escaping from low bug counts


Daisy presents a persuasive argument for focusing on test coverage discussions when bug counts are low:

http://quality-assurance-software-testing.blogspot.com/2005/08/bug-count-vs-test-coverage.html

Comments (3)

  1. MSDNArchive says:

    Probably the easiest way to avoid low bug counts is not to fix any. Also, to hire incompetent developers. I guess testers have a different perspective.

  2. Joe Chung says:

    Derrick, the idea is to improve the quality of your software, not make it worse. What Eric and Daisy are talking about is helping improve software quality by expanding the scope of what your tests are testing.

    I have never liked bug counts as a metric of tester performance, because (a) it encourages nitpicking and (b) when it comes to crunch time, developers and testers alike start signing off bugs by finding "innovative" ways to close bugs instead of fix them.

    You have to be careful with test coverage too. Test coverage creep can be as insidious as feature creep.

  3. … for Architects

    Nick Malik – Enterprise Architecture Agility

    Roy Osherove – [Audio Interview] Ingo…