Hacker News Reader

>But then you make the mistake of thinking because you're really smart and good at this thing you must be smart about something completely different, like oh I don't know the financial system. That's crypto.

That's definitely crypto, but you also just described half of HN.

That's very "pigeonholey" way of looking at things though, no? Just because you're good at something doesn't mean you can't become good at something else too. If you're super knowledgeable about very technical things, learning other hard sciences is feasible as well given enough time.

At the end of the day, if you create anything that's worthless, it will be worthless independent of whether you're an insider or outsider.

I think GP’s underlying point (besides paraphrasing a popular YouTube video) is that for good reason, society elevated “computer engineers” because the work they did (e.g., build Google) was very impactful and value-creative.

A bunch of computer engineers are now burning through that good will in a misguided attempt to accumulate personal wealth.

Good will built when building Google (and others from that era) is already burnt by Google (and others ) of today. You rarely here positive stories about tech in general media these days compared to say 00's.

The argument isn't that they couldn't become knowledgeable in finance and economics, if they invested enough time and effort, but that they aren't, because they haven't, because they think that their expertise in one area makes them automatically experts in other areas. This is pretty accurate in my experience.

I can have all of those positive things on a blockchain for when I need them, and I can do without them when I don’t need them. Blockchains are about choosing the degree of trust you want to have in others and the amount of control you want them to have over you, not completely throwing out ideas that work.

> imagine you're a smart person who is super knowledgeable about something like software engineering, distributed systems and consensus. That's great. But then you make the mistake of thinking because you're really smart and good at this thing you must be smart about something completely different

This is such a bad take that I'll be honest, it makes me angry.

Someone can't be super knowledgeable at more than one discipline? More than two? That's complete nonsense.

Just because you are smart in one field does not mean you are smart in a similar but unrelated field.

This is not what he said

Actually, I disagree. People have said the same about paper voting systems, that the inefficiency is a feature not a bug… yet they have major downsides

eg to take the highest profile example with the most consequences, George W Bush got elected because people were confused with butterfly ballots and then it took too long to do a recount in Florida so the Supreme Court had to step in and stop the recount … as a result we got 9/11 (negligence) the invasion of Iraq (lots of excess deaths and destruction of large swaths of the middle east) the PATRIOT Act (erosion of civil liberties) and much more. Imagine if we had a secure system alongside the paper ballots that involved Merkle Trees and multi-factor verification for voting, so everyone could verify theirvote and participation would be higher. Banking apps are secure enough for people, why not this…

I wrote an article on CoinDesk about this very thing two years ago: https://www.coindesk.com/tech/2020/03/12/in-defense-of-block...

Anyway, going back to “feature not a bug”. The real problem is Blockchain. It uses the full power of the network on every transaction, so it costs just as much to transfer $500 million in one shot as it does $0.50 — that is the wrong pricing structure. It is too cheap to transfer huge amounts and circumvent capital controls, do money laundering etc. The “world computer” is a glorified mainframe with “gas” playing the role of renting time on the mainframe. “Flash loans” shouldn’t be possible, they only are because one transaction in the world can happen at a time.

When we move past blockchains, we will have proper pricing of securing transactions (in % of the amount being managed). And then we can build reversibility ON TOP OF the non-reversible architecture.

The non-reversible architecture and finality in the vast majority of transactions is a GOOD thing just as it is in most databases. Reversibility should be explicitly built on top of it.

Smart contracts and other things are great and we have only scratched the surface because Blockchain holds them back. Having the rules stated upfront and enforced without government violence is great. For example copyright and DRM is being replaced with NFTs and POAPs. Old boys clubs are being replaced with DAOs. New non-coercive business models are emerging. Open source and journalism and other digital content can be monetized without SWAT teams raiding grandmas. Just one of hundreds of innovations.

People just cling to the old ways because the technology is not mature yet.

Paper ballots are better than any purely electronic form. Like that's not even a debate except by people who are trying to sell electronic voting machines. The gold standard is either print outs from a voting machine or a ballot you fill in with circles with a pencil that a machine can verify. Counting these ballots is nondestructive (unlike punch cards) and highly accurate. Audits of elections involving these can be done easily by hand.

As for the 2000 election, this is a whole rabbit hole. You're right the butterfly ballot probably caused miscast votes. But that's user error. Punch cards are a ridiculous form of voting technology. All the recounts weren't about being more accurate. They were about passing punch cards through counting machines that would knock out chads every run through and changing voting standards after the fact (eg pregnant and dimpled chads).

Blaming 9/11 and the grand misadventure of the Iraq war on paper ballots is... a stretch. I mean, 9/11 had been in planning for years and probably would've happened anyway. I suspect a war in Iraq would've happened anyway with Gore in office. I mean a good chunk of Democrats (including Hilary Clinton) voted for the Iraqi war resolution. Many of these same people (including Biden) are now threatening to start a war with Russia over Ukraine.

> Just one of hundreds of innovations.

I disagree. Contracts enforced by the blockchain only have value when the entire context of that contract is on the blockchain. As soon as you leave that system you need traditional infrastructure and systems to enforce contracts and handle disputes.

Ha ha ha. Not even a debate? Why would something so non-obvious not be up for debate? Are you afraid to lose the debate?

What's next, telephone switchboard operators are better than VOIP? We went from paying $3 a minute for long distance audio calls to nearly free calls with multi-way high res video broadband, simply by automating away the infrastructure to get things done. Why can't the same be done with voting? Today we spend 16 billion dollars on a national election with shitty turnout, tomorrow any small organization can have a poll and vote about anything, in any number of ways, and turnout will be much better too.

There was a time that chess playing programs were laughably bad. And within a couple decades they are able to beat any grandmaster. Give it time.

About the 2000 election: I regret even mentioning the butterfly ballots, because the real point is that it took TOO LONG to recount the votes, BECAUSE it was paper ballots, and thus GWB was elected because the Supreme Court stepped in. Paper ballots and their inefficiency have consequences and downsides, you just might choose to ignore them in order to not have the debate of pros and cons vs electronic form.

(9/11 may not have happened if Bush had acted on intelligence like "Osama Bin Laden Determined to Attack in the United States. https://www.youtube.com/watch?v=3L2513JFJsY) Agreed about the war with Russia over Ukraine, Democrats have their own issues, absolutely. (As do leaders of other countries).

> I disagree. Contracts enforced by the blockchain only have value when the entire context of that contract is on the blockchain. As soon as you leave that system you need traditional infrastructure and systems to enforce contracts and handle disputes

Even if that were strictly true, we can have all these applications on the blockchain today: https://intercoin.org/applications

But it's not even true. People don't need the entire source of truth in a byzantine fault tolerant network. They just need to secure that it wasn't tampered with. The key is that you can turn over control to a computer program + protocol rather than elected representatives, magazine publishers, librarians or telephone switchboard operators. And we have been increasingly doing that.

> Ha ha ha. Not even a debate?

You brought up paper ballots for voting and then blamed them for 9/11 and the Iraq war. If that's your definition of "debate" I don't think I can help you.

> What's next, telephone switchboard operators are better than VOIP?

If you're so into debate you might want to look up "straw man argument".

The key design principle for a voting system is that voters should be confident in the results. That means if there's a paper ballot, the voter should be able to understand what it says (ie not just some QR or PDF417 or similar code). The worst that can really happen with paper ballots is ballot-stuffing but there are lots of checks and balances in place such that there's no evidence of this having ever been a widescale problem in the US, let alone has changed the result of an election. A pure electronic count has no such safeguards and no real capability for an audit trail.

> There was a time that chess playing programs were laughably bad. And within a couple decades they are able to beat any grandmaster. Give it time.

Here you come across as the very kind of person I mentioned, a Blockchain Andy who has completely drunk the Kool-Aid. Chess is a compute power problem. Electronic voting is not a computing power of algorithm or even a technical problem.

> ... because the real point is that it took TOO LONG to recount the votes,

No, it wasn't. The real problem in 2000 was that multiple recounts were done selectively to a changing standard of what constituted a valid vote even contradicting the instructions given to voters to "divine the intent of the voter".

> and thus GWB was elected because the Supreme Court stepped in

GWB won because he got more votes with the rules that existed for that election. Period. I don't say that as a partisan (for the record, I'm a leftist closest to the Bernie camp). It's just fact. Even a comprehensive review of ballots by the NYTimes after the election showed GWB won even with the most favourable change of rules (eg dimpled chads).

> GWB won because he got more votes...

I am aware that there are various opinions on it, but once again, it's not so "clear cut" as to not have a debate. In fact, the sources I'm following, such as Wikipedia, say exactly the opposite about the media recounts:

https://en.wikipedia.org/wiki/Bush_v._Gore#Recount_by_media_...

In 2001, the National Opinion Research Center (NORC) at the University of Chicago, sponsored by a consortium of major United States news organizations, conducted the Florida Ballot Project, a comprehensive review of 175,010 ballots that vote-counting machines had rejected from the entire state, not just the disputed counties that were recounted.[3] The project's goal was to determine the reliability and accuracy of the systems used in the voting process, including how different systems correlated with voter mistakes. The study was conducted over a period of 10 months. Based on the review, the media group concluded that if the disputes over the validity of all the ballots in question had been consistently resolved and any uniform standard applied, the electoral result would have been reversed and Gore would have won by 60 to 171 votes.[4] On the other hand, under scenarios involving review of limited sets of ballots uncounted by machines, Bush would have kept his lead. In one such scenario — Al Gore's request for recounts in four predominantly Democratic counties — Bush would have won by 225 votes.[a] In another scenario (if the remaining 64 Florida counties had carried out the hand recount of disputed ballots ordered by the Florida Supreme Court on December 8, applying the various standards that county election officials said they would have used), Bush would have emerged the victor by 493 votes.[b][81]

The emphasis above was added by me... far more media organizations, and a comprehensive analysis, and so forth. How much more extensive can you get, and the result is that Al Gore would have won under any uniform standard at all. And again ... this was all because of the outdated technology. Call it what you want, paper ballots, rejecting by voting machines that counted them, etc. The fact is, the election would have been a lot MORE reliable if it was done with Merkle Trees and private keys, as I said.

And no, I'm not a Blockchain Andy who's completely drunk the Kool-Aid. I often critique Blockchain right here on HN, I think Blockchain holds Web3 back. But the actual applications are very viable (if we move past blockchain as the technology on which they are built) and we'll all be voting from our phones in 10 years. Do you think that somehow voting is one application that won't make the switch from manual paper-based counting to technology, because people can't be "confident of the results" from their electronics?

> About the 2000 election: I regret even mentioning the butterfly ballots, because the real point is that it took TOO LONG to recount the votes, BECAUSE it was paper ballots, and thus GWB was elected because the Supreme Court stepped in. Paper ballots and their inefficiency have consequences and downsides, you just might choose to ignore them in order to not have the debate of pros and cons vs electronic form.

Germany counts 50 million ballots in a single night (from 18:00 CEST to ~06:00 CEST). Even if the US had the same number of people counting ballots as Germany does, the US should be able to do a full recount in 36 hours.

Paper ballots aren’t the issue. The US just fucks it up in multiple different ways, and would find a way to fuck up electronic voting just the same (if not even more).

> Imagine if we had a secure system alongside the paper ballots that involved Merkle Trees and multi-factor verification for voting, so everyone could verify theirvote and participation would be higher

And rely on the same people who accidentally voted for Pat Buchanan on the butterfly ballot to understand merkle trees and verify their vote?

We already know how to increase participation if we wanted to - mandatory voting like they do in Australia.

Paper ballots work fine as long as they aren't silly designs like the butterfly ballot or the hole punch. Filling in circles with a pen and then running it through the machine yourself to verify the ballot is machine readable works very well.

Um, no. Do you expect people who use banking apps to understand bank-level security, elliptic curve cryptography, key generation and so forth?

You just give them a freaking app, to scan their vote on a website, confirm it's voting for the correct thing, and sign with their private key. Then they can check that their vote was included in the final tally (don't mention Merkle trees). OK?

The real question is, how do you make sure each person has only one vote (see this for example https://www.geekwire.com/2015/the-next-fraud-wave-when-banks...) without revealing the identity of every voter to the government. And also how do you prevent the final tally from being revealed to anyone until it's all ready? That last one is the hardest one (and we fail at doing it in our elections, the media already predicts the winner before the polls close).

> For example copyright and DRM is being replaced with NFTs and POAPs.

…no it's not, where's the contract that says someone gets "the rights" when you transfer an NFT? What if you split the rights up and sell Sony Pictures the movie rights to your ape then sell someone else the NFT?

Even actual NFT ownership isn't respected when it's a hack, it turns out. OpenSea blacklists it and the rest of the ape picture community doesn't start treating the hacker like they own it, even though the data says they can now set a hexagonal Twitter avatar.

The crazy part about crypto is somehow it looked for all the weird economic inefficiencies that don’t really make sense (Gold is valuable beyond its utility, buying art is buying the object not the image, etc) and turned those into the feature of the project.

Seems like it's half misconceptions (like Austrian hard money economics) and half regulatory arbitrage. There's no reason crypto things shouldn't be as regulated as traditional banking and securities, they just aren't because they don't stop to ask the SEC what to do first.

I wonder why BTC isn't taxed at the collectible rate (higher than ordinary income) like actual gold investment though.

Major newspapers performed the recount after the Supreme Court stopped it and found Bush won the recount. In fact they tried a few scenarios the Gore team wanted, and in each scenario Bush won.

While the butterfly ballot may have pushed votes to Bush, studies have shown he lost votes when Florida was called premature. Calling Florida before the polls closed suppressed the western panhandle vote which was highly Republican.

https://www.nytimes.com/2001/11/12/us/examining-vote-overvie...

Well Wikipedia says exactly the opposite, and it is based on far more extensive recounts and analysis by far more newspapers, years later:

https://en.wikipedia.org/wiki/Bush_v._Gore

New York Times is just one newspaper and often carries water for the Establishment.