File: MURPHY.TX of Tape: Sources/Other/new-16
(Source file text)
EDSEL MUSPHY'S LAWS:
GENERAL:
1. IN ANY FIELD OF SCIENTIFIC ENDEAVOR. ANYTHING THAT CAN GO WRONG, WILL.
2. IF THE POSSIBILITY EXISTS OF SEVERAL THINGS GOING WRONG, THE ONE THAT
WILL GO WRONG IS THE ONE THAT WILL DO THE MOST DAMAGE.
3. IF NOTHING CAN GO WRONG, SOMETHING WILL.
4. LEFT TO THEMSELVES, THINGS ALWAYS GO FROM BAD TO WORSE.
5. NATURE ALWAYS SIDES WITH THE HIDDEN FLAW.
6. GIVEN THE MOST INAPPROPRIATE TIME FOR SOMETHING TO GO WRONG, THAT'S
WHEN IT WILL OCCUR.
7. MOTHER NATURE IS A BITCH.
8. IF EVERYTHING SEEMS TO BE GOING WELL, YOU HAVE OBVIOUSLY OVERLOOKED
SOMETHING.
9. NEVER MAKE ANYTHING SIMPLE AND EFFICIENT WHEN A WAY CAN BE FOUND TO
MAKE IT COMPLEX AND WONDERFUL.
10. IF IT DOESN'T FIT USE A BIGGER HAMMER.
11. IN AN INSTRUMENT OR DEVICE CHARACTERIZED BY A NUMBER OF PLUS-OR-MINUS
ERRORS, THE TOTAL ERROR WILL BE THE SUM OF ALL THE ERRORS ADDING IN
THE SAME DIRECTION.
12. IN ANY GIVEN CALCULATION, THE FAULT WILL NEVER BE PLACED IF MORE THAN
ONE PERSON IS INVOLVED.
13. ALL WARRANTY AND GAURANTEE CLAUSES BECOME VOID UPON FULL PAYMENT OF THE
FINAL INVOICE.
DESIGN:
1. IN ANY GIVEN PRICE ESTIMATE, COST OF EQUIPMENT WILL EXCEED ESTIMATED
EXPENDITURE BY A FACTOR OF 3.
2. DIMENSIONS WILL ALWAYS BE EXPRESSED IN THE LEAST USEABLE TERMS. FOR
EXAMPLE, VELOCITY WILL BE EXPRESSED IN FURLONGS PER FORTNIGHT.
3. IF THE BREADBOX TRIAL MODEL FUNCTIONS PERFECTLY, THE FINISHED PRODUCT
WILL NOT PERCOLATE.
4. IN A MATHEMATICAL CALCULATION. ANY ERROR THAT CAN CREEP IN, WILL DO SO,
AND IT WILL BE IN THE DIRECTION THAT WILL DO THE MOST DAMAGE TO THE
CALCULATION.
5. IN ANY GIVEN COMPUTATION, THE FIGURE THAT IS MOST OBVIOUSLY CORRECT WILL
BE THE SOURCE OF THE ERROR.
6. THE PROBABILITY OF A DIMENSION OR VALUE BEING OMITTED FROM A DRAWING IS
DIRECTLY PROPORTIONAL TO ITS IMPORTANCE.
7. IN SPECIFICATIONS, MURPHY'S LAW SUPERCEDES OHM'S.
ASSEMBLY:
1. IF A PROJECT REQUIRES "N" PARTS, THERE WILL BE "N-1" PARTS AVAILABLE.
2. INTERCHANGEABLE PARTS WON'T.
3. PARTS THAT MUST NOT AND CANNOT BE ASSEMBLED IMPROPERLY WILL BE.
4. THE MOST DELICATE COMPONENT WILL BE DROPPED.
5. THE CONSTRUCTION AND OPERATION MANUAL WILL BE DISCARDED WITH THE PACKING
MATERIAL, AND THE GARBAGE TRUCK WILL HAVE PICKED IT UP FIVE MINUTES
BEFORE THE MAD DASH TO THE RUBBISH CAN.
6. THE NECESSITY OF MAKING A MAJOR DESIGN CHANGE INCREASES AS THE ASSEMBLY
AND WIRING OF THE UNIT APPROACH COMPLETION.
7. A DROPPED TOOL WILL LAND WHERE IT CAN DO THE MOST DAMAGE (ALSO KNOWN AS
THE LAW OF SELECTIVE GRAVITATION).
8. A PART SELECTED AT RANDOM FROM A GROUP HAVING 99% RELIABILITY WILL BE A
MEMBER OF THE 1% GROUP.
9. TOLERANCES WILL ACCUMULATE UNIDIRECTIONALLY TOWARD MAXIMUM DIFFICULTY
OF ASSEMBLY.
10. THE AVAILABILITY OF A PART IS INVERSLLY PROPORTIONAL TO THE NEED FOR
THAT PART.
11. IF A PARTICULAR RESISTANCE IS NEEDED, THAT VALUE WILL NOT BE AVAILABLE.
FURTHERMORE, IT WILL BE IMPOSSIBLE TO DEVELOPE SUCH A RESISTANCE WITH
ANY SET OF SERIES AND/OR PARALLEL COMBINATIONS OF RESISTANCES.
12. AFTER AN INSTRUMENT HAS BEEN ASSEMBLED, EXTRA PARTS WILL BE FOUND ON THE
BENCH.
WIRING:
1. ANY WIRE CUT TO LENGTH WILL BE TOO SHORT.
2. MILLIAMMETERS WILL BE CONNECTED ACROSS THE POWER SOURCE, WHILE VOLTMETERS
WILL BE CONNECTED IN SERIES WITH IT.
3. THE PROBABILITY OF AN ERROR IN THE SCHEMATIC IS DIRECTLY PROPORTIONAL TO
THE TROUBLE THAT IT CAN CAUSE.
TEST:
1. IDENTICAL UNITS TESTED UNDER IDENTICAL CONDITIONS WILL NOT BEHAVE IDENTICALLY
ON THE FINAL TEST, AFTER HAVING BEEN BURIED UNDER SEVERAL OTHER PARTS
AND MASSES OF WIRE.
2. A SELF-STARTING OSCILLATOR WON'T.
3. A CRYSTAL OSCILLATOR WILL OSCILLATE AT THE WRONG FREQUENCY --- THAT IS IF
IT OSCILLATES AT ALL.
4. A P-N-P TRANSISTOR WILL BE FOUND TO BE A N-P-N
5. A FAIL-SAFE CIRCUIT WILL DESTROY OTHERS.
OPERATION:
1. IF A CIRCUIT CANNOT FAIL, IT WILL.
2. A TRANSITOR PROTECTED BY A FAST-ACTING FUSE WILL PROTECT THE FUSE BY
BLOWING FIRST.
3. THE PROBABILITY OF FAILURE OF A PART IS INVERSELY PROPORTIONAL TO THE
EASE OF REPAIR OR REPLACEMENT.
TROUBLE SHOOTING:
1. AFTER THE 24TH CABINET-TO-CABINET SCREW HAS BEEN REMOVED TO REPLACE THE
UNDER-CHASSIS FUSE, IT WILL BE OBSERVED THAT THE LINE CORD PLUG HAS
BEEN DISENGAGED FROM THE A.C. RECEPTACLE.
2. AFTER THE 24TH CABINET-TO-CABINET SCREW HAS BEEN ASSEMBLED. THE DRIVER
TUBE WILL BE FOUND UNDER THE SCHEMATIC ON THE BENCH.
3. THE BLEEDER RESISTOR WILL QUIT DISCHARGING THE FILTER CAPACITORS AS THE
OPERATOR REACHES INTO THE POWER SUPPLY ENCLOSURE.
FINAGLE'S LAWS:
1. NO MATTER WHAT RESULT IS ANTICIPATED, THERE IS ALWAYS SOMEONE WILLING
TO FAKE IT.
2. NO MATTER WHAT THE RESULT, THERE IS ALWAYS SOMEONE EAGER TO MISINTERPRET IT.
3. NO MATTER WHAT HAPPENS, THERE IS ALWAYS SOMEONE WHO BELIEVES IT HAPPENED
ACCORDING TO HIS PET THEORY.
FINAGLE'S CREED:
SCIENCE IS TRUTH; DON'T BE MISLED BY FACTS.
ALLEN'S AXIOM:
WHEN ALL ELSE FAILS, READ THE DIRECTIONS.
GUNNERSEN'S LAW:
THE PROBABILITY OF A GIVEN EVENT OCCURING IS INVERSELY PROPORTIONAL
TO ITS DESIRABILITY.
GLASSER'S COROLLARY:
IF, OF THE SEVEN HOURS YOU SPEND AT WORK, SIX HOURS AND 55 MINUTES
ARE SPENT WORKING AT YOUR DESK, AND THE REST OF THE TIME YOU THROW
THE BULL WITH YOU CUBICLE-MATE, THE TIME AT WHICH YOUR SUPERVISOR
WILL WALK IN AND ASK WHAT YOU'RE DOING CAN BE DETERMINED TO WITHIN
FIVE MINUTES.
FINAGLE'S FACTOR:
(SOMETIMES CALLED THE SWAG -- SCIENTIFIC WILD-ASSED GUESS CONSTANT)
THAT QUANTITY WHICH, WHEN MULTIPLIED BY,DIVIDED BY, ADDED TO, OR
SUBTRACTED FROM THE ANSWER WHICH YOU GOT, YIELDS THE ANSWER THAT
SHOULD HAVE GOTTEN.
FURTHER HINTS ON WRITE-UPS
1. IN ANY COLLECTION OF DATA, THE FIGURES THAT MOST CLOSELY CONFIRM THE
THEORY ARE WRONG.
2. NO ONE YOU ASK FOR HELP WILL SEE THE MISTAKES EITHER.
3. EVERYBODY WHO STOPS BY WITH UNSOUGHT FOR ADVICE WILL SEE THE MISTAKES.
4. AN EXPERIMENT MAY BE CONSIDERED SUCCESFULL IF NO MORE THAN HALF THE DATA
MUST BE DISCARDED TO AGREE WITH THE THEORY.
5. NO EXPERIMENT IS EVER A COMPLETE FAILURE. IT CAN ALWAYS SERVE AS A BAD
EXAMPLE.
PECK'S PROGRAMMING POSTULATES:
1. IN ANY PROGRAM, ANY ERROR WHICH CAN CREEP IN WILL EVENTUALLY DO SO.
2. NOT UNTIL THE PROGRAMMING HAS BEEN IN PRODUCTION FOR AT LEAST SIX
MONTHS WILL THE MOST HARMFUL ERROR BE DISCOVERED.
3. ANY CONSTANTS, LIMITS, AND/OR TIMING FORMULAE THAT APPEAR IN THE COMPUTER
MANUFACTURER'S LITERATURE SHOULD BE TREATED AS VARIABLES.
4. THE GREATEST MOST VITAL PARAMETER IN ANY SUBROUTINE STANDS THE GREATEST
CHANCE OF BEING LEFT OUT OF THE CALLING SEQUENCE.
5. IF ONLY ONE COMPILER CAN BE SECURED FOR A PIECE OF HARDWARE, THE
COMPILATION TIMES WILL BE EXORBITANT.
6. JOB CONTROL CARDS THAT POSITIVELY CANNOT AND/OR MUST NOT BE ARRANGED
IN IMPROPER ORDER WILL BE.
7. IF A TEST INSTALLATION FUMCTIONS PERFECTLY, ALL SUBSEQUENT SYSTEMS
WILL MALFUNCTION.
8. INTERCHANGEABLE PARTS WON'T.
9. IF MORE THAN ONE PERSON HAS PROGRAMMED A MALFUNCTIONING ROUTINE,
NO ONE IS AT FAULT.
10. IF THE INPUT EDITOR HAS BEEN DESIGNED TO REJECT ALL BAD INPUT, AN
INGENIOUS IDIOT WILL DISCOVER A METHOD TO GET BAD DATA PAST IT.
11. DUPLICATED OBJECT TAPES WHICH TEST IN IDENTICAL FASHION WILL NOT GIVE
IDENTICAL RESULTS AT REMOTE SITES.
12. MANUFACTURER'S HARDWARE AND SOFTWARE SUPPORT CEASES WITH PAYMENT OF THE
COMPUTERS LAST INSTALLMENT.
13. AT LEAST ONE CRITICAL TEST TAPE WILL BE LOST, MISPLACED , DESTROYED
OR WRITTEN OVER.
14. WHAT GOES UP MUST COME DOWN -- AND CAN BE EXPECTED TO DO SO IN THE
MIDDLE OF YOUR JOB.