>Speaking personally... I am a better developer today because I studied for and passed the VFP MCP exams. The exams forced me to review what I thought I knew and to learn things that I hadn't gotten around to yet.

Evan - I have never even looked at the newest VFP/VS exams - but based on past ones - the problem lies in the multiple choice aspect of it. I happen to be a good test taker - and will usually do much better on such a test than my knowledge of the subject would dictate. On the other hand, there are those who are simply not good test takers and do worse than what they really know.

The only way I would see an exam being a fairly realistic assesment would be with the following criteria:

1 - it is completely open book, open VFP, open help files, open UT, web, etc. etc. etc. Why? because the ability to rapidly find, discern, test, and apply information is more important than the ability to memorize it. Ability to use all resources available to rapidly get answers is one of the greatest skills a developer (or anyone in business) can have.

2 - it is timed - with specific tasks/questions also timed. This separates those who know what they are doing from those who are only looking up answers. If you do not have a good mastery of the principles involved, you will get wiped out by this criteria.

3 - It should be based on solving specific business problems - not on specific knowledge of commands, syntax, etc. Note that there could be more than one right answer with this.

4 - "grading" should be based on how many tasks were completed and how well the solution works in terms of addressing the problem, performance, accuracy and completeness, clean code, etc. most of these are subjective, but some benchmarks of performance, code 9app) size, etc. could be used.

My .02,