A new calculator and why it is necessary∗Harold ThimblebyComputing ScienceMiddlesex UniversityLONDON, N11 2NQ.Email: [email protected] 30, 199
(a) Copy and complete these number sentences.7 × = 2828 ÷7 =× 4 = 2828 ÷4 =(b) Copy the following series of fractions and fill in the missing numbers.3
Key press Display after each key press4 × −5∆= −20⇐4 × −∆5 = −20DEL4 ×∆5 = 20=4 × 1.25 =∆584 × 21.25 = 8∆5⇒4 × 21.25 = 85∆DEL4 × 2 = 8∆DEL4 × 1 =∆4Fig
you can, and what’s more you only need to input the formula once, thereafter you can just fillin the knowns and the calculator works out the unknown. S
Key press Display after each key press33.8∆= 1 × 1.8 + 32−−∆−33.8 = 1 × 1.8 + 324−4∆= −20 × 1.8 + 320−40∆= −40 × 1.8 + 32Figure 5: Converting Fahrenhe
Problem New calculator Casio fx-82lb ‘fraction’ calculator4/3 =? 4/3 = 1 + 1 / 34 abc3 =, answer: 1 1 31.1 =? 1.1 = 11 / 10 = 1 + 1 / 10 cannot handle
An important property ofRCLis that whenever the memory is used its value can be seen (this isnot the case on any other calculator reviewed here). The
or to beep and not change the display?9(The keyboard has a keyeso that large numbers can beentered exactly as they are displayed.)The prototype calcul
7.9 Other important design detailsCurrent calculators can be criticised on their non-technical design. A good calculator would be ruinedby poor design
log10(3) + log10(4) = log10(12) 10 ↑ log10(42) = 4271 × log10(10) = log10(10 ↑ 71) 2000 = 3 + log10(2)Figure 7: The calculator as a ‘chalk board.’ The
8.6 Non-specific complaintsMy final response to objections is that the new design is mathematical. It is mathematical in twoimportant ways that no other
decimal point, when not in an error condition . . . ). Furthermore, as users may make slips — pressing thewrong button, omitting a press, or pressing
It may be desirable to introduce modes for working in degrees, changing the logarithm base, or settingother preferences. This can be done consistently
into algebra rather than numerical coincidences, would cause more educational damage than the cal-culators we are trying to supercede. (That wouldn’t
[11] P. Latham and P. Truelove (1983) Nuffield Maths 3, Pupil’s Book, Longman.[12] R. E. Mayer and P. Bayman (1981) “Psychology of Calculator Languages:
'&$%22/7∆−π = 0.001264 →⇐F IX DEL⇒(√π)7 8 9÷4 5 6×1 2 3−0•=+Figure 8: The new calculator — example key layout. The calculator has a protectiv
All might be harmless design variation, except manufacturers’ claims suggest otherwise, as thefollowing typical example from a market leader makes cle
Model4 × − 5;1 − 5 %;1 + 5 %NotesCanon WS-121H −1 −80† 1.05 †1 + 5 ± %calcu-lates 0.95.Casio MS-70l −1 −80 1.0526315Casio MS-270l −1 −80 1.05‡ ‡1.0526
This suggests 2−πmight be worked out by2 yx− πor perhaps2 yx( − π ). Neitherapproach works.It is sensible to do some experiments, to see how a calcula
marketted for at least ten years and its serial number suggests it has been revised many times.4Calculators have inadequate constants (recall the inco
3.4 Over-functionalityButtons mean lots of different things. Button with four meanings, depending on the mode, are common.Many of the calculators (espe
evidently means 4 − 5. But if there is a ‘correction rule,’ it does not generalise:83√ √takes a6th. root, not a square root; and2 + + 3creates an ‘add
according to the rules of arithmetic.” [24]The purpose of a calculator is to do correct calculations, and to do so efficiently. It is clear thata calcul
Kommentare zu diesen Handbüchern