Really? Can it be adopted for Site Selection? Or would data entry be the bottleneck?
Yes and yes. If I'd thought it would have sped things up, I would have put you in contact with the programmer, since we used it for Hugo ballot counting.
For a single race with <5 candidates where you only need to know the winner (that is, you don't want or need 2nd, 3rd, etc.), you can count 2000 IRV ballots by hand faster than you could key the information into the counting program. The Hugos need the program because there are 15 categories, at least six candidates (usually) for each category, and we want to know each placement. Oh, and the administrator has a couple of weeks between close of balloting and when s/he has to get the plaques engraved, which gives him/her time to do the data entry.
Even in the notorious 14-hour ballot count of 1991, it only actually took us two hours to count the 2,107 ballots -- the other twelve hours were spent validating them.
I was NASFiC site selection administrator in 1992 when we had the four-way race for the 1995 NASFiC and had to go to the third round to determine the winner. Even then, I think it might have taken us maybe 90 minutes or so (I forget the exact time) to do the counting, and I was able to announce the results during the Masquerade half-time.
![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png)
timeasmymeasure
Counting IRV Ballots
Date: 2007-01-24 02:53 am (UTC)For a single race with <5 candidates where you only need to know the winner (that is, you don't want or need 2nd, 3rd, etc.), you can count 2000 IRV ballots by hand faster than you could key the information into the counting program. The Hugos need the program because there are 15 categories, at least six candidates (usually) for each category, and we want to know each placement. Oh, and the administrator has a couple of weeks between close of balloting and when s/he has to get the plaques engraved, which gives him/her time to do the data entry.
Even in the notorious 14-hour ballot count of 1991, it only actually took us two hours to count the 2,107 ballots -- the other twelve hours were spent validating them.
I was NASFiC site selection administrator in 1992 when we had the four-way race for the 1995 NASFiC and had to go to the third round to determine the winner. Even then, I think it might have taken us maybe 90 minutes or so (I forget the exact time) to do the counting, and I was able to announce the results during the Masquerade half-time.