Hacker News Reader

Comment by zilchers | Hacker News Reader

I think this is the point though - for something like bitcoin, this tension is ok. It's when you start getting into things like blockchains for real estate in developing countries (I think there are a few startups working on this) when it becomes an issue. Someone still fundamentally needs to issue the blockchain deeds, or ok that you, in fact, own the property you are claiming to own. At that point, you have a central authority, and a normal database with some sort of public backup or checkpointing would be as good as a blockchain solution.

So, I'd agree bitcoin doesn't want central governance, and that's a feature not a bug, but for other block chain solutions this article makes a very good point.

View on HN

Agreed, and it is a reason why I am holding back on Etherium so far. I don't know if they have a proposed solution to that, tbh, but it seems pretty impossible to solve.

Although, to be fair, "classic" contracts also need that someone to enforce them. Maybe Etherium & the likes only replace one aspect of it, not the whole process. If company X publicly announces that they'll adhere to a certain contract in the blockchain, they can be held accountable for it and transactions can be completely transparent.

Haha, my issue with Ethereum has always been that, as a software engineer, I know it's impossible to write bug free code. It's a great idea to have a code driven contract utopia, but our first experiment with that model proved we still need human intervention. If it were that simple, we would just have hyper formalized real world contracts (in an almost mathematical notation) - but that obviously never happened. Frankly, hand written contracts are already incredibly decentralized, it's only enforcement that is centralized.

Slightly pedantic, but you and your parent did this, and people do it all the time.

It's Ethereum, not Etherium.

True, I worry about that, too. And in case of the DAO, that issue hit home big time.

There are efforts under way to create contracts with formal verification, but I remain a bit skeptic. It is difficult to write complex software that way afaik.

On the other hand, a lot of contracts are maybe not very complex. I thought about a bank account recently (receive and send money). That could perhaps be done in 100 lines of code? (I don't know the programming language yet, but their basic token creation example seems to be short, and it is kind of like a bank account). That seems impressive compared to the effort banks probably spend on IT.

You'd only be able to use a blockchain for land ownership of the government agreed to unconditionally honor whatever appeared in the blockchain. And even then, the government found change its mind at any moment.

So really, it would not work well unless your whole government was also properly decentralized.

agreed on land (to an extant, I think a digital contract would be a good legal basis if things go to court), but what about virtual land? Or digital tokens that control access to things like a house, car, logins, etc for that matter? Now must there be M:N escrow keys etc.?

It is not even ok for bitcoin if you want to use it as your primary currency, for both conducting transactions and for storing liquid wealth. It paints a nice picture to say 'the market will work it out', but as an individual, that market can wipe many people out. Sure, it might work out as a whole, but the volatility is too much for something that needs to be as stable as currency.

Reply on news.ycombinator.com