Stan mitigates the odium in comparison by being, as usual, marginally scrutable.
In the other-worldly jargon of dreaded math examinations, where each non-union, disembodied Stakhanovite digs X ditch-meters in Y days, and anachronistic milkmaids churn X imperial-gallon drums into Y liter bottles, many a poor distracted, literate innumerate examinee wastes time, and fails, pondering the irrelevant narratological subtext. The Len(MarxEng) students wonder which oppressive socio-economic regime has set the given, arbitrary ditch-digging targets, while others, swimming fresh in the George-Eliot-ModEngLit stream, dream of a buxom, white/blue-starched-attired, fine-fingered milking maid (called Chloe?)...yet an udder story?
More seriously, we also meet outright ambiguous testing formulations that have led to injustices, student/parent-grader litigation, and even GBH (Grievous Bodily Harm). Thus, Charles Baker reminds me of the prestigious MBS (Margaret Bryce Smith) exam that settled your fee-paid Liobian entrance to the exclusive Liverpool Institute (a.k.a. locally as de Inny_ [1]). One of the questions confronting Charles was "If 3x + 2 = 14, find x." Unexposed to algebraic smalltalk, he thought "What a stupid question" and "circled the x." Prof Underwood Dudley (famed exposant of circle-squarers and angle-trisectors [2]) confirms this as a real examiner trap, and reports that "find x" has now been safely replaced by "find the value of x." I suggest that the Supreme Court might require the addition "Having found your putative value of x, write it in the box provided." No doubt, slippery lawyers (forgive the tautology) have endless dispensationist interrogatories at $560 per nit, demanding clarification of "value," "it," "write," "box," and "provided." Not to mention the covert, politically-dubious "true."
In a simpler test, the scenario is that A has five oranges and one lemon while B has four oranges and two lemons. Clearly, A has more oranges but fewer lemons than B. You get a point (possibly a Ph. D. in some nudge-nudge polytechnics) for confirming this proposition, but in the elite campi [3], you may next be asked who leads in "fruit" ownership. The naive reaction, assuming that "orange" and "lemon" belong to the superclass "fruit," is that A and B both possess six fruity objects. In the strict binary domain of "less than" or "greater than," the smart examinee could well answer that "leader" is ill-defined. Unrevealed in the question's givens, we have no idea how many other fruits, e.g., apples, if any, A and B may have in their baskets. (Recall the mock G. E. Moore Beyond the Fringe sketch: "Is it or is it not the case that you assert or deny that you have some and/or no apples...?") We hit the teasing "argument through so-far silence" that infests all known epistemolgies and Hamlet soliloquies. We posit, declare, believe, hope, demand, imagine facts, yet each, in our inevitably sequential exposure thereto, leaves open current, gainsaying doubts. Quine, for instance, has a disturbing, paradoxial "proof" that most of our cherished beliefs must be "knowingly false." (Not to be confused with the Popperian "scientifically falsifiable.")
To answer the fruit-ownership question, we would also need a widely accepted definition of the taxon "fruit" via botanical cladistics [4]) or via a volatile enumeration of "fruit" instances (yes, we have no tomatoes), together with provable algorithms for determining "who-owns-X," counting "how-many-X," and, finally, the easy bit, "comparing integers." (Lawd, let's ignore A's melon slices and B's pine-apple cubes.)
Lest I've lost your trench-coding attention with these metaphyisical musings, recall the crucial OO debates on operator== and operator>. Can you measure programmer productivity by LOC (lines of code)? Or measure "encapsulation" by the number of member functions? [5] Those of us brung-up on topology know the precise ordering quirks of "metric" spaces (and some of us versed in Thom/Zeeman quasi-metric, catastrophic "tolerance" spaces even more so), see the dangers of inept comparisons. Billions of dollars have been wasted on dubious if Real1 == Real2 tests. Despite the obvious type-safety checks, we've recently seen major (obscene) NASA losses due to unit-safety lapses. If Real1 measures Newtons and Real2 measures Pounds, even a mathematically-correct IEE-FP-ISO comparison is expensively fatal.
Finally, returning to the LitHum domain, Shakespeare oft risked ordering disparate objects. Thus, one of his sweetie-pies (possibly a woman) was rated "more lovely and more temperate" than "a summer's day" (Sonnet XVIII) [6]. Later, invoking the operator !=, Bill the Quill scorned: "My mistress' eyes are nothing like the sun" but then, for once, switched to a more [sic] OO-acceptable comparable predicate: "Coral is far more red than her lips' red" (Sonnet CXXX).
We endure many implied operator> orderings. The regular light-hearted Letterman Top Ten has been swamped recently by serious millennial lists in various categories. But how serious when you see the Beatles and Elvis rated above Beethoven, Mozart, and me? Shakespeare (just a nose above Barbara Cartland!) predicted this travesty:
"Be thou the tenth Muse, ten times more in worth/ Than those old nine which rhymers invocate" (Sonnet XXXVIII).
Notes and References
[1] See www.southernx.com.au/liobians.html/.
[2] Underwood Dudley. Mathematical Cranks (MAA Press, 1992).
[3] Cade Roux (Caius' 1989) was the first to chastise me (Downing 1950) for my error in placing Tony (C. A. R.) Hoare in the retirement home at the Microsoft Research endowment at Oxford (this column, February 2000). In fact, alas, mixed feelings, Bill's largesse has fallen on Cambridge. As my Liverpool Institute tutor used to say: "It doesn't matter which university you get to, they're both very good."
[4] Bertrand Meyer. Object-Oriented Software Construction 2nd Edition (Prentice Hall, 1997), page 865 et passim. For a recent big-stakes update of the "taxonomy of taxonomists" debate (Hennigians vs. The Rest), see In Search of Deep Time: Beyond the Fossil Record to a New History of Life, Henry Gee, Free Press, New York, 1999. Gee, in the new evolutionary classifications (e.g., reptiles and birds), we extend is-a and has-a to was-a and had-a.
[5] Scott Meyers. "How Non-Member Functions Improve Encapsulation," C/C++ Users Journal, February 2000, p. 44.
[6] Bill's rival, Chris Marlowe (a Cambridge guy) introduced a formal measure of pulchritude: the millihelen = the amount of facial beauty sufficient to launch one ship.
Stan Kelly-Bootle has been computing on and off since 1953 when he graduated from Cambridge University in Pure Mathematics and hacked on EDSAC I (the first true stored-program computer). He is a contributing editor for Linux Journal and a Jolt Judge for Software Development Magazine. With the demise of UNIX Review/Performance Computing, his 16-year-old Devil's Advocate column has moved online to www.sarcheck.com. His many books include 680x0 Programming by Example, Mastering Turbo C, Lern Yerself Scouse, The Devil's DP Dictionary, The Computer Contradictory, and Unix Complete. Under his nom-de-folk, Stan Kelly, his songs have been recorded by Cilla Black, Judy Collins, the Dubliners, and himself. Stan welcomes email via skb@crl.com and his website http://www.crl.com/~skb/.