Subject: | |
From: | |
Reply To: | Pickett, Michael |
Date: | Thu, 20 Mar 1997 12:07:00 +0000 |
Content-Type: | TEXT/PLAIN |
Parts/Attachments: |
|
|
(I missed the start of this thread because my mail host appears to have lost
everything since last Friday, but here goes ...)
I interviewed a chap recently, and I asked him about his approach to
debugging. He told me that he rarely used a debugger, finding it much more
effective to examine his sources and, from them, to understand why an error
occurred, rather than poke around to see what was happening.
I discovered that he was ex-army, trained in mine clearance.
Perhaps I should have a word with our training department. :-)
--Michael--
------------------------------------------------------------------------------
REPLY FROM: Pickett, Michael
FROM: Stephen Schwarm (SMTP:[log in to unmask])
TO: SMTP:[log in to unmask]
SUBJECT: Re: Info request: Code development cycle Ada vs C++
------------------------------------------------------------------------------
I always defined a hacker as a person who enters:
main()
{
}
with an editor, compiles and then envokes the debugger to do
the real work.
I remember in a previous job when someone came to me to get
help running the debugger because I was the UNIX expert and were
very suprised to find out I could not answer their question
because I used it so little. In that case I was writing
in C. I delivered about 3.5KSOCs and made 38 errors. 37
of them would have been caught for me by an Ada or Pascal
compiler. BTW all the errors were caught in unit test.
Steve Schwarm
|
|
|