[EM] 3 or more choices - Condorcet

[Kristofer Munsterhjelm]

On 11/16/2012 04:52 PM, Kevin Venzke wrote:

> Yes that's an excellent marketing approach. I think advocates of *all*
> methods should try to boil down the rationale to a single sentence.
>
> I don't think it is a decisive argument though. Many things in the
> world sound good in overview but end up having problems that weren't
> obvious from the definition.

The converse is also often true. The first example that comes to mind is 
DES, the encryption algorithm: it used strange constants picked for 
reasons IBM kept secret after discussions with the NSA, and for a long 
time, people wondered if the NSA had built some sort of backdoor into 
the thing through picking just the right constants. In the end, it 
turned out that these constants made DES significantly more robust to 
differential cryptanalysis, an attack not known to the public at the 
time of DES being made standard.

Thus methods or algorithms with low complexity may have hidden 
weaknesses, complex methods may use complexity to protect against 
weaknesses, and in both cases, there 's a contrast in how the method 
"appears" (simple and robust, or complex and brittle) and how it really 
acts.

It's going to be a lot harder to have people accept a complex voting 
method than a complex cryptosystem, however.