Hacker News Reader

Comment by jeffreyrogers | Hacker News Reader

I'm very skeptical of the value of blockchains. It seems that people proposing them don't have any idea about when/why you'd use them over a more centralized system, and in many cases (e.g. financial applications) there are existing regulations that make it difficult for a decentralized system to work (e.g. know your customer laws and centralized counterparty systems).

View on HN

> It seems that people proposing them don't have any idea about when/why you'd use them over a more centralized system

I think that this diagram is a good starting point...

https://media.licdn.com/mpr/mpr/shrinknp_800_800/AAEAAQAAAAA...

According to this diagram, IPFS should use (or be) a blockchain.

I think very many IPFS use cases do need / benefit from blockchains.

With the same logic also a BitTorrent tracker should use a blockchain, but obviously it's not the case. You are distributing trust-based units (money, legal and financial instruments, etc), not files (video, images, etc).

This diagram is to answer the private blockchain hype. It's not meant to address any piece of technology- just qualify when a blockchain doesn't provide clear value.

Perfect starting point. Thank you for posting this.

Thanks for sharing this. A picture paints...cuts through all the hype.

Note that "private blockchain" means "gives you a Merkle tree", which you literally get just putting all transactions into git. I suspect the "private blockchain" products that actually exist (and aren't just vapour) are git underneath.

This exact argument seemed compelling when the internet was rolling out. Private networks were fine and even more useful—who would send financial information over a public network? It turns out that technologies that unlock permissionless innovation are incredibly important. Regulators want blockchains to succeed because they aren't faceless boogeymen who stand in the way of progress for fun.

Blockchains allow me to build economic protocols that require scarcity without inserting myself as a middleman to enforce that scarcity. This is where the most potential lies. There are lots of problems that exist because the transaction costs to solve them are too high: finding someone to solve them, agreeing on a price, and setting the terms for when something goes wrong can be cost-prohibitive. Each time a new economic protocol is built, mankind is permanently given the power to trade frictionlessly in that arena. This is a new thing. Join us in figuring out how to use this new thing.

I think you're right, it doesn't make sense to use blockchains in traditional FinTech - the real value comes from the concept of distributed trust model. The base case (bitcoin, altcoins) though seems sound, but whether or not the value of distributed trust (which is nonzero) exceeds the 'natural' hassle of the respective APIs/limits on the network, or the 'artificial' hassle as will be imposed by increasing regulatory burden moving forward, is an open question.

Also a lot of people don't seem to realize the obvious problem with spending huge amounts of electricity on "busy work" to generate hashes (i.e. "mining" is a very poor term for what's going on).

It's a class of fiscal blindness in the same category as the people who are positive gold is a great investment that will always retain it's value (it won't, the existence of a spot gold price proves this trivially).

> there are existing regulations that make it difficult for a decentralized system to work (e.g. know your customer laws and centralized counterparty systems).

Those regulations don't apply to peer-to-peer electronic cash transactions, just like they don't apply to physical cash transactions.

They apply when you try to convert from a fiat currency to a digital currency and vice-versa. And trading derivatives contracts on a blockchain would almost certainly violate US laws.

edit: there are also regulations that apply to physical cash transactions if they are over a certain amount.

Yes conversion is subject to regulation but I don't see how that prevents cryptocurrency/blockchain applications from being useful.

>And trading derivatives contracts on a blockchain would almost certainly violate US laws.

These laws will eventually be made obsolete, the same way obscenity laws were made obsolete by the internet.

In the EU the e-money regulation applies to any kind of electronic cash transaction.

If you become an Electronic Money Institution, in most cases you'll still have to follow KYC and AML laws. Bitcoin companies in the EU all do after certain thresholds.

Yes, but that's not peer-to-peer. That's like a company ferrying physical cash from one party to another. They become a financial intermediary and become subject to financial regulations. You don't have to take the role of a financial intermediary if you're an enterprise that uses cryptocurrency. Microsoft accepts BTC, but is not a money transmitter.

Reply on news.ycombinator.com