[RFC] tweak to voting rules

[Kent Gibson]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1



Matthew D. Fuller wrote:
> On Thu, Nov 30, 2006 at 08:10:21PM +0800 I heard the voice of Kent
> Gibson, and lo! it spake thus:
>> I don't see any need for the votes to be integers.  Why not just
>> use decimal votes such as one of 0.1, 0.5 or 0.9 (and a negative
>> equivalent)?
>
> The problem with this is the same problem as going to +/-2; you're
> trying to express two dimensions of response (suitability of
> concept / suitability of implementation[0]) in one dimension of
> representation. At best, it'll be a hacky and poor fit.
>
I thought all we are trying to express how close a patch is to being
accepted.
That can be measured as a distance independent of the dimensions involved.
> The decimals are somewhat handy in person-to-person communication
> to give some idea of degrees of like/dislike between +-0/+-1, but I
> think they're even less suitable for any sort of auto-grokking...
>
If the values in use are formalised they are just a symbol.
Auto-grokking just needs the symbol table.
I don't see how making those symbols integers is advantageous.

>
>
> [0] And maybe even a third dimension, since "suitability of
> implementation" encompasses both "this general way of achieving the
>  end" and "details of this implementation of that general way".
>
Oh god - we are on our way to the M theory of patch approval.


Cheers,
Kent.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.3 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFFbufCgoxTFTi1P8QRApSqAJ9W6vxPeDMsKm8aAOSXjqDlALLzxwCg6b2I
p8T8e6NV11doMYUn50HHXpQ=
=Lbya
-----END PGP SIGNATURE-----