Hacker News Reader

Comment by jdrc | Hacker News Reader

What kind of research is needed in cryptochains?

None. They are noticing the NFT hype starting to die down, and now need to keep the ferris wheel running with something. Crypto is only marketing, there is no actual value created.

NFT hype is still going strong.

https://dune.com/rchen8/opensea

Is it? https://news.ycombinator.com/item?id=30745139

Yes, it is!

https://news.bitcoin.com/moonbirds-nft-sales-skyrocket-captu...

It's just not the same NFTs from 1-2 years ago that are doing high sales volume. Interest rotates around. Just look at the sales volumes overall.

My link explains that the wash sales prevent us from deducing anything from these numbers.

Here's a thorough analysis of NFT wash trading. The activity is overstated, it's there, but the majority of volumes are not wash sales:

https://blog.chainalysis.com/reports/2022-crypto-crime-repor...

These analyses can only come up with lower bounds rather than actual estimates.

I'm not sure which part of the article you rely on to assert that "the activity is overstated", but it claims:

> With blockchain analysis, however, we can track NFT wash trading by analyzing sales of NFTs to addresses that were self-financed, meaning they were funded either by the selling address or by the address that initially funded the selling address.

Any wash sale not doing exactly that will be missed. The article itself is titled "Chainalysis Detects Significant Wash Trading and Some NFT Money Laundering" which does not bode well, on top of the arguments made in the HN comment I linked.

I wish you go through Foundation of Blockchain lecture series by Tim Roughgarden and also understand what cryptography is. I bet you will change you mind on - there is no math in crypto.

He did not say anything about cryptography.

So how has that mapped into actually useful projects?

Very wide range. On the cryptography side: new signature schemes, cryptographic proof systems. Research on voting systems, economic incentive design, coordination and trust problems with value transfers (bridges) across different chains, VM + compiler design for smart contract execution

all things crypto has been pretending to work on for more than a decade while riding the work of an individual research paper from the 1970s which explained, at the end of the paper, why this should never be used for money

y'all some junior koch brothers

To get an idea you can search for blockchain on Google Scholar. There is plenty to be done. Some naive questions could be... How to scale them, how to increase privacy, what are the right incentive structures, what happens if you change those structures, how can you reduce the size of the chain while still keeping the chain secure, what kind of attacks are possible and how do you mitigate or prevent them and so on.

Answer to 95% of the above questions: have a trusted central authority and permissioning. Then you can trivially scale, increase privacy, keep it small and secure, and prevent attacks.

That's why they're not researching how to do it when you have a trusted central authority and permissioning.

Come on man, this kind of comment doesn't add value to the discussion at all. You can do better than this.

You'd be surprised, though, how often people forget that.

Working without a trusted central authority and permissioning is extraordinarily expensive and wasteful.

Think about it, the actual useful compute power of Bitcoin (or Ethereum) is about what one average computer can do. But because of massive duplication (can't trust the other one) plus PoW, you need around 100,000,000 times the compute power.

The real question then is when or under what circumstances it is worth it, and that question is not being asked enough.

Research on finding a legitimate use case ;)

Reply on news.ycombinator.com