Prev | Current Page 346 | Next

Peter Farrell-Vinay

"Manage Software Testing"


171
9
The Test Team and Its
Context
9.1 Recruiting the Test Team
The Gold Team
here was once a team of testers. They were a mix of programmers and operators. The
operators knew similar systems backwards, and the programmers knew how to build
systems and write scripts. They all knew that testing meant finding bugs early. So they
wrote lots of test scripts to test a bank system. In doing this they found lots of bugs in the
specifications themselves. (
???Which we know about anyway, and were just going to tell you about,???
said the management. Every time.) They liaised very closely with the developers ??” so closely that
some of them later
became
developers. And the management saw the scripts and were very pleased.
Then they started testing the first build, and the second, and the third. Then they stopped
and said,
???Aren??™t three untestable builds in a week overdoing it a bit????
and management twitched
a little, and they agreed that an experienced tester would check out each build before the team
loaded it onto all their machines. And another 10 builds were made before the team started
testing again.
And they found bugs galore. And developers fixed, and management sweated, and time passed.
And every deadline under the sun was missed, and customers got really edgy. And it got so bad
that the project manager slammed the test manager up against a wall and screamed,
???You??™re doing
it on purpose!??? ???Doing what???? ???Finding these errors!???
But senior management were under no illusions ??” releasing a poor build could provoke a rich
crop of very senior redundancies, because they had built badly in the past and the customers
wanted no more of it.


Pages:
334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358