software development, QA, and the tech writer

[Cathy Carr <ccarr -at- OVID -dot- COM>]

To:   techwr-l @ listserv.okstate.edu
cc:
Subject:  software development, QA, and the tech writer


"Testing every window and button will tell you whether those windows
and buttons work, but not whether the software can be used to perform
useful work.  For example, it won't tell you if a developer forgot to
(or decided not to) implement a key feature."

Starting out doing software documentation, it was a real shocker to me how
determined (or stubborn--pick your word) some programmers can be about
implementing the features *they* like the way *they've* decided is best.
Once one of my supervisors was reviewing a set of release notes I'd
prepared.  She picked out one feature and said, "That's not how this
works."  "But it is," I said, "John told me so."  (Names have been changed
to protect the innocent.)  She stared at the paragraph a moment, then
picked up the phone and called the programmer and--as one might say--got
him back with the program, in a few well-chosen words.  I returned to my
cube and John was waiting for me, so to speak.  "You shouldn't have told
her," he said, perhaps not altogether logically.  "It works better the way
I did it."  (As if I could just *sneak* a feature by in the release notes,
for crikey's sake, or would do it if I could.)  He was genuinely annoyed at
me for several days.  Talk about killing the messenger!

Cathy