Rules, Concepts and the Dreyfus Model

[textile]A friend of mine, Kerry Jones, recently drew my attention to "an article by Dave 'pragmatic' Thomas": Dave references the 'Dreyfus Model of Skills Acquisition', a model used in training medical professionals.

There are five levels, from Novice to Expert:

* Novice
* Advanced Beginner
* Competent
* Proficient
* Expert

(I found this "PDF on the Dreyfus Model": useful in better understanding the levels)

Dave explains...

...high-quality experience turns us from a novice to something approaching an expert.

One thing that Dave observed really hit home for me...

We process information in a qualitatively different way in the two states: novices need rules, experts need concepts and contexts. Novices need external feedback, experts generate their own. Stuff that works when teaching a novice drives an expert up the wall, and vice-versa. Experts are often the worst teachers of novices.

This has helped me to better understand some of the long-standing arguments in the community, such as "automate all tests vs. you can't automate all tests" or "follow best practices vs. there are only good practices in context". I've believed for a long time that these arguments weren't about the subjects being discussed but actually about two different philosophies... "commandments vs. concepts".

I now see it as being about targeting two different audiences: Novice-to-Competent or Competent-to-Expert.

So how do I provide rules to the novices without annoying the experts whilst providing concepts to experts without annoying the novices? Hmmm...

Browse by topic:


I think there is insight there, but I'm not sure I agree that we should go so far as providing commandments to novices... at least not *those* commandments.

The reason I say that is because I believe that testing is a scientific art (or maybe an artistic science, I'm still kicking around which should go first). Either way testing involves creativity, even at the novice level. We simply can't teach testers things with names like "best practices" without quashing that creativity and driving those with the potential to become experts out before they even make it to advanced beginner. Maybe they do need rules, I don't know, but I do know that *if* we are going to give them rules, they'd better be good rules. Rules that the "experts" may cringe at, but not ones that make them break out in hives and have heart attacks.

On the flip side, what does this theory say about all the self-proclaimed "experts" who try to shove commandments down the throats of those who have been granted "expert status" by the industry and their peers?

Scott Barber
Chief Technologist, PerfTestPlus
Executive Director, Association for Software Testing

For the same reasons I had resisted idea (advocated by James Bach) that novices should do exploratory testing. I gave up a year ago or so. You know what - I prefer motivated novices over seasoned scripted testers.
Situational leadership describes the same phenomena of novice who

Generally lacking the specific skills required for the job in hand, and lacks any confidence and / or motivation to tackle it.

However, I just realized following: while blogging I generally target Competent-to-Expert. I believe beginner wouldn't read it. I think providing concepts to novices frighten them (annoyance is only sub-reaction).