I'm up for it.

On Thu, Jul 17, 2008 at 1:28 PM, Alan D. Cabrera <[EMAIL PROTECTED]>
wrote:

> I have a crazy idea that I have always wanted to try out just for fun.  If
> you don't want to participate feel free to ignore this thread since this
> experiment will be non-binding.
>
> This is not the first time that I've seen disagreement over feature sets
> and priorities.  I'm sure it's happened to all of us.  There's a technique
> that I use to make the issues plain.
>
> When one evaluates a solution, they usually have a set of criteria used to
> perform the evaluation and so each solution gets a certain score depending
> on how well it meets that criteria.
>
> S_i = sum_j C_j,i
>
> But of course, there's usually no agreement on how well each solution meets
> that criteria.  What has worked well in the past is to average everyone's
> criteria assessment.
>
> C_j_i = average_p C_j,i,p
>
> And also there usually is no agreement on what criteria is relevant and so
> we let everyone submit their criteria and then we weight each one by an
> average of how much people think it's relevant.
>
> W_j = average W_j,p
>
> So the solution gets a score of
>
> S_i = sum W_j * C_j,i
>
> It would be interesting to see what we get with regards to logging.  Any
> one care to try this experiment?
>
>
> Regards,
> Alan
>
>

Reply via email to