Timeline for Do Australian and Irish elections use ballot scanners on ranked-choice ballots?
Current License: CC BY-SA 4.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 7, 2021 at 3:38 | vote | accept | QuantumWalnut | ||
| May 6, 2021 at 9:20 | comment | added | Eric Nolan | In Ireland there isn't that much enthusiasm for this sort of system. The current system works pretty well and many people don't see any real advantage in having the results on the same day versus the next day for most constituencies. There is a certain amount of reassurance in having a room with dozens of people counting and hundreds observing them. People are, justifiably in my opinion, concerned that errors or interference with computerised systems is harder to detect. | |
| May 5, 2021 at 19:29 | comment | added | code11 | Fascinating! I think you should include this in the answer since it explains clearly why even qualitatively acceptable recognition systems are unacceptable for this application. | |
| May 5, 2021 at 18:57 | comment | added | Joe C | 0.25% error can still equate to nearly 1700 votes in the average Australian electorate, when you account for the number of candidates. Six electorates were won on smaller margins in 2019. | |
| May 5, 2021 at 18:12 | comment | added | code11 | Is that really the case? Certainly true before 2010s. Modern CNNs can get below 0.25 error rate on handwriting, granted on curated datasets. Perhaps that is still not good enough? | |
| May 5, 2021 at 13:37 | vote | accept | QuantumWalnut | ||
| May 7, 2021 at 3:34 | |||||
| May 5, 2021 at 8:54 | history | answered | Joe C | CC BY-SA 4.0 |