>Jess, I would agree it was a little harsh. Until Craig can
>prove to us that he got a perfect 5.0 out of 5.0, the test by
>definition was challenging to him.
>
>One excuse would be some questions/answers were confusing.
>However, I would say that is the case with the M$ test more
>so than this one. No multiple questions about Visual Source
>Safe at Brainbench. How is VSS related to VFP when that developer
>works alone?
>

Version control is vitally important during the project lifecycle, if for no other reason than to be able to determine the exact sets of changes between specific version of a project or back out poor implementation decisions. And in most cases, I'd expect someone who claimed to have mastered the product to be familiar with all aspects of project lifecycle and development. But perhaps that's more than a reasonable expectation of what the claim of being a master-level VFP developer represents...certainly the claim that somneone was a competent mentor for VFP developers with a score of 4 was not an accurate assessment, since in no way was the capacity to explain the concepts behoind VFP to other less skilled people tested. The evaluation of capability based on the score received was not accurate.

The logic that lies behind the topics presented or excluded for this exam could be used to include or exclude any range of topic; what you're saying is that this exam tested what you're fairly confident of having mastery of now, OTOH, we really have no idea of what it tests was given to us before or after the exam. It did not state in any way, shape or form what specific skill sets were tested or were required, which makes the test about as meaningful as a used piece of tissue.

Face it, the average cadaver with a couple of days coaching could score in the low 4s on that test! With no standards of question quality, answer accuracy, or subject topics represented by the exam, it's a complete fraud.

>Come on Craig - we know you are one of the better VFP developers
>out there. I say that with great sincerity. You helped
>me out with an NT network error several months ago.
>
>Put your money where your mouth is and show me your perfect score...

I'm sure he will once you show him a pass of 155 and 156; he already passed those. The BrainBench exam is not an adequate test of mastery of the product; question quality was poor, and at least in the beta, two of the questions repeated. The range of tested material was weak, and the question posing mechanism was extremely limited compared to the MCP exams. The taxonomical classification of the type of knowledge testing (cf "Bloom's Taxonomy") was significantly lower than what was expected for the MCP exams. Several of the BrainBench beta questions did not have a clear answer; wording of the questions was poor in some cases. Certainly the BrainBench exam's expectations of required abilities were not based on the criteria established for the 155/156 exams, and the distribution of questions across the breadth of the domain of coverage was not based on the extensive studies that MS did to determine the topics that an MCP candidate should have mastered in order to earn certification.

To give a fair perspective, both Jim Booth and I were involved in the development of the MCP exams. There were very stringent requirements for question distribution, presentation and accuracy, which were checked and rechecked by other SMEs as well as the VFP development team. The exact range of subject material expected for passing the exam was given in detail ahead of time - nothing on the final cut of the MCP exam was not mentioned in the range of topics covered and laid out ahead of time. And passing the MCP represents a well-defined set of acquired skills.

Believe me, a large number of the questions on this exam would not have met the question requirements for the MCP exam. You can't really state what set of skills were needed to pass this, or what the relative values of a pass with a 4 represents vs a pass with a 3 or 5.

YMMV. The BrainBench exam is a walk in the park compared to the MCPs. Without some statement of needed skills, or clarification and validation of the test material, it just isn't useful as a certification tool. I certainly take exception to the statement made in the beta exam that someone with a score of 4.14 (better than 67% of those tested to date) "Demonstrates a clear understanding of many advanced concepts within this topic. Appears capable of mentoring others on most projects in this area." That's a crock of cowpies. No significant thought was needed to determine the right answer - I took about 36 seconds/question when I took it on 10mg of hydrocodone the day after surgery, and I couldn't have drooled on myself with any degree of accuracy that afternoon. Certainly I didn't see evidence of mastery of advanced concepts from my POV, but perhaps my POV of what we should expect from the average VFP developer differs from yours.

No worries...enjoy your score if it makes you feel good.