5 ms·
I mean, if you're willing to spend that much, and it'll be very expensive, then sure. It's just technophobia - machines are going to be more accurate than a hum
by sadeshmukh 2y ago
I mean, if you're willing to spend that much, and it'll be very expensive, then sure. It's just technophobia - machines are going to be more accurate than a human (who also can make a mistake!).
- Dalewyn 2y ago>machines are going to be more accurate Says who? Also, what does "accurate" here actually mean? Speaking as someone who actually understands computers and machines: I agree with the commons (who are simpletons with regards to computers and machines) that machines cannot be trusted to be "accurate" (whatever that means) or even trusted in general. Especially when a simpler, confirmable-by-anyone method exists: Having someone count paper ballots by hand in the presence of anyone and everyone. That includes mistakes and errors. The value here is anyone and everyone can and will immediately understand (and thus accept) what is going on. Also, why are we even putting the integrity of the very foundation of our democracy on the table in exchange for convenience and cost of all things? Are we serious? It should be a good thing we are taking precious time and money to make sure our democracy is working properly. I thought democracy was actually fucking important.
- BenjiWiebe 2y agoMachines are amazing at counting things without losing their place. I'd trust an ATM's counted stack of bills over a human's (for sure if they only each got one try). I've written some code at a previous job to simplify data entry. The previous method was adding numbers from a stack of papers, with a calculator. I trust my code to add up the numbers on the computer over a human reading them from a printout and entering them in a calculator. Humans make mistakes. A lot.
- the_gipsy 2y agoIf the technical problem was solely about counting then obviously everywhere in the world we would be using machines by now. But we don't. Because the technical problem is trust, not counting.
- ethbr1 2y ago> Humans make mistakes. A lot. To put some numbers on this, from my experience. Health insurance manual claims processors (who usually average ~5 years of experience) can do 95+% accuracy, at speed (a few minutes), at scale. That's counting and verifying multiple things against processing rules. General data entry, from less trained folks, tends to average around 85% accurate (i.e. 15 mistakes + 85 entries correct, out of 100 entries).
- ajb 2y agoWe do it in the UK Volunteers count the votes because they want to see a fair election (and there are ways of checking if someone partisan slipped some votes into the wrong pile). I agree with GP. Transparency is more important than precision in democracy. Good engineering is about choosing the right technology, not just the more recent one. Sometimes the right technology is paper.
- Dig1t 2y ago[flagged]
- tsimionescu 2y agoAlmost every democratic country on Earth today does it like that, and all democratic countries have done it like that for the last 100-200 years. Counting paper ballots is just not that hard. Machines are infinitely more complex and exploitable. Plus, you have the extra layer of public perception: it's much easier to convince a chunk of the public that all the machines in some area are miscounting, than it is to convince them that all human vote counter in those areas are miscounting, and all in the same direction.
- Aerroon 2y agothat all human vote counter in those areas are miscounting, and all in the same direction. And you can send observers that can watch the entire process.
- Dalewyn 2y ago>watch the entire process. "Entire" is the keyword here. Any programmer worth their salt knows that it's practically impossible to vet that what is executing is 1:1 the code that someone at some point in time audited somewhere, or that the code is worthy of trust from the commons in the first place. Anyone and everyone can watch someone count paper ballots, noone can watch a computer count electronic ballots.
- ethbr1 2y ago> Any programmer worth their salt knows that it's practically impossible to vet that what is executing is 1:1 the code that someone at some point in time audited somewhere, or that the code is worthy of trust from the commons in the first place. What? There are entire systems built around doing exactly that. Embedded, military, high-trust. It's never state of the art performance or mass deployed, because most people would rather have performance and cost optimized over assurance, but it exists and is in production use. You verify hardware, chain of custody from production to delivery, track every deployed piece of hardware, then lock the firmware and enforce restrictions on anything that executes after that. It's not easy or cheap (or foolproof, as anything can be exploited), but it's also not impossible. And substantially hardens security. And for simpler systems with lower performance requirements, completely achievable. F.ex. voting machines don't need to be running 16-core, hyperthreaded CPUs running multi-process operating systems
- mattclarkdotnet 2y agoThe disconnect is that in most of the world we only vote for one or two candidates on a ballot. In America you vote for everything from the president to the dog catcher on one ballot. While I think of it, the USA and UK should both stop holding votes on working days. That is nuts! Do what Australia does and vote on a Saturday and make it compulsory.
- worstspotgain 2y agoBelieve me, we've been aware that this is a non-bug feature for a long time. The Tuesday law was passed in 1845. Instead of changing it, many legislators are pushing in the opposite direction: trying to selectively suppress their opponents' votes further. If it hurts them more than us, it's a worthy goal!
- mr_mitm 2y agoAre you sure? The last time I voted in Germany they gave me five ballots (EU, state, county, city, district), some with dozens of candidates - per party. I had dozens of votes to give. Here is a similar example: https://www.volksfreund.de/imgs/scaled/28/1/8/3/7/5/7/5/0/5/w1100_h825_x1796_y1347_Kommunalwahl_Trier_2019_Stimmzettel-1a38573d4a4dfd30.JPG https://www.volksfreund.de/imgs/scaled/28/1/8/3/7/5/7/5/0/5/...
- pedalpete 2y agoIn Australia, (I believe) you have to pick your top 5...for a bunch of different items you're voting for. Here's an example ballot https://www.ecsa.sa.gov.au/images/article/2018_LC_Above.png https://www.ecsa.sa.gov.au/images/article/2018_LC_Above.png
- bigger_cheese 2y agoI'm Australian, that screenshot is from the State election in South Australia, it is an example of how the Upper House ballot paper looks, it is similar in my state New South Wales. The vertical columns (labelled as Group A to E in screenshot) divide up the political parties. The Greens will be one column, Labor Party another, Liberal Party another column and so on. There are two horizontal rows separated by a thick line. You can choose to either vote "above the line" or "below the line" but not both methods. Above the line is used if you would like to vote based upon the wishes of a political party and below the line is used for "finer grained" voting for individual persons. For example the Labor party might have 3 Candidates "Fred", "Mary" and "Bob" if I vote above the line I can put a 1 next to the Labor party and then the Labor party's wishes will determine how my vote is distributed. Or if I Vote below the line I must number 12 different people in the order I want them to be chosen. So I could number Bob from Labor first, Peggy from the Greens second, then Fred from Labor third and so on and I exert exact control over how I want my preferences to be distributed. edit: Our elections are staggered, The State parliament is elected on different day to the Federal Parliament, which is different to Local City Council elections.