On Wed, Jun 3, 2020 at 3:40 PM Kerim Aydin via agora-discussion < agora-discussion@agoranomic.org> wrote:
> > On 6/3/2020 2:45 PM, Aris Merchant via agora-discussion wrote: > > On Wed, Jun 3, 2020 at 2:20 PM Kerim Aydin via agora-discussion < > > agora-discussion@agoranomic.org> wrote: > > > >> > >> On 6/3/2020 2:12 PM, Kerim Aydin wrote: > >>> > >>> Aris may have a little more difficulty with eir proposal because that > F/A > >>> ratio doesn't have a clever term-of-art name defined in R955 - and I > just > >>> realized it's broken anyway, because the "greatest F/A ratio" doesn't > >>> error trap division by 0 the way R955 does, even if it directly refers > to > >>> R955. > >>> > >>> (In terms of "greatest", for A=0 you might need an explicit tiebreaker > - > >>> level of F maybe? - that's not in the rules anywhere). > >>> > >> > >> Just a little more on this: the way we used to do it is to define > "Voting > >> Index" (VI) using the F/A ratio such that: > >> > >> If A>0 then VI = F/A; else > >> if F=0 then VI=0; else > >> VI = 'Unanimous' > >> > >> and we had additional text that said a proposal was ADOPTED if VI > AI, > >> where 'unanimous' was considered greater than any AI. > >> > >> Still doesn't help with ties for A=0... > >> > > > > I'll go with (F - A)/total. > > You might want to limit to "adopted" proposals (maybe?), in the above > formula you could have a proposal rejected at AI 3 beat a proposal that's > adopted at AI 1. Maybe that's ok though. Other wordsmithing: I don't > think "open" is a defined concept for decisions in "no decision whether to > adopt any proposal distributed in the same message remains open" (it could > easily apply to either the full decision or just the voting period). > > I hereby solicit opinions on 1) Whether it should be limited to adopted proposals, and 2) Whether it should be calculated in terms of strength for and against or in terms of ballots for and against. -Aris
