Re. Certification

[Geoff Hart <geoff-h -at- MTL -dot- FERIC -dot- CA>]

 I suggested some time ago that the easiest form of
 certification would be to formally evaluate someone's body
 of work. The evalution would consist of the following
 steps:

 1. Obtain a work sample (or several for more senior or
 prestigious levels or grades of certification).

 2. Identify (through the employer?) what the candidate's
 role was in producing the work.

 3. Identify the goals of the work and the constraints under
 which the person worked.

 4. Determine the extent to which the candidate met each of
 these goals, in the context of the existing constraints.

 The evaluation would be performed by a committee of peers,
 perhaps including audience members to provide a reality
 check, and the certificate would be awarded or denied based
 on the committee's consensus. There would likely be a right
 of appeal (perhaps to a second committee) too. There are a
 few problems to work out in this scheme (e.g., how to
 determine objectively if the candidate met a criterion),
 but the overall outline seems robust.

 I'd submit to such an examination willingly. For example,
 most of the technical reports that I edit are handled
 rigorously, and stand on their own without need for
 defence; on the other hand, a newsletter we publish gets
 put out every year under very tight deadlines, with a
 format I'm not fond of, and with lots of other problems; I
 wouldn't want to be evaluated on that publication without
 being able to defend myself for several aspects that I
 consider substandard; I'd get my certificate for the tech.
 report, but not for the newsletter. C'est la guerre!

 --Geoff Hart    @8^{)}
 geoff-h -at- mtl -dot- feric -dot- ca

 Disclaimer: If I didn't commit it in print in one of our
 reports, it don't represent FERIC's opinion.