Hacker News Reader

Comment by siosonel | Hacker News Reader

"... Citizen Scores are a near-perfect expression of reputation economics: like most other forms of currency, they are issued by a central bank that uses them to try and influence social outcomes."

It seems to me, and I could have misread, that the biggest problem with the examples of currency systems in the article is that the 'pooled' currencies are unquestionably accepted by participants. Take that guarantee away, and someone could design a reputation currency where participants could reject payments from disreputable participants. In which case accumulation of units does not imply the long-term ability to use them; instead, there will be a long-term incentive for participants to maintain a good reputation.

See this overview of a counter-example to the article's point, a reputation currency system without a central issuer: https://tatag.cc/ui/home-about. (Disclaimer: I'm the developer of the linked site.)

View on HN

In which case accumulation of units does not imply the long-term ability to use them; instead, there will be a long-term incentive for participants to maintain a good reputation.

There's an important distinction here: the score of your peers versus the score of your government/authority.

It's possible your peers love you while your government hates you. Maybe you do great work, but are a pain to the power structure. In that case, you end up with high peer ratings, low "political office" ratings.

If the goal is "orderly citizens," then the political ratings become more viral (as in, the people you associate with get boosted or discounted based on your Official Political Office Rating) and try to reinforce people conform to how The Political Office wishes people would behave.

We can even make this local to HN. HN has secret "political rankings" on accounts to restrict the reach of people the "HN establishment" doesn't like (HI DANG), but the public vote counts are more peer-oriented rankings (which still aren't perfect because HN has 3 to 5 distinct sub-cultures fighting against each other, so a +100 from the "VCs are evil bastards" subculture in the morning gets canceled out in the afternoon once SF wakes up and you get -200 from the "VCs are genius darlings and we should all kiss their feet" subculture).

I think the best way to address the issue of how reputations are determined is to let "sellers" or payment recipients decide on who they want to benefit from their goods and services.

So, in the digital currency system that I have prototyped [1], each team decides on which recommender system they want to provide advise, in real-time, on whether to accept or reject a payment offer from another team. Borrowing from your examples, one team could use an "advisor" developed by the HN staff, another team could use an "advisor" endorsed by the subculture you identify with, etc.

There are many issues that I have worked out in the prototype, such as making sure payments are always traceable to the issuer and inflation is decentrally regulated, and most of the solution comes from the budgets-as-currency approach. I'm still in the process of improving the advisor options with better data-science techniques (hopefully with contributions from others). [2]

[1] https://tatag.cc/ui/home-about [2] https://github.com/siosonel/tatag-api

No, that's not quite how Sesame Credits work in China.[1][2] The system computes a social score for each citizen. But it's not run by the Government or a bank. It's run by Tencent and Alibaba. It tracks what you say online, what you buy, and who your friends are. Having a low-scoring friend brings your score down. New feature: Baihe, China's largest dating service, now uses Sesame Credit scores in matching. It's voluntary now, but will become mandatory in 2020.

This is the successor to the Dang'an, a permanent record kept for each citizen of China since it went Communist. The record was typically maintained in a book by the employer, in conjunction with the Public Security Organization. The Dang'an system became less relevant as more non-state employers appeared. Now it's getting an upgrade.

[1] https://www.youtube.com/watch?v=lHcTKWiZ8sI [2] http://www.bbc.com/news/world-asia-china-34592186

Thanks for the clarification. I think by "central bank", the article just meant a non-decentralized currency system.

It's good that you pointed out the possibility for scores to go down in that system, and I assume that would affect the ability of low scoring participants to transact in the system. If that's the case, it makes me doubt the article's strong assertion regarding that example of a reputation currency.

> Take that guarantee away, and someone could design a reputation currency where participants could reject payments from disreputable participants. In which case accumulation of units does not imply the long-term ability to use them; instead, there will be a long-term incentive for participants to maintain a good reputation.

+1, considering that the empowerment of individuals to issue the currency themselves creates a trust graph that should be enough to mitigate many of the downsides he brings up. Here's my attempt at social, reputation cryptocurrency (but it's currently being rewritten from scratch): https://github.com/sunny-g/whuffie/blob/master/ABOUT.md

You might be interested in this idea of routing payments through a trust network, http://p2pfoundation.net/Ripple.

I'm just a curious onlooker and not directly involved with the idea, but there is a blockchain related implementation at https://ripple.com/, and other implementations of the original idea at https://classic.ripplepay.com/ and https://villages.cc/.

Interestingly enough, I started this project after reading about Ripple and Whuffie, hence the name.

This project however is being built on Stellar https://www.stellar.org, which is actually a fork of Ripple started by the Ripple founder!

I had to dig a bit which founder you were referring to. No controversy here, but just want to give a background: I always think of Ryan Fugger as the person who initially proposed the network of IOUs idea and used the name Ripple for it, and later on worked with Jed McCaleb's team who implemented Ripple as it's known today. [1]

Just amazing how alt-currency/alt-payment ideas have grown so much from being on the fringe to more or less widely accepted. It was hard to imagine back in 2006 how these ideas would finally take off.

[1] https://groups.google.com/d/msg/rippleusers/IVin3Qwrp7k/urza...

Reply on news.ycombinator.com