ertainly. 'Undersea exploration mission, please. "Yes?" ℃ omputer!" on the surface but down here it looks like a Kansas tornado. The drilling rig has broken right at the base and the plume Of escaping 0 ⅱ , rising rapidly tO the surface, has set up a vortex threatening tO pull us into it. Emergency reverse thrust SIOWS ou 「 advance towards the rig, but is not sufficient to completely pull us away.l should have requested more infor- mation from topside before getting so close, but that's just a judgement error. What matters is that the sub can't resist the force Of the current.l could ask the computer to modify the engine design but at this point that would be cheating.l could take care Of thatlater. FO 「 now ー push the sub to its limit and take my chances. ー ask Jack for a status check Of the explosives deployment system. "Everything nominal,' he responds. "NominaI," ー repeat tO myself. l've got tO give that guy more vocabulary. ー change thrust tO half-forward and aim the sub at the base Of the rig. The vortex quickly takes ho 旧 and we spiralinward, being swept up with debris from the damaged rig, sand, and a stray shrimp (there are fish after all). The sub shudders against the strain as ー try to keep us headed directly at the base Of the rig. は is no easy task. We are now circling at one revolution every 45 seconds. The view is beginning tO b ⅳ「 . T00 much sand, t00 much 0 ⅱ . And my inner ears are screaming for relief. ー have to take the chance thatl am finally close enough. Deploy!"l yell back tO Jack. ("Deploy?"l think to myself. 'Fire" 0 「 "bombs away" would make more sense. ButI had fallen into the mundane vocabulary of my assistant. ) The explosive canister shoots out from the deployment tube directly towards the base of the billowing 0 ⅱ . But the vortex is stronger than the canister's propulsion system. は spirals faster and faster into the very heart of the black cone. A steel beam, jutting out from what is left of the rig, catches the canister with full force. Did we make it? Jack vanishes. The control panels blink once, then twice, then dissolve away.ln their place my keyboard and CRT once appear. "SimuIation ove 「 . ' ' The voice Of the computer finally breaks the silence. ー sit motionless fO 「 several minutes taking deep breaths to regain what little composure ー can before responding. 旧 OW did ー do ? ' ' コ ask. 、、 You stopped the 0 ⅱ spill. 'Great!" linterrupt. "But you and your assistant were killed. T00 close tO the explosion. The minisub just can't take that kind of force. "Damn,"l think to myself, ・ was sure ー was far enough away. "And you apparently forgot about the three missing men. ' ー had, but hOW many things can two men (well, one man and one 、 'assistant") accomplish in such a dangerous situation? ー was about tO argue my point but reconsidered. The computer just wouldn't understand. lreach forward and flick the abort switch on my keyboard which then dissolves away to join the rest of my imaginary hardware, safely tucked away in the memory of my simulation computer. Overhead lights flash on, revealing the four walls 0f my evaluation/training cubicle.l get up from my chair and carefully remove the telepresence suit,leaving it tO dangle from the myriad Of position control units suspended from the ceiling. ー begin tO leave the 「 00m but pause momentarily at the d00 「 . 。、 Tomorrow,"I say quietly, 、 'l' save that 0 ⅱ well and the three missing men. Back tO reality By now, if you are a regular reader Of this column, you are probably wondering whether the editor goofed. Or maybe you think l've gone off the deep end. After 訓 , can ー possibly imply that a simulation system as sophisticated as the one de- OCTOBER プ 980

9. Interface Age 1980年10月号

Other samples include a game which has a colorful array Of letters march on the screen tO form a jumbled word. The player has tO unscramble the letters before they with fun and eliminate the confusion march intO their correct position. a child feels when he learns tO Another game, the Face Maker, operate a computer. Hakansson, uses the computer as a creative WhO began her career With com- tOOl. The player chooses from puters at the Lawrence Hall Of various facial shapes and features Science at the University Of Cali- tO create unique characters. fornia, BerkeIey, has been working The majority Of the machines on the Sesame PIace project fO 「 have a menu Of tWO tO three pro- grams. But there are 引 SO dedicated almost tWO years. ・ felt it was important that, machines, running one program since some games access a lOt before a WhOle new generation grew up in fear and intimidation, it more files. needed hands-on experience with AII the computers are token- computers. The more ー saw children operated with the games lasting ap- use computers and interact with proximately 4 minutes. “に s a them, the more convinced ー became friendly 4 minutes," Hakansson that it was a magnificent way tO says. 'Rather than cutting you Off, the timing mechanism will check teach. Hakansson goes on tO explain and, if the playershould hit a natural that even the most patient teachers ending a few seconds after 4 place expectations upon their minutes that's fine; if he hits it students. "A student feels a IOt Of before, that's fine t00. は allows you stress and pressure When dealing tO complete the exercise. with the adult community—he The reason for the token- always feels the need tO perform operated machines Hakansson and meet up tO the teacher's claims, iS more for crowd control expectations. You take that away than anything else. "We feel that if when you 訓 OW the child t0 work people pay t0 get in they should with a computer, which deals with have an opportunity tO use the en- him on an individual basis. The com- vironment, but if we dO not have a puter—being non-judgemental tOken operation, visitors will sit in —lets a child take as long as he front Of the terminals for hours, not giving others an opportunity. needs tO answer a question. WhiIe many think Of a computer What are future plans for Sesame as a CO 旧 , complicated machine, Place? Hakansson claims they Hakansson feels that it can would like tO open more Sesame Places around the country. "The generate a very personal ex- perience, especially in programs more we expand, the more children that deal with each user on an in- are exposed tO 'technological' dividual basis. learning. Computers should be another resource in the school en- Types 0f games vironment and Sesame Place iS one step towards making the public Some games in the gallery are designed tO aid a visitor in develop- more aware. Hakansson hopes tO mass market ing hand-and-eye coordination, the software developed for Sesame another in logical strategies, or PIace next year. She wants tO in- skills in reading. Perhaps a player tegrate it with Sesame Street's would like tO try a music game or a learning concepts. 、 Then a child at creative writing exercise. For exam- home could watch Sesame Street, ple, youngsters use the computer turn off the TV, turn on the com- tO write their own Sesame Place puter and play games, drawing on mystery. げ they don't like that, they the ideas he just saw on TV. may want tO try and guess the pat- 。、 As the micro population ex- terns forming Oscar, Big Bird, plodes and becomes more afford- C00kie Monster and the rest Of the able, parents are going tO start ask- muppets. A business-oriented ing whether their children are using game ShOWS a young entrepreneur computers in schOO コ believe they operating a lemonade stand what will make choices dependent on decision-making choice he must that answer. ' ' ロ make tO stay profitable. Sesame PIace Continued from Page 15 You now order ardcle reprints 伝 om University Microfilms lnternational, in cooperationwith publishers ofthisjournal,offersahighlycon- venientArticle ReprintService. SingIearticIesorcompIete issues can now beobtained in their 0 「 iginal size(upt081/2 x 11 inches). FO 「 more information please com- plete and mail the coupon below. ARTICLE REPRINT SERVICE University MicrofiIms lnternational 90 品寸 Ue6!UO!b•N - 」 oq 」 V uuv peou qeez UUON 009 の 0 一 ~ ののを一」 d のエ型 0 モ V 一 0 32 一 S のの e もっく 、 uedwoo 、 uo 一ち = u 一 eweN ・の三一型 qe = 2 》 e 一 0 e コ 60 一 2 0 9PnlOU! eseeld ロ ・」の℃」 0 ueo 一 uo SI!EIOP lln; ew pues e の eeld 60 一 ~ の S を一」 d のエ一 0 モ V ①をち oqeelow ou 0 一の一 PlnOM 一あ山Åロ e 三ト OCTOBER 1980 7 20 INTERFACE AGE

10. Interface Age 1980年10月号

HEATHKIT@ computer products and Zenith Data Systems get even better when you read Buss: The lndependent NewsIetter Of Heath CO. Computers. Buss spreads 石 6 ーー 1 五、「一病、一 6 ・「 - ー 66 市 0 6b16 ・一 hard w are and software from many vendors. The newsletter is not affiliated with any Of these and does not depend on income from ads. SO Buss has tO please only its subscribers—over 2900 OT -- fhem. Readerst reports Of their discoveries save Other subscribers headaches—and money. lnnovations published in Buss have included hardware modifi- cations and fixes. And, since it iS not a company-controlled publication, Buss can deal with bOth the weaknesses and strengths Of Heath CO. computer products. Buss is a proven newsletter that keeps getting better—itts been published since ApriI 1977. you subscribe this month yo Ⅱ receive a free directory Of over 50 suppliers Of hardware and software for Heathkit@ computers and zenith Data Systems. You have the choice Of starting your Buss subseription with the latest issue or available back issues (about 14 ). AII this information will be mailed first class ()y air mail overseas). Full refund guaranteed if youtre not satisfied. pay- ment must be in U. S. dollars payable on a U. S. bank or by Master Charge or Visa. lssues: U. S. & Canada Overseas $ 17.97 25.00 B uss $ 24.89 30.00 24 $ 29.90 35.00 325- A PennsyIvania Ave. , S. E. Washington, DC 20003 CIRCLE INQUIRY NO. 3 1 0 fo 「 $ 29 毛 0 Verbatime Verbatim FD 34 8000 Diskettes f0 「 the SCOTCH 0 「 MEMOREX: 10 f0 「 $ 31.00 TRS 80 MOD 10 fo 「 $ 51.00 STORAGE MOIded plastic. Hold 10 disk- Fit any 3- ring binder. Phone ( 215 ) 439-1717 Free CataIog Checks and Bank Cards 0K green bar 0 「 plain. One part. 1 / 2 " X11 " 0 「 1 / 2 " X11 ” , M ⅲト P 「 e 「 STOCK cards. VinyI. 10 f0 「 $ 5.75. HO 2 diskettes and 旧 ettes. P. 0 Box 327 Disks Etc. Maxi ( 8 " ) Case: 図 .25 Mini ( 1 / 4 ' ) Case: N75 CenterVaIley. PA 18034 CiRCLE iNQUlRY NO. 72 58 INTERFACE AGE mapped display within the inherent limitations Of a serial terminal With varying degrees Of success. One Of the newer products is Wordmaster 日 , a set of programs that ties the text editor intO a page formatter and menu-driven command processor fO 「 Office appli- cations. Wordmaster Ⅱ can perform nearly 訓 the func- tions Of the operating system in a simple, generously- prompted manner by novice operators. For example, the file directory can be altered as if it were another text file, and the files are renamed, deleted 0 「 copied instantly tO match the altered directory. Basic is the majorlanguage on the system. This one was described tO us as 。 your standard enhanced super- set Of Dartmouth Basic. "ln a world dismally lacking in Basic standards, we find a IOt that is similar among the various offerings. ThiS version iS neither more nor less capable than most. However, it hOlds the bOOby prize in execution time for our prime number cruncher bench- mark program ()A Jun 80 ) : 3 , 300 seconds. TWO versions Of Pascal are available. One is the fami- liar UCSD Pascalt0 execute the P-code of any program developed fO 「 a UCSD-compatibIe computer. The other is PascaI/MT, a subset Of UCSD PascaIrunning under the normal operating system for 88. As a result, the data files created in PascaI/MT are compatible with those used by a Basic program. PascaI/MT programs are assembled right down tO 8080 machine code, mak- ing them execute very fast. A version Of ou 「 prime num- ber cruncher zipped through in 119 seconds. Macro-88 is included for 8080 assembly-language programmers, including macro definition capabilities and Other t00 that the factory has developed fo 「 its own use while creating the operating system. FinaIIy, Digital Research's CP/M is now available on 訓 POIymorphic computers. The consumer should be able tO purchase any Of the thousands Of canned CP/M-compatible applications programs being offered. Varied applications programs The applications software catalog is thin. は includes a manufacturer's inventory package and mailing list p 「 0- gram, bOth in daily use at the company's factory. A 0 listed is a financial modeling t001, Plan, a "what if?" infor- mation handling program sharing some Of the features Of Visicalc by Personal Software. A 0 , there is a listing Of about 50 programs available from local dealers. We witnessed a preview Of Datamaster, a forthcoming language/applications package. は will be a database man- agement system combined with a super high level appli- cations program development tOOl. は will 訓 OW a non- programmer tO make intelligent inquiries intO a database and create immediate customized management reports. は is being translated from Basic intO Pascal/MT. A Twinsystem 88 with a pair Of double-sided, double- density 51/4-inch diskette drives lists for $ 13 , 200. This includes tWO terminals, interconnecting cables, disk- ettes containing the operating system, editor, Basic, and the macro assembler and user documentation. は SO includes a set Of comprehensive confidence test programs. TWO double-sided, double-density 8-inch diskette drives cost an additional $ 4 885. Software pricing ranges from $ 95 for the mailing list program t0 $ 195 for Wordmaster ル lt's feasible tO purchase a single-user, single-disk- ette system 8810 for $ 5 , 208 and build it up tO the full capabilities Of the Twinsystem 88 later. ロ OCTOBER プ 980