Ask HN: Should I learn the tech behind crypto even if I don't want to own any?
I have to be extremely picky about stuff I want to learn. I lack time, and moreover, if I actually want to reach a non-trivial level of skills in a particular field, I have to spend serious amount of time, abandoning other wishes/ideas.
So, my question is- is the Crypto tech, like blockchain, worth learning? Will there be applications beyond decentralised finance, web3, etc.?
Will learning the tech make me a better "technologist" in any way?
I want to be clear: I do not want to buy or own cryptocoins. Web3 does not look interesting to me at this point.
Should I still learn about blockchains?
Yes. Read the original bitcoin paper; its a really interesting read and won't take long.
I can't say what % of cryptocurrency tech you'll learn, but it's enough to make sense of a lot of the information out there.
I found it useful -- for a rough idea what's _not_ a possible application for these technologies as much as what is.
Satoshi’s paper https://bitcoin.org/bitcoin.pdf
Very Easy to read
and read the Ethereum yellow paper/beige paper.
Nothing there
Just clicked, there is a pdf of the whitepaper that opens just fine for me, even on mobile safari.
Weird - I get a blank page with just
"It takes advantage of the nature of information being easy to spread but hard to stifle. - Satoshi Nakamoto"
on it. Got the PDF from elsewhere. Chrome on Ubuntu 20.04.
I'm going to be that guy that shills his own stuff. I wrote Programming Bitcoin published by O'Reilly. It has specifics and exercises on all the building blocks that make up the Bitcoin blockchain. You'll learn Finite Fields, Elliptic Curves, ECDSA, Transactions and more. It'll be worth your time since it's a gentle introduction to a lot of cryptographic concepts.
I wouldn't waste time on Web3 or DeFi, though.
Check out Tim Roughgarden’s course on Foundations of Blockchains, Jimmy. Current iteration of web3 and DeFi is a lot of hot air and BS but I think the general idea of trustless computation is bigger than just Bitcoin and that new applications will arise given time.
That is what I tend to think.
I do not know enough to have an educated opinion.
But, intuitively it feels like the technology can be used in a lot of novel fields. And might find new applications.
I found it very interesting, and it’s a very interesting application of cryptography (and, though far less appreciated) economics and game theory which really motivated the learning for me.
Start with Bitcoin, it has a good white paper, is the simplest, and is proven at scale. You’ll want to be careful chasing hype, even if you aren’t buying any. It can lead you down a dozen different highly complex proposals for the next big thing which haven’t been proven and often just monetize buzzwords.
Applications beyond is a bit of a misnomer. If the tech is built from blockchain as its backbone or exploits its benefits there is inherently an economic factor serving as a deterrent for tampering, so even if your application is not specifically money, it may take advantage of a system (blockchain) whose fundamental property relies on being scarce and transferable.
I wish more people who did invest in crypto took your approach, then invested after.
Take a look at this list of academics, all working in or around crypto, producing research that materially solves real problems in computing, not just in blockchain or crypto.
https://iohk.io/en/team/#team=research
Then realise that eschewing it all because you don't like crypto is probably one of the most closed minded and regressive decisions you could make.
IMO yes, since big banks and governments are looking at replacing cash with digital coins. These coins are going to use blockchain tech, most likely. At the beginning adoption will be regulated, banks will push back but I expect innovators to push banks away (we don’t need banks for gov controlled eMoney, every citizen will have an account to the Fed, ECB, etc).
So I believe it’s a pretty valuable skill. The other side of the coin is current reality: it’s still a solution looking for a problem to solve :-)
This doesn't make sense. The main incentive to use Blockchain for payments is that it is decentralized. If that's not a priority, and it's not for governments and big banks, Blockchain doesn't get you much. In fact, the cost of mining is downright harmful to the environment and wouldn't scale to billions of people using it for their daily needs .
> The main incentive to use Blockchain for payments is that it is decentralized.
Some might say that the main incentive is not decentralization, but immutability of records and maybe trustless computing.
Blockchain is not inherently decentralized. Decentralized consensus protocol is what makes Bitcoin decentralized and ineffective, Nothing to do with blockchain. Governments and big banks generally use blockchain with more centralized permissioned consensus for their infrastructures.
There is literally no reason to use a block chain if it’s centralized. You’d just use a database. Maybe a light bit of PKI on top if you want.
Database is heavily designed system, not a kind of data structure as blockchain does, so database is the one that's overkilled in this case. Centralization and decentralization is not black and white, there's a spectrum between them. Permissioned blockchain can have enormous nodes, such as 10000 nodes, involved to achive consensus. How do you make database permanent and achieve consensus between enormous users? You may end up with a data structure called blockchain and consensus protocol in that database with extra layers and barriers.
No, not really. The US financial system does it using SFTP and tech from the 70’s just fine.
All the stuff you’re talking about is only useful when it’s nothing but distrusted counterparties with no trusted intermediaries, no place anywhere someone can store a copy of the data they trust, etc. which pretty much never happens in the real world.
All these issues got worked out a long time ago in ways that really do work fine. It’s a bit slow, but it works.
Most of the things you’re talking about also have big disadvantages for real world players (no one generally wants all transactions stored forever and visible to everyone for instance, or for transactions to be immutable even if there was fraud or fat fingering).
Where it CAN be useful is where the existing power structure is actively trying to destroy trust in any active institutions or actively driving the economy off a cliff bandit style, but is also unable to clamp down on internet connections.
Which does have some niche uses, and it might be more than that at some point. But right now it’s just not well fit for purpose compared to the equivalents.
It depends expensive data centers, engineers and whole legal enforcement involving judges, lawyers, prosecutors, overseer, agents and jails to make public servants not corrupted. Even hiring one human are much more expensive than blockchain. Governments and banks research blockchain to reduce all these expenditure. Blockchain data can be updated with old data permanent at the same time, not a problem at all.
I honestly think you're not informed enough on this topic. What you are criticizing are mostly solved now. Gov and big banks are adopting blockchain opposite to your point.
Lol, ok enjoy the crash. I’ve been in crypto a long time my friend, and written a decent amount of code related to it.
This is a major (but common) misunderstanding of central bank digital currencies (CBDCs). CBDCs are actually more centralized money than has ever been conceived of before, and therefore are especially not a good fit for blockchain.
Working with encryption more broadly will equip you to understand the fundamentals that underpin the web3/crypto world: hashes, asymmetric crypto, signatures, signature chains, etc. And this knowledge also has many uses, of course, outside the web3 realm.
If you're working with Node, tweetnacl.js (https://github.com/dchest/tweetnacl-js) gives you secure defaults and a nice API to start learning and building with. Otherwise, look for a nacl/libsodium library in your language of choice.
Once you're familiar with this stuff, blockchains/cryptocurrency/web3 lose a lot of their mystery. They're essentially all just different takes on using key management and signature chains to verify identities and maintain immutable action logs.
Blockchain is an interesting computer science achievement in secure distributed systems. It's worth learning about as an intellectual curiosity for sure.
Imho it’s awesome to learn. I don’t personally like crypto, but the idea of a blockchain is somewhat genius.
Satoshi’s paper https://bitcoin.org/bitcoin.pdf is _very_ easy to read.
Agree, it is important to separate the cultural/socio-economic aspects of Crypto from the engineering/technology behind it.
Sadly, both issues get boxed up into a giant package, bow tied with hate, greed and emotions. Even on HN, let alone the rest of the tech community.
> I do not want to buy or own cryptocoins. Web3 does not look interesting to me at this point.
> Should I still learn about blockchains?
Yes.
Have a try for yourself and just learn whatever you want to and do not listen to any extreme anti-crypto or crypto-maximalist telling you either not to learn it completely or look at only their 'holy' and 'true' blockchain project.
The undeniable fact is; it is always going to be around and there always blockchain projects (NOT silly meme token projects) always emerging in the background, especially in payments and the internet itself.
> The undeniable fact is; it is always going to be around and there always blockchain projects (NOT silly meme token projects) always emerging in the background, especially in payments and the internet itself.
And they will always try to pay OP using stock options, because even the big guys at the helm who party in Miami on Bored Apes Yachts , they aren't sure about this thing and want to share risk with employees.
Frankly most crypto tech can be understood in an hour or two of focused reading. Understanding how it works greatly demystifies the malarkey out there and lowers your cognitive load when interacting with crypto nonsense. For this reason alone, and since the news is replete with crypto nonsense, it’s worth spending an hour or two. If you already have a basic knowledge of computer science and perhaps distributed systems you’ve got all you need.
They are networks worth understanding. Though the signal is flooded with noise, and there are many scams to every sincere effort, the evolution of strategies to adapt to the many issues we face are interesting, and will become increasingly more important.
I assume that if you don't bother now, you'll eventually feel compelled to. I would definitely recommend understanding how Bitcoin works before investing in any cryptocurrency.
Yes. A lot of the tech is very interesting, despite all the hype and fluff.
Yes because then you'll understand 99.9% of the industry is fraudulent. If you don't get there yet then keep studying
A bullshit stat that you happened to pull out of your ass through an emotional reaction to crypto. Even companies like Chainalysis, which work specifically in the industry of detecting and tracking fraudulent vs. ordinary crypto transactions, have repeatedly claimed that a majority of transactions aren't actually of criminal origin or money laundering. In other cases, people (ordinary people) have used BTC or crypto to avoid draconian controls on capital by governments. Illegal? Sure, but in a very absolutist moral sense. There indeed real business and financial activities in this space that don't involve fraud.
Con artists often gravitate to new technologies, see the internet itself as an example, and it applies especially when said technologies can be used to move information or monetary value around in some way. This doesn't make blanket condemnations of these technologies fair or valid. IT's absurd to see such idiotic condemnations constantly keep being parroted on a site like HN, full of people working in technologies that are constantly used by criminals too.
What real product is there in web3 or defi? Explain without using abstract buzzwords and handwaving explanations.
Here's a simple one: Remittance fees take a non-zero percentage of a developing nations GDP. [1] "The World Bank estimates that globally US$16 billion can be saved annually if remittance costs were reduced by a mere five percentage points." Regardless if the solution utilizes a blockchain or a cryptocurrency, you can agree that there is a problem, and ideally we could get remittance fees down to 0 or near 0.
[1] https://www.lowyinstitute.org/the-interpreter/bang-buck-gett...
yea bitcoin solved that and had nothing to do with the crypto movement or web3
What's the .1%?
It's obviously the....well could be....I don't know
Make decision by considering blockchain as data structure of distributed resilient auto-clearing accounting system. I don't think learning that would make you more "technologist" if such system doesn't serve you any purpose, cryptography and data structure rather make one "technologist" in general.
There's some interesting research going on. As of right now it's not really from a software/coding POV that it's interesting, but the mathematics being used and the implications: zero-knowledge proofs are a pretty hot area and have applications for both privacy and data compression. And there is some P2P networks and front ends stuff to be done too, but those are more well-trodden parts of the puzzle. Implementing a cutting edge example of the tech would be similar to working in graphics programming or digital audio; lots of first-principles stuff implemented from papers.
Personally, I just observe the research at some distance and hold tokens. The software side of things is churning immensely as everyone comes up with their own protocol concepts, so there's definitely work to be done.
Yes, people like to hate but many are silently working on some variation of it
Azure cloud already has a beta blockchain of some kind, saw it as an option while configuration a database https://azure.microsoft.com/en-gb/services/azure-confidentia...
More probable than not than in some years all it systems will have some sort of crypto ledger
The financial industry will keep an eye on it, and will definitely implement it internally and B2B
There are very small project tutorials, that implement small blockchain in your favourite language. But there are also small machine learning tutorials that are also interesting
A Merkle tree is not a bad data structure to have in your toolbox. That's learning the tech behind source control as much as learning the tech behind crypto I guess.
Depends. How well do you understand the internet? I know nothing about how networking works, yet I use it everyday. In 20 years, we'll all probably be making blockchain transactions on a daily basis, either directly or indirectly. Just like I'm creating packets, and I don't really know what that means I just hear networking people say it.
You should study crypto, blockchain, tokens and NFTs as an example of the type of fraud that is the signature of mania.
There needs to be investigation on the circumstances that make groups irrational. What exactly is the causality in the cycle of pandemic, mania, recession, war? We need a model for this.
> as an example of the type of fraud that is the signature of mania
That's the social aspect of it. I clearly mention that I am asling about the technical aspect of it.
you did, my bad
> distributed, cryptographically secure, permissionless ledger > is fraud
wat
If you're interested, watch a youtube video or two about it (e.g. https://www.youtube.com/watch?v=bBC-nXj3Ng4). You don't need to get super deep or spend too much time on it.
No.
Can you explain why you feel so.
Why bother, if you are (rightly) not interested?
Are you curious about the tech? If yes, what more reason do you need? If no - why proceed?
I’ve been searching for Rust development jobs and most things I find are related to web3, which im not interested to get involved in. I guess I’ll keep looking.
Good decision. These web3 companies can be very sleazy. They don't have real products and only get revenue by hypering up their token sales
If it's not interesting learn something else. Not everyone wants to work on payment systems or swap / stake web pages.
Why would you learn about something you don't find interesting?