Dear all,
In our Mumbai meeting we discussed project prioritization. I noted that all of the proposals on the table were problematic, and that there might be a better way to do it.
Essentially all voting and ranking schemes are problematic.
First, results are sometimes surprising, unintuitive - e.g., adding an irrelevant alternative can change the result.
Second, they are susceptible to gaming, or strategic voting: voting honestly by your preferences is not optimal for achieving the results you want.
Some systems are better than others but none are perfect, and the better ones tend to be more complicated and harder to understand.
But for project prioritization, where the intent is not to get final results by ranking but just to use it as a tool to help discussion, at least some of those problems could be avoided by doing it differently.
The idea is simple: have something like a spreadsheet, where people can enter their rankings and that computes the averages or whatever automatically. (Google Docs should work.)
I don’t think it matters much exactly how the averages would be calculated, but in general the simpler the better.
Key points:
(1) Votes are not secret, everybody can see what others have done.
(2) Ranking is not a one-off, but people can change their choices after seeing what others have done, as often as they like.
It is critical that there is no hard, predetermined cut-off point: that would bring all the problems back.
Consequently there's no final outcome either, unless and until everyone agrees.
So this would not be a method for making final decisions, only a tool to facilitate discussion.
I’m not sure how well this would work, but it should be easy enough to try and see.
--
Tapani Tarvainen
