🚀 go-pugleaf

I see two drawbacks:

First, a false sense of security.  You may ensure that a programmer CAN
program in a competent way (or at least pass whatever test the state
comes up with) but not that said programmer WILL consistently act com-
petently.  I smell malpractice and malpractice insurance for programmers
on the horizon, with associated bloating of software production costs.

Second, how do we ensure that (a) the programmer certified competent in
year N will be competent to program in year N+10 or N+20?  Do we set an
expiration date on these licenses? and (b) that a programmer fresh out of
grad school with the latest in software methods instilled in his head will
not fail the certification exam because the material with which he has
worked has not reached the examiners yet?  As a side issue, who decides
what competent programmer's must know?  And how will the requirements affect
industry?  And how does one administer a truly meaningful competency exam
without extending it over such a long time period that the risk of cheating
becomes unacceptably high?

Ciao,
  Mike

--
Mike Zraly				``If you give me six lines written
mzraly@ldbvax.dnet.lotus.com		  by the most honest man, I will find
or c/o seaotter@athena.mit.edu		  something in them to hang him.''