Prev | Current Page 80 | Next

Peter Farrell-Vinay

"Manage Software Testing"


1.5 Something to Do When You Really Hit Opposition
Here are some of the things they don??™t want to look at (and you do). Use these as the basis of a report
to provide an outline for discussing
1. How much newer projects have slipped as a result of having to pull people off them in order to
try to fix the mess created during older ones.
2. The cost of fixing post-release bugs, possibly because that doesn??™t come out of their budget or
they don??™t care how frustrated the developers get in having to build patches rather than design
new bits of the system. Get the data from time sheets. Look at:
??? The equation in section 18.13
??? Table 3.4
??? Section 7.4.6
??? Section 14.8.2
4
I??™m sorry. We had this tacit agreement that we were going to use the English language, and here I go breaking
it. What I
think
the person who said this meant, was that writing requirements specifications was something he
associated with high-ceremony, process-oriented companies. And that he wanted his company to become just like
that really fast. Er, next year.
6
Manage Software Testing
3. The relationship between stopping testing ???too early??? and delaying getting the next release out
(because the developers are spending all their time bug-fixing). Measure this by looking at the
bug detection curve, the test runs, the time sheets, the overtime, and the 3rd-line support calls
(problems per user month). If it looks like the one in Figure 1.


Pages:
68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92