I just took the test. It did for me what I thought it would do, it
made me read the docs from cover to cover (very quickly, and not the
appendices) with a firm deadline. If I hadn't scheduled the test, I
would not have done that. I learned some things.
Here are my comments on the test as an evaluation of someone's
knowledge of PostgreSQL:
1. The bar is set a little low at 64%. I would think the passing
score ought to be a bit higher.
2. The coverage of tasks and knowledge as advertised was pretty appropriate.
3. There were too many typographical errors. However, I was able to
reason through them, to the extent that the typo would have meant no
answer was correct, and at least one answer had to be. For example,
two tables are shown with their data. One has a field "t1" and the
other has a field "ti". The answers would all have been wrong had
they not both been t1.
4. There is too much improper grammar and misuse of terms. This test
was obviously translated. A native English speaker should have been
able to clean it up pretty easily, but I assume there is a huge
question bank, so a few slipped through. What is the feedback loop
for this? I wish I could have left comments for each question
indicating how it could have been written better. It didn't bother me
that much, because I could usually deduce the meaning, but there were
a couple that it really made difficult. A person who is impatient
with imprecise grammar might have gotten upset.
On the issue of "What is certification worth?" I don't have a
technical degree. All I can put on my resume is what I have
accomplished on the job, which prospective employers know is not an
impartial evaluation of my skills. They can call references, but
again, they may just be trying to get rid of me. The only unequivocal
assessment of someone's knowledge (at a point in time, anyway) is a
degree or certification. By themselves they don't say much, but I
think in conjunction with experience they are seen as a good thing by
PHBs.
Thanks to SRA for making this happen!
- Ian