A testing hierarchy?

From Javaworld

The next level is populated by performance testers.

Performance testing has something to do with concurrency and achieving good performance is hard and requires intimate knowledge of shared variables, thread locals, monitors, java.util.concurrent.*, etc. A performance tester or performance engineer may even be responsible for tuning the JVM garbage collector and that’s really a profession in its own.

At the bottom of the pyramid we have functional (system) testers and software quality assurance people.

Functional testers are meant to execute only GUI-based test tools, so their job is doing not more than performing a few clicks and comparing the results. QA people has the smell of “controlling real programmers” and that’s not a good reputation in any case. QA people are running, e.g., “static code analyzers” and real developers have to fight with all the “false positives” that may result of such a “big brother is watching you” procedure. Moreover, all these QA people that want to tell real developers what’s wrong with their code: did they ever implemented a large application when being pressed for time by the project leader? Or are they only “quality administrators”, just being able to start Findbugs or Checkstyle for a couple of JAR files (or PC-Lint in the C area) thus producing endless reports nobody is really interested in? BTW: When I’m talking about “testers” I mean people who are executing tests, not famous people who are talking about testing on so much test conferences that you may wonder when they have time do do real testing.