Re: cobol code assessment

Louis Krupp wrote:

>
> Slightly OT, but is average experience necessarily a meaningful
> metric?

It's considerably more meaninful than not mentioning the skill sets at all.

> Are ten people who can claim a year each of COBOL experience
> really equivalent to one person who can claim ten years, or two
> people who can claim five each? Is ten years' experience worth twice
> as much as five years, or is there a point of diminishing returns?
>
> This question calls for some handwaving: I would sum a*log(b*e + 1)
> over the whole team, where 'e' is years of experience and 'a' and 'b'
> are constants of your choice, and use the result as a measure of
> available skill. 'a' and 'b' could be different for different skills;
> for years spent writing spaghetti COBOL with ALTER statements or
> FORTRAN with arithmetic IFs and Hollerith editing descriptors, 'b'
> could be very small or even zero while 'a' could be zero or even
> negative.

"Average" is probably understandable by the factotum who assigned this
absurd exercise. I'll wager "Median" is beyond his grasp and regression
analysis (or, for that matter, your in-depth study) is an example of the
perfect being the enemy of the good.

Hell, I bet he couldn't even grasp
Ln(e) = Sum(1/n**2) + (sin**2 (theta) + cos**2 (theta))



.

Relevant Pages

  • Re: Well done Netherlands
    ... everything I could grasp from the cricinfo comms (and what I know ... about some of the Dutch players) suggests they have enough skill and ...
    (rec.sport.cricket)
  • Re: Well done Netherlands
    ... everything I could grasp from the cricinfo comms (and what I know ... about some of the Dutch players) suggests they have enough skill and ...
    (rec.sport.cricket)