Ask HN: Should I Learn Blockchain?

https://news.ycombinator.com/item?id=37339964

by itsmekali321 • 3 years ago

6 13 3 years ago

news.ycombinator.com

I am half way learning it, through the MIT lectures. But, most of the reddit is against it. I asked a question on reddit about it and many were against it. I sometimes feel that might me true and what i am learning might me totally useless in the future. Should i really continue or leave it?

PaulHoule• 3 years ago

If it really speaks to you, if you find it interesting in and of itself, go ahead.

I don't believe it is the next big thing that will change the world though.

I've revisited the "Web3" concept lately because I've gotten involved in the Fediverse and it seems to me that it is missing a universal system for authentication. To add new services to the Fedi it seems fundamental there should be single sign-on so I can log in and prove I am a certain person on mastodon.social.

The traditional way to do this put the server in charge and I guess I would trust mastodon.social as much as I would trust Facebook or Google but there will always be people running small servers who are dishonest who might reset my password and steal my account.

So the "Web3" concept where I have cryptographic keys and some immutable database that verifies those keys is appealing to me, but (1) it is so tied up with scams and I think the whole idea of "ownership" bothers people, it seems something like a timeshare scam or Amway or

https://en.wikipedia.org/wiki/Enclosure

and (2) boy people hate managing their own cryptographic keys, I mean Che Guevara was one of the most disciplined revolutionaries in the world but he reused his one time pads so they tracked him down and put a bullet in him. I look at how people have really struggled with gpg, whenever I have published some open source project to maven or something I've always made up a new set of keys because I lost the keys I made three years ago when I published another project.

Something like "Web3" that is low resource consumption, cheap to run, still secure and doesn't have all the bullshit, grifters, ugly apes, and all that might appeal to me but people will still hate managing keys

version_five• 3 years ago

Luckily there's a decision framework that can help: https://cdn-images-1.medium.com/v2/resize:fit:800/0*850OATxY...

WallyFunk• 3 years ago

A blockchain is just a database. Why not learn various non-blockchain related database systems instead?

corinroyal• 3 years ago

It's not a very useful data structure, despite all the hype and investment. You'd be better off learning general programming with an emphasis on data structures that are widely used. The book "Designing Data-Intensive Applications" by Martin Kleppmann is a mind-blowing modern survey of data structures and their applications.

thesuperbigfrog• 3 years ago

James Mickens gave an entertaining presentation about why blockchain is poorly suited for most application use cases:

https://www.youtube.com/watch?v=15RTC22Z2xI

mcdonje2• 3 years ago

This is a good lecture. It gets at the heart of the matter.

The best use-case I've heard is adversarial companies who need to cooperate on a project and share data. However, in real life, they don't need blockchains because if a party can't be trusted to share data then they don't meet the minimum requirements to participate in the project.

Real life has incentives that mitigate the need for blockchains to the degree that the overhead isn't worth it.

It's a solution in search of a problem.

ipaddr2• 3 years ago

The big application is b to c transactions. A store owner can be trusted because of reputation level but a new customer less so. This solves the chargeback problem stores face.

retrocryptid• 3 years ago

Blockchain technologies are interesting in their own right as the application of math to a perceived problem. But they're even more interesting in how a technology can be hyped beyond belief. And the sociological aspects of how easy it was to convince people to part with their money for a dubious "investment."

I think to understand the human / sociological aspects, you should grok the underlying tech, even if you don't plan to use it. Maybe not at a detailed implementation level. Like understanding how people use ETH to spin up their own coins is a good level of abstraction to understand. (i.e.-why are ETH and BTC different.)

And I would love for all the eCos / Mark Miller / Nick Szabo e-contracts stuff to be useful someday.

colesantiago• 3 years ago

[ Do I need blockchain? ]

No.

dtbuchholz• 3 years ago

Totally depends on your desires! I think what’s cool about the tech is that it touches so many different areas. Ignore the grifters…those happen in any sector (albeit, web3 has definitely had its issues).

Look into the core infrastructure layer. It incorporates generally useful, interesting topics that are hard problems to solve, including: computer science, cryptography, game theory, distributed systems, p2p networks, finance (tokenomics), social engineering, etc. There’s a lot you can learn from that alone, and you can always apply it elsewhere. Plus, most of the industry is open source, which helps with the learning aspect.

The more interesting protocols and applications imo are in the “DePIN” space (physical infra networks), which take advantage of web3’s permissionless nature (open identity framework with cryptographic, payments, consensus-based rules, run your own nodes, etc.). These can be built on top of chains like Ethereum or totally standalone blockchains. Plenty of companies are building things like decentralized satellite imaging, weather data, compute, storage, IoT, renewables, etc. Admittedly, it’s still very early; the user/dev experience is subpar, but abstractions will continue to improve things (e.g., no common user should need to worry about private keys).

The difference here with traditional web2 infra is that web3’s incentivized p2p network allows applications to take more of a crowdsourced approach, in a sense. Users can contribute but also earn for their contributions through protocol fees or reward mechanisms, which helps scale a protocol itself while—in the long term—changing how data moves on the internet.

Also—slight tangent, but calling blockchains a database is a misnomer. For example, Ethereum is a global state machine that provides shared state & execution, allowing anyone to run arbitrary programs on it. There is definitely (expensive) data storage as part of this, so you can only store a limited amount of data using smart contracts. Also, databases process data a bit differently and more efficiently. The native query features for blockchains (RPC calls) are nothing like that of a database (e.g., SQL), to the point where new networks on top of blockchains are built solely for indexing and querying that data (Tableland, The Graph, etc.).

And of course—there are other benefits like censorship resistance, but the tech itself is more intriguing!

speedgoose• 3 years ago

It’s interesting to understand how it works. It’s not completely useless knowledge and it’s a good exercise. A bit like learning advanced maths at school. You will probably never need to do the same maths after but it’s good training and brain exercises.

Now, as others said you will probably never want to use a blockchain so you may not have to go too deep into the topic.

Ask HN: Should I Learn Blockchain?

Related Stories