🚀 go-pugleaf

In article <1172@meaddata.mead.UUCP> mead!gor...@uccba.uc.edu writes:
>In light of the stupid user discussion, how about a discussion of stupid and
>embarrassing programming errors.
>
While I attended University I worked at the Library Automations, primarily
doing applications with some systems programming stuff.  I had just
finished writing an interface between the big IBM's and a little
Honywell 5 running the Pick operating system, which was the university
check writing machine.  This was for the financial people at the library
and the University.  Someone in the accounting department notices that
they are loosing all kinds of money because the old application did not
handle some case or another, and because I had just been in there, I
was asked to provide some new whizzy stuff.

Everybody wanted this real bad, so I went into hacker mode and turned it
out in a few days, with extra bells and whistles that I could see they
would need soon.  It went into production after we proved that it would
do the job with the previous months data.  After that, everything
went real well for a month or so.

One day I walk into work, and there are the library and University 
MIS people, along with a couple University vice presidents, and they
are not happy.  They explain that the previous night a check had
been written for $216,420,138.90, and they wanted an explanation.

Turns out that someone decided to process a credit into the system by
writing a check for a negative amount.  My program didn't know what
to do with the '-', so it just deleted it from the ascii input,
effectively concatenating two amounts together; a $216.42, and a $138.90
with a leading zero.  University people have no sense of humor...

In article <1172@meaddata.mead.UUCP> mead!gor...@uccba.uc.edu writes:
>In light of the stupid user discussion, how about a discussion of stupid and
>embarrassing programming errors.
[Story of a recursive function overflowing the stack on a long input.]

There are programming mistakes that can result, not from stupidity,
but simply from moving to a different type of programming language.  A
few years ago a friend at school showed me a program that had worked
fine when she'd tested it, but came back from grading with the
supremely unhelpful message, "This dumps core."  The assignment was a
short filter written in C.

The problem was that her previous programming course had used Scheme,
emphasizing general programming techniques.  Scheme guarantees that it
will compile a tail-recursive function into a loop.  Her filter code,
while beautifully written, was handling the main loop of the filter
tail-recursively, and the program overgrew the stack as soon as the TA
gave it too many characters of input.

She looked rather shocked when I told her that most C implementations
wouldn't handle that case nicely, but she recoded it into a while-loop.
--
============================================================================
David Wald                                           wa...@theory.lcs.mit.edu
============================================================================

An acquintance of mine was hired to speed up a COBOL accounting program
which used some 8 hours to process some 1000's of transactions. She
discovered that the program processed the first transaction, then
REWOUND the file, sought the second, bypassing the first, rewound,
sought the third, etc.. Instead of some 1000's of file operations the
program performed some 1000000's!
-- 
Matti Nyk{nen			CS Student at Helsinki U, Finland
Trepanation!			email: mnyk...@cc.helsinki.fi