|Title||Myths, Over-Simplifications and Unconfirmed Claims in Software Engineering|
|Afilliation||Software Engineering, Software Engineering|
|Publication Type||Talk, keynote|
|Year of Publication||2014|
|Location of Talk||Keynote at: EUROMICRO DSD/SEAA, Verona, Italy|
The software engineering discipline contains numerous myths, over-simplifications and unconfirmed claims. Some of them may be harmless, but others may hamper the use of good practices. Examples of myths, over-simplifications and unconfirmed claims are presented to demonstrate the mechanisms contributing to the creation, spread and survival of them. Ten years ago we (Barbara Kitchenham, Tore Dybå and myself) introduced evidence-based software engineering. The goal was to support a move towards a discipline with practices based on valid scientific and experience-based evidence, i.e., to reduce the influence from myths, over-simplifications and unconfirmed claims. I report from my experience in providing training in the use of evidence-based practices to software professionals and computer science students. The experience includes the need to emphasize and train software professionals in the collection and evaluation of practice-based experience and in the design and completion of studies within their own context, not so much in collecting and evaluating research-based evidence.