Son of Devil's Advocate, March 2004

Click here to make UNIX performance tuning easier!
advertisement

Previous SODA Index of SODA Columns Next SODA

Son of Devil's Advocate

Stan Kelly-Bootle


Picture of Stan Kelly-Bootle

Melting Pot Luck?

Each year Intel Corporation returns some of its well-gotten gains by sponsoring a Science Talent Search among USA high-school students [ref 1].

As reported in Science News, January 21, 2004 (The Chosen: A New Crop of Scientific Minds), this year's contest started with 1,652 entrants, aged between 16 and 18 years. The 41 finalists were announced on January 28 by Science Services, the publisher of Science News who administers the competition. The finalists' stats are intriguing: 45% are female and 66% are bilingual (at least). And 45% are from New York (quick chorus of "It's a Wonderful State") No breakdowns by ethnicity are given, which is understandable and praiseworthy in these hopefully "color-blind" days. However, I extracted the finalists' surnames (as best I could [ref 2]) and sorted them alphabetically:

Alexeey, Anathur, Auster, Banneker, Chauhan, Chen, Chi, Choi, Collier, D'Ascoli, Deutsch, Glukhovsky, Goldberg, Ha, Halperin, Hang, Harel, Hartman, Hedberg, Heller, Karnik, Levin, Lubin, Michta, Munteanu, Nachbaur, Nettimi, Raj, Rau-Murthy, Reznik, Richelson, Storm, Su, Suri, Thakur, Wagner, Wang, Westrick, Wolfson, Yen, Zhou.

Is it indelicate, misleading, irrelevant, and/or dangerous to observe the distinct paucity of what earlier generations dubbed WASP (White Anglo-Saxon Protestant) appelations?

The wondrous waves of immigrants (some more huddled than others) passing through Ellis Island during the 19th and 20th centuries often met ethnic intolerance based on names and stereotypes. Fritz Spiegl [ref 3] once told me, and I have no reason to doubt him, that one NY Immigration Officer studied the papers presented and said to the would-be entrant "Igor Stravinsky? Igor Stravinsky? I can get that changed if you'd like." Thus, many a Lakatos Imre sought the American Dream as Imre Lake. Often the strange-sounding names were simply misinterpreted and misrecorded in the hurly-burly, with a final "-ski," "-stein," or "-ov" curtated to remove traces of the shtetl.

There are no doubt doctorates in Sociology past and pending to draw the politically sensitive conclusions. My tentaive spin from the Intel finalists is that the scientific wealth of the USA owes much to immigration, and it would be easy to make a similar case for the wider cultural and artistic benefits of the "melting pot."

Re-reading Richard Rhodes's "The Making of the Atomic Bomb" (Simon & Schuster, 1986) reminds me of the particular contribution of the European physicists who were expelled from Nazi Germany (and Fascist Italy, of course) or who left to escape the impending Shoa [ref 4]. Hitler's big mistake was to think that "Jewish" Physics were inferior to "Aryan" Physics. When Einstein left, the party newspapers declared "Germany is well-rid of the Jewish scum." Ironically, it was Einstein who helped persuade FDR to initiate the Manhatten Project. There are some echoes here of the current Iraqi Gulf War 2 debate. There were diverse signs that Germany was developing plans for an atomic-bomb. The basic facts of nuclear fission were well-known (there's Leo Szilard's "epiphany" September 12, 1933 -- a date that will live in all atom-bomb biographies), leaving only "a few technicalities" on the road to massive explosions. Further, one of the leading particle physicists, Werner Heisenberg, had elected to stay in Germany (for reasons still under debate), increasing the fear that Hitler might well have a team capable to producing a WMD (to invoke the fashionable TLA) before the Allies. Now, hindsight is all well and good, but the Allied consenus was that a NAZI WMD first-strike was such a horrendous prospect that no risks could be taken. The one certainty, amid all the smoke and mirrors of counter and counter-counter intelligence, mis- and dis- information, was that Adolf Hitler (or par pari refertur, or even non nova sed nove, his prime pupil, Saddam Hussein) could not be trusted with such weapons, real or potential.

Misfiled Books

This heading, well-known to my older fans, started way, way back with the discovery of The Bourne Conspiracy tucked in the UNIX section of my local library. Readers rushed to report similar misfilings, actual or imagined. Thus: The Book of J (under APL); Equimultiplicity and Blowing Up (Balkan History); Five Graves to Cairo (Microsoft OS); Case Closed (Software Engineering). And my own salacious favourite: Springer-Verlag's Motions of Coupled Bodies about a Fixed Point (Adult).

Newly released and destined for the wrong shelf is The Expert Versus the Object -- Ed. Ronald D. Spencer (Oxford University Press, 2004).

Alas, not another of the "Beyond X" genre, where "X" can stand for "OOP," "UML," "AI," "The Future," or even "X." [ref 5]. But, is the real subject given away by the subtitle: "Judging Fakes and False Attributions in the Visual Arts?" The librarian might well be even further misled? Could this book be a tasty attack on the diverse Visual Studios and their hyped promises: "Your payroll suite is just a click away?" (The Advertising Council is coming down heavily on such deceptions. The preferred phrase is "...just a few clicks away.")

From the 1970s comes the tale of a company founded by Richard Pick. No, he didn't mind being called Dick! The Pick system was and remains forever ahead of its time. The hardware and OS were inextricably entwined using a remarkable Data Base with, as I recall, some advanced associative access tricks. They used to say that the DB is to Pick as the array is to APL. Anyroad, a consignment of Richard's manuals was confiscated by the South African customs. Because...the title...was "The Pick Pocket Guide." (see ref 1 for the perils of non-hyphenated ambiguation.)

We Have Mail

Hiding between the Spam, which resists our smartest filters [ref 6], I am pleased to acknowledge some heart-warming email, a recurrent theme of which is "Stan, we thought you were dead!" I rush to gainsay you, whence the biblical opposition between "the quick and the dead?"

Neville Hunter aims to instruct us with a remarkable animated powers-of-ten trip through the cosmos -- from macro via micro to nano, more!

http://micro.magnet.fsu.edu/primer/java/scienceopticsu/powersof10/index

To which I replied:

Dear Neville: quite nice!! (Candidate for Litotes of the Year?)

BUT had Archie Thorpe, our Physics teacher, lived to teach Quantum Chromodynamics, he would surely have mentioned the quarks' annoying "confinement principle," vitiating the final zoom of your tour beyond the nucleons!

And why not a plug for Liverpool Football Club? A group of Up-Red-Charmed aces chanting "You'll Never Quark Alone?"

From famed Welsh Methodist and Methodologist, Meilir Page-Jones:

Well, I must rush now to catch the post - I miss those endings of yore (whenever that was). Let me end by saying that I heartily welcome the return of the highly esteemed, all-leather, amusing and cerebellum-tickling SoDA. * (I observe that the most recent installment tickled me just in front of the cerebellum; it makes me nostalgic for those ante-cerebellum ticklers. "Have you ever been tickled just in front of the cerebellum, missus?" [(c)1957 K. Dodd Esq.])

I also welcome your return to vigor. As they say in Wales: Lang may yer lum reek (unless you're in a reekless zone, that is).

It goes without saying [so why say it then? - Ed.] that I beg the honour to remain, as always, your most devoted servant. Meilir.

* Not to be confused with the wretched Rational [sic] Rose reporting = add-on also named SoDA.

Dr Nick Jacobs writes:

Dear Stan, You remark in a footnote that ... "The usual, satirical spellings, Graudian or Graduian (etc!) are now unfair since the paper's typesetting system now seems to be equipped with a reasonable spelchock, at least for its own name."

The Alan Coren Omnibus, published in 1996, revealed what's behind this. Or rather, regular readers of either Alan Coren or The Guardian would have found out much earlier. Mr Coren's articles are often suggested by snippets of news he sees. One of the articles in the Omnibus, "Just a snog at twilight", cites as its inspiration an extract from the Guardian:

> 'The Head Proof Reader at the Guardian retired last
> night'. - Guardian
>

Space/Time precludes adequate acknowledgement of entertaining input from Rainer Brockerhoff, our Man in Brazil. I'll relate his insights on the WeBLOG pandemic next month. Meantime, check his site: http://www.brockerhoff.net/bb/viewtopic.php

Finally, I thank Al Anway for compiling a list of typos that escaped our careful profredding. He is awarded a bar to his Grande Palme Doryphorique. He is reluctant, however, to accept this honor, since doryphores are defined as "taking an excessive delight in pointing out small errors." He claims only a mild pleasure in correcting a column that often reports other's mistakes.

Poetic Interlude

"Trust Me, I'm a Poet" by Roger McGough

Your husband upped and left you After years of playing the field? My heart goes out, I know the type Of course, my lips are sealed. Let me be your confidant I'm generous, let me show it Champagne, I think is called for Trust me, I'm a poet.

Put my wallet on the counter When I turned round it had gone But I've got to meet my agent In town for lunch at one To sign a five-book contract I'll be back before you know it Can you lend me fifty quid? Trust me, I'm a poet.

The Way Things Are, Penguin Books, 2003.

Speaking of honours gained and honours declined: Roger was awarded an OBE in 1987 and the Cholmendeley [ref 7] Prize for Poety in 1998. Hamish Hendersom refused an OBE; hardly surprising for one of the most fervent Scottish Nationalists.

ref 1: The surprising best-seller,EATS, SHOOTS & LEAVES by E. S. Turner (Profile, 2003) stresses the ambiguating power of misplaced hypens. Thus, "extra-marital sex" and "pickled-herring merchant" take on new lives without careful punctuation. And "high school-students," although the norm, are not encouraged by Intel.

ref 2: Terms such as "surname" and "given name" are by no means self-evident without prior knowledge of their holders' cultural backgrounds. The use of "Christian name," of course, needs extra care. Walter Martin (The Bible Answer Man) would gently correct people who assumed from the designation Jesus Christ that His parents were Mr. and Mrs. Christ. Then there are the Hungarians who reverse the order with Kovacs Andras, for instance.

ref 3: Regular readers will be aware of Fritz Spiegl's profound influence on my life and career. The widely-published musicologist, sociolinguist and humorist had escaped the horrors of Nazi Austria, arrived young and penniless in England with little command of English, and crafted a new life in Liverpool. He discovered rich resonances between the dialects of Vienna and Merseyside, becoming fluent in the Liverpool Scouse funicular (known elsewhere as vernacular). After many years as principal flautist with the Royal Liverpool Philharmonic, he founded Scouse Press to publish the famous "Lern Yerself Scouse" guides. The timing was exquisite: the Beatles were, to borrow a phrase from Grady Booch [speaking of UML], "poised to emerge!" The world would soon be panting to interpret every malapudlianism and velar fricative uttered by the Fab Four. Alas, Fritz died suddenly last year while sipping his favorite Earl Grey. "Ah fill the Cup; what boots it to repeat that time is slipping underneath our feet...?" Or in my Scouse rendition of 'Omer Street Stan's Rubaaiyaat (which Fritz published in Lern Yerself Scouse, Volume 1, now in its 16th reprint):

"So come me lads, and fill yer boots wid beer; "Youse may be in Ford Cemitry next year. "Termorrer? Listen la' it never comes, "Let Fally drown yer sorrows, it's de gear."

Fritz & Stan

ref 4: Johnny (von Neumann to outsiders) was fond of the story that the AEC (Atomic Energy Commission) would hold its meetings in Hungarian if the two non-Magyar members didn't turn up.

ref 5: X as the familiar unknown quantity became problematical with the arrival of the product X in 1984 (an MIT windowing system for UNIX).

ref 6: To fool the filters, spammers introduce bizarre syntax and spelling. Thus, Viagra becomes Vi:a:gr^a:. Also Roget has been raided for all known synonyms for "penis enlargement." When they ask me if I'm troubled by the size of my "love-muscle," I am tempted to tell them "Well, I often trip over it rushing to the can."

ref 7: One of the joys of dysphonetic English in the famed tradition of "'My Fair Lady' pronounced Success." I recently met an American pilot whose proffered business card said "Randy Cholmendeley." I pleased him no end by saying "Chumley, old chum!" He confessed that he had learned to avoid the agony by calling himself, literally, "Chol-mend-e-ley."


Fifteen Long Years Ago

UNIX Review, March 1989 -- Devil's Advocate -- Stan Kelly-Bootle 12/28/89

Limning the Footprints in your Platform Environment

As your self-appointed, unpaid...well, let's say underpaid...OK, then, inadequately remunerated word-watcher and undefender of the jargon (See me immediately after this column...Ed), I bring you yet more searing exposures of the vagaries of DP grammar and its lexicon.

The word lexicon is perhaps a mite inappropriate to the computer industry since it carries certain implications of a vocabulary deployed with some semblance of understanding and consistency. Considering the semantic joyride from Software Engineering to the Sales Department, whereby any ratbag of conflicting modules becomes a seamlessly integrated system, I propose the word laxicon as a possibly new portmanteau description of DP verbiage.

I say "possibly new" to avoid challenging that army of citationeers whose days are "made" daily by determining an absurdly ancient use of any given word or phrase, especially if that usage predates the earliest given in the OED. The citation game is now fully computerized, by the way, so the prizes go to the scholar with the biggest database and the fastest grep.

Undoubtedly, Tom Paikeday, editor in chief of the Winston Dictionaries of Canadian English, pioneer in micro-computer-based lexicography, and frequent pre-citer over the OED, will immediately declare that laxicon occurs twice in Chaucer's Roman de la Rose (ca. 1369); further, that Chaucer was quoting Sophocles, who borrowed it from Sargon, who stole it from Shalmaneser V.

I feel on safer ground proposing the resurgence of a sweet old word, limn. Those who were hurt by my removal of paradigm from the laxicon (this column Vol 6, #10) may be pleased that my other hand promptly giveth back. A book reviewer, in Theology Today of all places, bristled politely at author Donald Weber: "He combines literary analysis with insights derived from historical and anthropological research to limn (one of his favorite words) a finely nuanced portrait..." (James H. Moorhead, op. cit., January, 1989, page 458)

One can sense that limn has outstayed its welcome in hermeneutic circles, so let's talk trade! They already have swaines of paradigms O boy, do they have paradigms: from Adamic to Zoroastrian, but theologians desperately crave our seamless integration. The deal is done: limn is now ours; religion gets seamless integration; we decide to throw in structured walkthrough to get apocatastasis (handy for restoring lost files); they agree to keep finely nuanced portrait until our pixels get smaller; plus, we get a minor deity to be named later subject to medical exam. (The latter provision was added upon learning that a Japanese god had recently died of cancer.)

Limn has what it takes to become fashionable in the computer tabloids: few know its meaning or pronunciation, lockjaw derivatives are easily formed, OED lists five spelling variants, and it packs a lot of pretension in just four letters. In addition to its drawing and painting connotations, limn resonates with the glories of large illuminated medieval fonts, the ones that fortunately require more RAM than the average amateur laptop publisher can afford.

Now for usage, repeat the following paradi...whoops, exercises, until the right measure of selfconscious fluency is achieved:

The window is limned by calling dowin(x, y, h, b); The refresh rate is limning-tool dependent.
We need to reprioritize the limnetics interface.
I limn; you sketch; he/she/it doodles (irregular).
SpeedLimnIt has replaced QuickDraw.
Let me briefly outlimn for you, if I may, our sales drive flow chart.

Launching the Platform

I move on to report with pleasure that environmental abuse has been declining of late. We still, alas, meet people and machines "operating in the UNIX environment" rather than simply working with or under UNIX. Furthemore, it remains difficult to avoid SETting environment variables under DOS, since they have been thus baptised in the canonical literature. And corporative spokespersons remain prominent offenders. Here's Jerry Sanders, Chairman of Advanced Micro Devices (AMD), quoted in the San Francisco Chronicle of 1/14/89:

"We expect that the demand environment for these products will remain poor until the inventory correction has been completed."

(Translation: "Customers will not buy more of our chips until they have to.")

However, a welcome word is mercifully decellerating the environmental bandwagon: let's hear it for the platform. Although platforms are usually of the hardware persuasion, one sees them worming their way into software environments too. Consider the following press release from Microsoft announcing Quick C v. 2.0 (I hate to do this to you, but who ever said that word-watching was all slap and tickle?):

"...the system software platform of the future -- MS(r) OS/2 Presentation Manager."

Platform has a reassuring, down-to-earth yet curiously romantic ambience, reminding one of schoolboy, anorak days, steam-train spotting at King's Cross, sweet milky tea and stale buns. Or, later, the Paris-Warsaw Wagons-Lits, double Pernods, diamond earrings, twisted stems of papirosi, sables, mistresses,... (C'est assez!...Rédacteuse.)

Environments are all static vapor and sterility. Platforms give a rarin'-to-go, rocket-launch-to-Mars feeling. "Upon this Sunix platform will I build my system." "The machine-ayah now standing on Platform 3-ayah will stop at Clapham Junction, Southend-on- Sea, Control-C, Mumble-in-the-Marsh, Ramsgate, Foobar, and all stations to Completion."

I do see a few impediments to the exclusive use of platform: the environmental lobby will parade the evils of offshore drilling and the cries of oil-soaked ospreys. Others will remind you of the antics at a Democratic convention: "This platform recognizes the drunken delegate from South Carolina." We also face a clash with the bizarre but growing use of footprint to indicate the acreage occupied by your hardware and mousepad, whether they sit on floors, desks, airplane tables, or laps. Our laxicon must now accomodate such expressions as, "We have decreased the footprint of our platform."

I also fear that some writers will confuse their prepositions and start running under platforms. Note the following the patterns:

Running in (aliter: under) the X environment Operating on (aliter: from) an X platform

and the allowed exception: "Running from an OS/2 platform" in the sense of away from.

Metapunctuation

You may recall my earlier discussion of the "stops-within-quotes" problem. I pointed out that the use of quotation marks to signify character strings in many computer languages can offend the punctuational conventions adopted by US (and some British) typographers and editors for printing normal literary quotations. If you are writing about a string in C or Basic, for example, every character shown within the double quotes cannot but be taken by the reader as an element of that string (I exclude the complications of special escape sequences like \n or \\, where two printed characters end up internally as one character, usually one byte). The compiler, too, will make that assumption when meeting literal strings in your source code, so it is important that our books on programming display strings correctly under all conditions.

The catch is that

...you will find that whenever a comma or period follows a quotation mark, the printers put it inside, regardless of the logic of the sentence...By all the principles of common sense, such practice is absurd...Somewhere, apparently, there was once a type designer who decided that his aesthetic sense was more pleased by the sight of a period inside a quotation mark than by that of one outside. Printed outside, it creates a small white space which he considered ugly. He started a tradition which has become so firmly established that today many printers seriously believe that there is some kind of divine law governing their practice...We have certain conventions in the use of punctuation that are not strictly logical. This is the only one I know of that is completely absurd.

I quote from Punctuation for Clarity by Robert Brittain (New York: Barnes & Noble, 1950), an excellent guide to rational punctuation. His war cry is:

Punctuation is not a strait jacket designed to make you think and write exactly like everybody else; it is a useful instrument that helps you to communicate exactly the shade of meaning you intend.

Brittain does acknowledge that proofreaders and printers are unlikely to change their sordid quoted-full-stop-comma habits, so he advises writers to kowtow, don't worry, and be happy. (The comma before a final, cataloguing "and" is another bone of contention, to be picked at on some future occasion.)

And so it comes about that the author's

If X$ is set to "x", the string Y$ is set to "end".

is usually mispublished as:

If X$ is set to "x," the string Y$ is set to "end."

This, of course, is "no big deal," until the billion-dollar space probe reaches the wrong galaxy. (Wasn't that a misplaced Fortran colon?...Ed.)

Here's a real-life example spotted recently in a programming manual (sorry, I failed to make a note of the source):

Comments must be placed on a line by themselves and are preceded by a "%."

Further reading revealed that only one character, "%", (not two: "%," or "x."!) was needed for a comment. Since commenting strategies do vary a lot, and have been known to require two (even three) symbols, this typo might lead to spurious (but harmless) keystrokes.

Over the years, I have crossed swords and stops with editors at McGraw-Hill, Sams, and Sybex to prevent spurious symbols appearing inside my precious strings. Of course, I had already accepted the aforementioned US rules governing the placement of stops with normal, non-computer-string quotations. British habits tend to be more logical: quotes usually contain only those elements that truly belong to the quote. (See Modern English Usage, H. W. Fowler, Oxford: Oxford University Press; 2nd ed.,page 591 et painstaking seq.)

My rehashing of the problem was triggered by a letter to the Editor from Dale B. Hymes of Lanham, Maryland:

Although I thoroughly enjoy UNIX Review, some serious grammatical errors throughout a recent issue have got me taking a second look at the subject matter presented in your magazine. I'll be the first to admit that I'm no expert, but there is one law of grammar that I'm sensitive to: The usage of quotation marks along with other punctuation. The following are the American conventions taught and STRONGLY enforced in this land's academic institutions:

1. With commas and periods - Always place commas and periods INSIDE the quotation marks. There are no exceptions to this rule.

2. Semicolons and colons - Always place semicolons and colons outside the quotation marks. There are no exceptions to this rule.

3. Question marks, exclamation points, and dashes - Place these marks inside the quotation marks when they apply to the quote only; or to the quote and the entire sentence at the same time. Place them outside the quotation marks when they apply to the entire sentence only."

Dale concludes by wondering if my earlier quibbles about punctuating literal strings somehow misled the UR editors when dealing with normal quotations. Without specific citations of the supposed errors, it is not easy for me to comment, except to query whether punctuation conventions and similar questions of style should be classified as grammar.

Now Dale does have a good point: when one encounters gross grammatical and orthographic solecisms, especially in computer literature, one is tempted, however unfairly, to distrust both the subject matter and the writer's general competence. After all, programmers (of all drudges) are martyrs to the cause of arbitrary, nit-picking rule-enslavement, so why should the time- honored laws of their mother tongue be so lightly disregarded? I would not, however, join those who write off the whole of H && S (giving the C: A Reference Guide of Harbison and Steele the same honor accorded to K && R) because "supersede" is spelled with, what else, a "c"! Note Dale's law #3 here, although the scoping rules for "!" lack a certain ANSI-type precision!

Here are some examples that tease or stretch Dale's rules.

Situation 1: A reports to B that C said "No." A wants to know about C's intonation; was it flat or interrogative? There are two basic ways for A to pose the question, but at least four ways to record it for posterity:

Did it sound like "No" or "No?" Did it sound like "No" or "No?"? Did it sound like "No?" or "No"? Did it sound like "No?" or "No?"

Situation 2: Consider four books: My Life My Life: My Life! and My Life?

God, what a book, "My Life!" Lord, what a book, "My Life"! Wow, that book, "My Life!"! Moses, that book, "My Life?"! Have you read "My Life?" No, but I've read "My Life!" What? You've read "My Life!?" No, I've only read "My Life." Oh the joys of "My Life": the plot, the characters, the punctuation!

Oh the joys of "My Life:": the plot, the characters, the punctuation!

I leave the reader to compare Dale's dogma (shared by the My Kinda Town style book) with the gentler views of Robert Brittain.

Postscript: Numerical Analysis can be FUN

Simpson's Rule: area h/3(y2 + 4y1 + y0) Mrs Simpson's Rule: Marry a prince.

Next month: Shopping around for a Mousepad with John Dvorak.


Stan's bio:

Liverpool-born Stan Kelly-Bootle has been exposed to computing, on and off and vice-versa, since 1953 when, after graduating in Pure Mathematics at Cambridge University, he switched to impure post-grad work on the wondrous EDSAC I. After some trenching with IBM and Univac in the 1960s and 70s, Stan opted for self-employment as a consultant, writer, folk-song revivalist, after-dinner entertainer, and cunning linguist.

His monthly DA ("Devil's Advocate") column ran and ran in UNIX Review (aka Performance Computing) from 1984 until January 2000 (a date that will live in infamy) but lives on as SODA ("Son of DA") via www.sarcheck.com the homepage devoted to UNIX performance.

The latest of his umpteen books are "The Computer Contradictionary" (MIT Press) and "UNIX Complete" (Sybex). More on his biblio- and disco-graphy can be found on http://www.feniks.com/skb/ soon due for its millennial update.

Stan welcomes reader reaction: skb@atdial.net

To the SarCheck home page

The URL of this page may change in the future. Please bookmark the home page instead of this one.

Portions © copyright Stan Kelly-Bootle 2004.
Portions © copyright Aptitune Corporation 1997-2004, all rights reserved.
This column is sponsored by Aptitune Corporation, makers of SarCheck. The opinions of Stan Kelly-Bootle are his own and do not necessarily reflect the opinions of Aptitune Corporation. Other products and companies referred to herein are trademarks or registered trademarks of their respective companies or mark holders.