Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I'm an educator and right now, I'm evaluating the crypto space.

A huge problem here is that it's very fast moving.

Take the fast moving Web of the 90s and multiply it by how much more people use the Web today.

At Eth Amsterdam I talked to many engineers from that space and didn't get a satisfying answer to many questions devs usually ask about new tech.

"Where should you store your data?" - "It depends!"

Back in the days, and even today, in Web2, you would spin up a relational DB and it would be good enough for 99% of use cases.

I know, that changed with the rise of NoSQL, but in Web3 it's even harder.



The reason the "engineers" at that conference couldn't give you straight answers is because blockchain doesn't have good answers.


No cryptocurrency or Web3 platforms actually store their data on the blockchain. That would be ludicrously expensive; it'd cost tens of thousands per gig at least. Relational databases, cloud object buckets, and CDNs are pretty much all you should need, storage-wise, unless you're solving an especially interesting problem.


Isn't the data technically stored on-chain with proof-of-space chains?


Arguably, perhaps, but nobody actually uses those.


So... what's the Web3 part for then?


Branding.

I'm not even joking; everything they do with blockchain can be done cheaper, easier, and just as well without it. They do the minimum they need to to hype themselves up as being "Web3" and that's it.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: