Evidence-based cost estimation better-quality for software

Our work on COSEEKMO is hardly enough to change experimental methods in the cost estimation community. So, we're also running a new workshop series called PROMISE devoted to repeatable software engineering experiments. Evidence-based reasoning is becoming common in many fields. Evidence-based approaches demand that, among other things, practitioners systematically track down the best evidence relating to some practice; critically relating to some practice; critically appraise that evidence for validity, impact, and applicability; and carefully document it. The software community can bring to bear many methods that could further improve evidence-based cost estimation.