Is Crypto Bullshit?

The big issue with this article is that once you take a deeper look at any of the examples, they are complete BS.

A good example is Filecoin:

> So, for example, you have things like Filecoin, which describes itself as “an open-source cloud storage marketplace, protocol, and incentive layer.”

If you actually look at the finances for Filecoin, it costs the network about 10x times more to store data than the centralized alternative Backblaze. Who is going to pay 10x more for storage?

The reason for this cost discrepancy is obvious once you look at the technical details of Filecoin. Because of the decentralized overhead, the mining requirements are ridiculous. Just look at https://docs.filecoin.io/mine/hardware-requirements/. 8+ core CPU. A powerful GPU. For a data storage node!

Details for the cost computation for Filecoin:

Computing the cost of the Filecoin network is quite simple. https://filfox.info/en reports that miners are being paid 326,077 FIL a day to store 13.673 EiB. This is 0.0229 FIL/TiB per day. Which corresponds to 0.687 FIL/TiB per month. Which is $34.84 / TiB per month. For a single copy. So you probably want two at $70 / TiB per month.

Backblaze is $5 / TiB per month for a redundantly stored copy.

> Replicating everything AWS or Azure or Google Cloud can do on a blockchain isn’t redundant because the decentralized collective “ownership” and governance of blockchain networks supports security, privacy, stability, and scalability features that Web 2.0 applications running on centralized corporate clouds can’t replicate because they are centralized and control is concentrated.

I find this paragraph enlightening. The author seems to be missing understanding of both systems. Peer to peer and cloud both rely on a large number of machines to remain operating in order to work. I do not see the distributed version being cheaper or more performant in the majority of cases.

To logically jump from ownership and governance to the positive features listed does not follow. If it was better to run these cloud operations in a distributed manor, don't you think the major clouds would have learned and implemented this by now. Also consider that the vast majority of the value of the cloud is captured by their customers, not the cloud host.

To summarize what I understood from the (too long IMO) article...

There might be something useful in the blockchain, specifically in how it seems to make a certain class of transaction easier/more reliable and there might be a good use case for NFTs in that space.

I really hate that the term crypto has come to mean 'crypto-currency' instead of cryptography in general.

"One of my personal epistemic precepts is that if I disagree with somebody who is (a) at least smart as I am, (b) has generally sound judgment and (c) is better informed on a subject, then I ought to concede that I’m probably wrong and revise my opinion in their direction. "

I appreciate this point of view. I think it's also worth noting that your smart, of-sound-judgement friends have their own personal epistemic biases. Some of which can be for or against crypto currencies.

I've personally decided it's not something I value, despite knowing many smart people that invest or are involved in crypto. If I am leaving a bunch of money on the table, at least I know why.

Crypto is total bullshit, how could you ever think that it is practical to find three very large positive integers e, d, and n, such that with modular exponentiation for all integers m (with 0 ≤ m < n): m^d^e === m (mod n) and that knowing e and n, or even m, it can be extremely difficult to find d. Come on.

I think crypto-detractors and crypto-advocates alike would benefit from reading this article. The truth is always somewhere in the middle.

Crypto is neither a tupip mania scam nor a magic solution for everything. One thing's for certain, it isn't going away anytime soon.

dupe:

https://news.ycombinator.com/item?id=29319434

Some, not all.

Yes.

Is money bullshit? It's a shared myth