SWELL people visits the International Conference on Software Testing in Denver this week. More info about interesting new results will be coming shortly on this blog. On the first day Paul Ammann gave a tutorial on "Software Testing - A Newcomer's guide". It is a good overview of the area and defines the basic concepts. You can find the slides here:

http://cs.gmu.edu/~pammann/icst/ICST_Tutorial.pptx

On the last slides he also goes through the sessions and papers in the conference. This gives a good overview of what is happening in the field.

---
SWELL is a National Innovation Driver for Software V&V funded by Vinnova and a number of Swedish universities and companies.
http://www.swell.se

Fredrik Scheja
5/24/2009 09:51:17 pm

OK, sorry for my late response, but here it is. Paul's slides defines four different tasks, test design, automation, execution, and evaluation. Yes, fine, but then he states: "Why do test organizations still use the same people for all four test activities?? - This is clearly a waste of resources". I can understand why he states that, but in general I would clearly disagree! From my experience, the tester´s work is both more entertaining and effeicient if the tester does the design-automation-execution and evaluation. What would a car designer learn from his design och enjoy from his work, if he never gets to experience the end product by himself?! And also, in every handover from one resource to another, there will always be misunderstandings and misinterpretations from a designer to an automator to an executor to an evaluator....it is like the infamous swedish "viskleken" (Whisper game). Therefore, I would say that it is a dangerous statement that Paul presents and it is not at all clear to me what he states.

Reply
Robert Feldt
5/24/2009 10:06:00 pm

We can't be sure since these are only slides; maybe he presented it more clear. However, I can agree with you to some extent, Fredrik. I think we should avoid creating "better/higher status testers" and "lower level/status testers" with corresponding salary differences à la Architect/Designer/Coder and view it more holistically with a team of people sharing a responsibility. OTOH, with your/my argumentation why should development and testing be separate activities; there is a risk misunderstandings... ;) Also I think Paul's statement can be seen in the light of a general quest of broadening the view on testing activities and thus allowing for different roles and specialities. So it could be seen as a case for a maturing sub-discipline.

Reply
Fredrik Scheja
5/24/2009 11:45:21 pm

Ah, yes! I see your point. And I have thought about this tester/developer issue earlier.....I think, in the future some testers will also fix the problems they find, decreasing the possible risk of misunderstandings in bug reporting. I see our profession in the future devided into different expertises, similar to how the medical doctors work in the hospital.....I know that you, Robert, don´t like when I campare V&V profession with other professions, but hear me out... =) Some doctors are generalists, and are very good at performing diagnosis (testing) and easy fixes (stop a bleeding or sign a reciept for losec), but sometimes he/she has to send the patient to a surgeon, with a expertise in heart deseases...Like when a tester finds a problem in the real time kernel and have to send it to an architect....are you following me?

Reply

Your comment will be posted after it is approved.


Leave a Reply.