Web3, DApps, DeFi?

DISCLAIMER: This is not a wage slave thread, neither a finance thread. Keep you money talk somewhere else. The only /λ/-related take would be doing data science on historical data, but forecasting is partly just palm reading done by white collars if you ask me. What do lainons think about the web3 paradigm, DeFi and DApps in general? Looks like there's plenty of room for being creative. I am trying to dive into it all but don't know where to start. So far, all I know is that knowing Solidity is p much canon an this point (https://cryptozombies.io/en/course). Sharing books, stories, and opinions is welcomed.

>the web3 paradigm well web3.js itself is a pretty garbage piece of hack webcrap code and it's all downhill from there 99% of the stuff done on this "decentralized tech" is done using remix.ethereum.org and whatever blockchain explorer, largely due to the offline tools being gimped crap solidity is a terrible NIH language (lol inheritance in a finance language based on java, even after java was repeatedly shown to be insecure under mutually suspicious code mostly due to inheritance ["javapocaylpse" et al]). every gotcha in ethereum is well known stuff modern PL's have avoided like that soykaf where it used to automatically make some variables into a reference without explicit syntax only in some case (copy/pasted from some enterprise braindamaged person's blog): contract Sample { struct info { bool doOnce; } mapping (uint => info) infoMap; uint _actionNumber; function() { var a = infoMap[_actionNumber]; if (a.doOnce == true) { throw; } a.doOnce = true; msg.sender.call(); _actionNumber++; } } should be self explanatory. some variables are references, some aren't. they patched this out with gimped crap but at least you don't get cucc'd by retard semantics anymore sending money to another contract from your own contract is a vuln by default (this could have been avoided by several alternative designs) there was some dumbass sh1t surrounding delegatecall or some other related operation, where your variables and memory will be reinterpreted as that of another contract. literally C tier nonsense which has no place in high level security-first language. constant useless bullsh1t lint-tier compiler warnings ABI design is sh1t too. bytecode is pointless and makes things more unreliable as well. it's literally no faster than executing from an AST and makes no difference how gas computation would work. the idea of having a language agnostic layer is also bloat (= more bugs = more $billions of losses) and braindamage since the unnecessary serialization and crippling of language A and B leads to more vulns. ethereum's one good idea is applying smart contracts (old idea) to blockchain. which anyone could have done. solidity can be summed up in one sentence: they literally think they have a secure language because they took some crap with no hopes of being secure like Java/C, and added "pure" annotations to functions and a few compiler warnings

>crippling of language A and B leads to more vulns. also what i mean here is that A cannot call $otherlang without crippled sh1t. you'd know what i'm talking about if you've experienced JVM or CLR whereas a pure single language system will not make you do weird unreliable vuln-prone sh1t to workaround serialization concerns

>>23902 What about Cardano?

>>23891 I discovered DeFi this summer using Argent wallet, it has the tab invest that made me discover aave, compound and others defi protocols, simply the banks can't compete with it (apart from the fact that banks are regulated and give you guarantees also if they get hacked), but a part from this Defi absolutely win. At the moment is like the far west. One interesting thing that I'm tempting to learn are flash loans and I found this interesting website for now https://money-legos.studydefi.com/#/

>>23917 Not that lain, I've seen stuff like their hello world example not compiling or some soykaf on r*ddit, but I couldn't discern if it was just FUD or that something is really wrong with their stuff. I've been thinking about getting into blockchain, seems to be a well paying field, but >>23902 makes some points that I've seen before and keep me from diving too deep into it. I've had enough bullshit at my current job to deal with crappy tech in my time off. In general, however, it does seem to be pretty interesting, considering the possibilities. If someone has a recommendation on a good beginner book, I'd be a good addition to the thread.

>>24997 I work in crypto, I recommend Mastering Ethereum by Andreas M. Antonopoulos for learning about the architecture and How to DeFi by Darren Lau for learning about the different smart contract protocols that you can use, learn from, or plug into

>>25234 How you found the job?

>web3 Not crypto but Urbit is the fever dream of an lisp or smalltalk image based OS but with the actual possibility of being adopted. The primitives are pretty nice once you get past the obscurity of hoon.

>>25582 I found it on Twitter actually, but AngelList and Discord are good too. Just poke around and ask questions. There are a lot of jobs right now in crypto and not enough people to fill them

Some thoughts: >Defi is going to be big, eventually will displace banks, stocks will be the blockchain, the genie is not going back in the bottle >With that said, most non-base layer coins are likely to end up worthless >It's still very early, if you buy a bit of this and a bit of that, you will make a lot of money. If you learn to develop defi starting today, you will probably get rich. >As this lainon >>23902 says, ETH is a mess, but it has such strong network effects that it will probably linger around indefinitely >I work at a startup that's building on Polkadot, only started recently but have positive impressions of that project. >People I respect told me ADA is a viable project and a strong candidate to win it all >Follow banteg on twitter or github >All kinds of crazy wild west stuff going down every day like this https://github.com/Defi-Cartel/salmonella

>>25616 I'm following many projects on twitter and until now I've seen two account I follow that searched for somone. But I don't use twitter everyday maybe I lost some opportunity, I also into some telegram group, but I think I will give discord a try. Can you raccomend some good discord or telegram chat filled with developers or people that don't think to much abouyt the price? Did you find a job as a solidity developers?

>>25620 The best channels are small so I'm afraid I can't help :) but yes I advertised that I have solidity experience and deploying on mainnet

>>23902 Developing a Web3 app using the ethereum ecosystem now and I word for word agree. If there wasn't already a strong ecosystem in Ethereum, I'd much prefer a different blockchain.

>>23891 How can I develop for ethereum without installing yarn, node and other bloat? For now I'm just using remix.ethereum.org to learn. I've already installed geth to run my light node. So I'm think I will try to deploy to some testnet (like ropsten? rikeby? Goerli?) to try out my smart contracts, and then I could use golang instead of web3.js. There are some good guides? Some suggestion? Should I embrace javascript? For now I don't see the point because I don't plan to do frontend development

>>25735 >How can I develop for ethereum without installing yarn, node and other bloat? >For now I'm just using remix.ethereum.org to learn. Unless you really need unit tests don't bother moving past remix, you will need to install and figure out how to use a mess of horribly broken overly complicated soykaf. Remix may be a soykafty browser IDE but at least it works most of the time as long as your code isn't too big. >I've already installed geth to run my light node Don't bother with this its only pain with no benefit, just use Infura. >There are some good guides? Official solidity and web3 documentation is pretty good. Overall the educational resources are very lacking. Some basic tips: * Never use structs or lists, only mappings, this will avoid a wide swath of pitfalls * Use uint256 for everything, mind the over/underflows and integer math * Never use loops of any kind * If calling other contracts or sending Eth, read about and understand re-entrancy * Don't do fancy things keep it painfully simple, fuarrrkups are easy to make yet unfixable with big consequences. If someone is asking you to do anything complex, push back hard. >Should I embrace javascript? For now I don't see the point because I don't plan to do frontend development Frontend developers tend not to know how to how to connect with smart contracts. If you plan on being hired, being able to do this part makes you a lot more useful. If you plan on doing your own project, it's mandatory since people use smart contracts via websites. Just learn enough javascript to make the buttons work.

>>25587 It seems interesting to me, but I have not found anything interesting going on on that ecosystem, and I find hoon a total putoff. Is it worth it?

>>25844 >Don't do fancy things keep it painfully simple, fuarrrkups are easy to make yet unfixable with big consequences. If someone is asking you to do anything complex, push back hard. This so hard. A lot of Ethereum devs have this obsession with doing everything on chain, but it makes their contracts vulnerable and the network expensive.

i've been thinking about how to simplify client-side web. here's my idea: just WASM and the DOM. the client gets a WASM (from either a http server or ipfs) and then that WASM is ran, and has the ability to create and manipulate elements in the DOM. browsers no longer need to parse HTML or run JS: if you want to use those you can always compile HTML parsers or JS interpreters as WASM and send that.

>>25617 ADA will not win it all. The fact you said that means you have no idea what you're talking about. Charles is narcissistic pathological liar and they have no one who wants to make soykaf on there

>>26087 >ADA will not win it all. The fact you said that means you have no idea what you're talking about. Don't take it from me, take it from Jim Donald, who has been working on cryptographic money since the 90s and was the first person to correspond publically with Satoshi. https://blog.reaction.la/uncategorized/back-again/#comment-2705381 >Charles is narcissistic pathological liar Probably good for the currency on balance to be honest >hey have no one who wants to make soykaf on there Doesn't really matter yet. The winner will be the protocol that can scale consensus bandwidth.

>>23891 I didn't know that it existed until this thread, and even after looking it up on duckduckgo I still don't quite know what it is. From what I understand it's about decentralization, which is cool, but I can't really form an opinion on something without knowing what it is. On the whole though, I think that the web has been stretched way beyond what it was designed to be; text documents accessed over the internet with links to other text documents. It ended up absorbing every other internet based technology and doing them worse.

>>26087 >>26087 >Charles is narcissistic pathological liar Do you know Charles personally? Do you know anyone who knows him? Do you have any reliable information on this because to me it seems like you are just repeating soykaf you heard once and stuck with it.

i have not dove into ada smart contracts as the mainnet has yet to be released still, but what gives it a serious chance of taking on solidity/EVM is that is has a built in oracle language that makes protocols such as Chainlink obsolete. that being said, mainnet still isnt out, so for all we know its still just vaporware. a few interesting projects have been deployed to ada testnets, but the majority of ada testnet projects released are NFT-related and repetitive. lots of bologna as for solidity/EVM, the bread and butter of work for development at the moment is building ponzi schemes/scams through contract work, and is rather discouraging. i picked up solidity as my first programming language sometime around last october, and while i've learned alot about what i DONT know, the majority of the work feels wrong. i would say some of the more interesting things to look into developing revolve around; >decentralized exchanges >lending protocols >different types of escrow services >stablecoin development and finally... >breaking all of the above, alla flashloan attacks

>>26281 Literally everyone knows how narcissistic charles is. He says dumb soykaf on twitter pretty often. Whines about being told to email support >DO YOU NOT KNOW WHO I AM?? Whines very publically about someone not adding ADA to their list of good L1s, and didn't even have smart contracts working lmao (they still don't).

I'm only briefly going to mention this. Kadena with braided PoW. Functional scaling without bloat, today. Not vaporware like everything else. Pact > Solidity.

>>26108 >I don't know what it is, I just looked it up, I still don't get it, and here's my opinion.

I've been dipping my toes into it for the past week or two, mainly on the back end with geth (I know I ought to learn JS/Solidity some day but I choose to put it off). Something that's struck me is how little information is updated to reflect the most recent developments in Ethereum mainnet code. I've had to use a 3 year old e-book and piece my way through it with very outdated code examples and a lot of research/asking people to get something working. I'm assuming this is unique just to Go based Ethereum development (is everything just word of mouth/based in Discord servers? seems inefficient) since web3.js and all that seems to be the dominant force with 1000 tutorials. I'm very confident in the tech, always have been since 2013, but I can admit it's mostly chock full of scams, hype, and horrible code (rekt.finance) right now. But I also think we're in the very beginnings of all of this and there's bound to be growing pains. Maybe I'm naive, but I'm still going for it anyway because I think the possibility of a decentralized internet is worth the risk.

>>27284 Not an expert but feel the same. Have a feeling there is a lot of spaghetti ideas at work.

starting with eth because it has the most resources. getting a dev environment going with hardhat+node+react is utter hell. hundreds of dependencies, broken code. APIs constantly changing. i just want to make a memecoin

>>28515 feel your pain bro, though been doing web development for long enough to get used to the hell. What's really pissing me off is the absolute state of the Uniswap2 documentation. Also ETH based tokens will work on BSC and MATIC pretty seamlessly. ADA and to-be vaporwave is not worth doing right now anyway unless you wanna take a chance on being an early adopter.

>>28515 >getting a dev environment going with hardhat+node+react is utter hell. hundreds of dependencies, broken code. APIs constantly changing. Well, yeah. You're dealing with Node. What were you expecting?

>>23891 I did some stuff for fun with web3 websites. I think the idea of wallet in a browser extension that interacts with websites is a very good one. Your extension wallet injects a JS object on window object in a page so the page can interact with the blockchain trough your wallet. I am planning to extend my fun little project with some smart contracts but I'm not there yet. I'll take >>25234 into consideration. >>25587 I love lisp and took just a brief look into Urbit. It was immediate put off with bullshit website full of esoteric names and phrases. I want a manual not a LOTR novel.

>>25870 It's a slow moving ecosystem from the outside, there's a lot more activity with the social aspect of it rn. >>28565 Really the interesting parts are the networking layer having identity baked in and the OS/interpreter having a functional style of design. Otherwise it's just more fuarrrking web3 hype in a slightly different package

>>28565 > I am planning to extend my fun little project with some smart contracts but I'm not there yet To give update on this topic...I've started writing Solidity contracts and I must say that tooling is the worst I came across in any ecosystem so far. I rate Remix IDE as the worst programming environment I had ever came across. It's worse than anything in Javaland, mobileshittery and even webshittery Also Solidity feels very brittle because you do lots of state mutations everywhere which wouldn't be such a problem in an enviroment where you can fix bugs, but here everything must be correct at launch. Sorry I had to vent like this. But I am kinda disappointed by the crypto space right now.