Hacker News Reader

There's a question that isn't answered and that's why would users care about those details of the backend. In the end, a solution can be as technologically elegant as you want, but the thing it really needs to do is improve from the users' point of view. And I don't think it does.

For most people, a decentralized platform will usually be worse (speed, latency), and only very rarely better, than a centralized server. The "paying to access a database" will introduce a lot of issues both for users and developers, no matter how you implement it, for barely any user-facing improvement. And centralized processors appeared to make things easier, so you'd need a payment platform that is as least as easy and, also, that protects you against scams (AFAIK, you can't do too much if someone steals your crypto wallet or you forget the password).

So while I agree that web3 could provide identify + payments, it doesn't need to be technically elegant, but user-facing elegant.

A BTC wallet is infinitely more user friendly than a banking website that only works in chrome xor firefox and constantly locks you out for reasons they don't want you to know about and severely limits your transfer sizes. If you do EFT and wait between 3 and 10 days each time, even bitcoin is better than this.

Centralized web is horrible and dead. Literally _anything_ is a better alternative. The user can be convinced to adopt superior tech even if he does not explicitly know the problems. It just has to feel better.

> it doesn't need to be technically elegant, but user-facing elegant.

dont worry bro. the "set user = happy" voodoo dance has already been done. every cryptocoin wallet is a browser plugin or mobile app with 90% of code for themeing with all the bugs and slowness smartcrap/web users are accustomed to. you are even discouraged from rnunning yoyr own node

> A BTC wallet is infinitely more user friendly than a banking website...

Until there is fraud, a dispute, or an innocent mistake.

Same for traditional banking + if you happen to be a journalist ( or whatever ) living under an oppressive regime, they can freeze your assets on a whim, because you criticised the wrong people. If someone still prefers trusting a central authority, for whatever reason, they can still use something like coinbase.

If you're living under an oppressive regime, they can freeze turning your crypto into state-approved currency at a whim, and you have the same problem.

Crypto is not binary.

The vast majority Og people will likely use a custodian for their crypto. For them it will feel like banking feels now.

Just like coinbase provide a visa card with collateral i your crypto.

So essentially these questions really are answered -- it is not an all or nothing. We will keep, eg. Visa to facilitate retailer transactions with collateral in, eg., bitcoin. Bow you can dispute and everything and you pay in crypto.

The parent is talking about using blockchain not just for payments, but as a building block for the web providing both identity and payments.

Also, why would we want bitcoin or any crypto if most people will just use Visa? How does that achieve decentralization? I doubt the goal of cryptocoins is just "replace the financial backend that big companies use".

> With web 2.0, you have to trust the backend (so again in most cases Google or Amazon) and that is a limiting factor for sensitive apps

And this is where the assumption is wrong. Web3 is about gatekeeping for profit, and additionally you not only have to trust a single vendor...but literally every other peer on the decentralized internet to be trustworthy with your data.

Computational security is not the same as data security, and this is where the marketing BS is happening.

As long as there's no permission management and no deletion feature built into the blockchain, your precious private data is as publicly accessible as it never was before.

You think encryption will hold forever? SHA might disagree.

You think 20 years ahead of time the password will still be secure and not leaked anywhere else? Ha, good luck.

The problem that I have with web3 is that 99% are ponzi schemes for gatekeeping ...and the 1% that are left try to market it as a privacy thing, which it cannot be by definition of how a blockchain works.

There are so many broken software programs out there already, and now we made those bugs persistent and unfixable for eternity (with respect to "how secure" user data is stored).

If the web3 would be about real distribution (not the same as decentralization), there could be much much easier solutions for having a local copy of your old myspace photos.

Yeah, even if the web3 moniker has been spearheaded by Ethereum and Parity/Polkadot related projects (it’s been a thing since ~2015 and until about last year its use has mostly been following Gavin Wood and crew), in my mind for a typical web3 app that doesn’t involve payments, the only blockchain/crypto-related functionality would be the built-in key management and using it for authn/authz. Possibly signing messages that are relayed p2p or c2s, no need to post transactions on the blockchain or hold any cryptocurrency at all.

For web2 (which was also abused to death as a buzzword), the defining technical component was asynchronous requests through js and dhtml. Decoupling the rendering of views from data.

For web3 it’s decoupled frontends and backends (the frontends should support being self-hosted/run locally), peer-to-peer and federated architectures, having not SPoF.

Js and Ajax calls where useful to everybody, including the companies paying for developers time. Web3 with P2P and federation seems exactly the opposite of how companies make money. So it's what I'd like to use but it's not what I'm expecting to be paid to develop.

Actually I've got a customer using Ethereum first and then Algorand to timestamp events but it was only a little more than calling a HTTP API for us.

Imagine a world where infrastructure, libraries and developer tooling are all there. Imagine the reduction in costs for infrastructure, security (sure you still have to do it but the body looks completely different) and all that goes with having a "backend". Suddenly you don't even need to run a business to afford supporting an app that loads of people use.

Sure, you will still have backend processes but the exposure to "the app" is greatly reduced.

That sounds great if it lowered costs, but it takes gas to run services on blockchains - it’s still “someone else’s computer” that has to be incentivized to do work, whether that’s executing a smart contract or paying to pin files on IPFS.

IMO there’s room for small benevolent communities where people choose to share in the cost of infrastructure (e.g. hosting seedboxes in torrenting communities where you are expected to keep your up/down ratio even) but expecting users to run full nodes is a tall order, computation and bandwidth-wise you’re offloading costs to your users. I always thought something like wikipedia could work like this since there’s thousands of committed super-users that could afford to run a background process, and indeed there’s IPFS mirrors of it, but IIRC the edits still happen on jimmy wales’ computer.

I kind of mentioned it already, but apart from payments or other transfer of digital ownership, there shouldn't be much need for on-chain transactions. In some years when, hopefully, L2 technologies have matured and grown to live up to their promise, those won't have to either (there is always the initial on-boarding of course, but credit cards and mobile payments have those too)

As for "who's going to run the nodes", if cryptoeconomics or users don't take care of it, I assume businesses will. The nice thing is that regardless, the payment for this infrastructure is also facilitated through cryptocurrency transactions on L2 networks. For someone who doesn't want or need to take ownership of that, it can just be proxied through to the client and not need to be the concern of the developer of publisher of the app.

Of course businesses can still choose to pony up for it proactively of they have the finances to support it, potentially running it themselves or through a third party. You can always fall back to "the old ways" for certain things when it makes sense.

If by L2 you are referring to proof-of-stake, I don't think that technology can ever be made to work without needing to exist alongside a proof-of-work coin due to the "nothing at stake" issue.

No, I am not talking about PoS. One example of "Layer 2" is Lightning Network for Bitcoin. You "lock up" an amount of BTC in a channel that you can then use over the p2p onion-routed liquidity network. Payments are practically instant and fees are low (and can be made 0 if you open a direct channel with the counter-party).

It's orhogonal to consensus so works equally well on PoW/PoS.

There are a couple of competing and complimentary solutions for L2 on Ethereum. I'm quite sure it will converge at some point.

The problem with web3.0 is decentralizing all the things is super hard and really user unfriendly. At this point i'd be happy with something more like web 2.5 where the important bits are decentralized from a central authority (identity management for eg) and able to integrate nicely with centralized platforms.

Indeed it's hard to build a truly decentralised app right now, that's why I referred to "future version of web apps"...

Most people deploy a backend proxy and use APIs for indexing, etc. Originally, these APIs were centralised (e.g. Infura) but decentralised alternatives now exist (like The Graph).

> where the frontend is a compiled SPA

This is pretty much just the same thing we do today, but wait...

> possibly hosted on decentralised platforms (IPFS)

We can use a filesystem that is extremely slow, and also loses all our files if we stop paying AWS to mirror them in S3!

> the backend is a decentralised blockchain

And we can use a really slow database where each write costs $100s in gas fees!

... see, much better!

Decentralisation is hard - never told the contrary.

However, transaction "gas" fees with some chains (Polygon, Solana) are less than 1 cent and you probably know that.

You can pay as much with DynamoDB if you don't optimise anything (and reads are free with the chains, still costly with DynamoDB). Of course, you can say that AWS is overpriced too... maybe that's just inflation going on.

1 cent per write to a database is insanely expensive, and no, you can't pay that much for DynamoDB. DynamoDB is free tier for 10 writes/read per second.

If you go with AWS's most expensive option for access (DynamoDB on-demand, infrequent access), the pricing is $1.56 per million write request units. That's 0.000156 cents per write. That doesn't include data storage, but that ranges between 10 and 25 cents per month, per GB.

Where does this preoccupation with payments come from all of a sudden? Personally I don't want "payments" implemented in my web browser, and I seriously doubt that that would be in anyone's interest.

Micro payments may prove to be a compelling alternative to. Or at least that is the hope.

Alternative to what? I think you forgot a word there

Ads

Ah! I can kind of see the idea, but crypto seems... kind of overengineered for that?

There was flattr [1] a while ago, with a similar concept, but it didn't really take off. I'm not sure a more complicated solution with a higher barrier of entry would.

[1] https://en.wikipedia.org/wiki/Flattr

ads, sorry the word got dropped

I found it pretty eye opening to be able to make payments freely and (practically) pseudonomously, in contrast to which all other digital payments systems are heavily permissioned. Crypto UI may not be great but try sending a wire internationally in the banking system and you are presented with 2-3 massively leaky abstractions (why should I the consumer have to research the difference in ACH, electronic wire) with various fees, delays, and possible invisible censorship issues (blacklists, whitelists, reports triggered) PER bank.

I make international payments pretty often and have never been presented with a leaky abstraction, suffered from "censorship" or anything like that. "Permissioned" is simply a consequence of that fact that most countries establish barriers to the free movement of people, goods and capitals in and out of their territory. You're not going to solve that with technology, because it's not a technological problem. It's not even clear that it's a problem.

I’m sure many people haven’t experienced that kind of censorship but lots of people have. I used to play online poker which was made illegal in the US through payment channel crackdowns(for reasons at best paternalistic but realistically just physical casino lobbying). Given that the internet facilitates pretty much any kind of information transfer I have to ask mysef is it more likely the future will be a future where I can only transfer value by requesting 20th C institutions to proxy for me or is it more likely I will be able to freely transfer value in my own right.

If you're talking about transmitting money, it's a regulated activity in the US. [1] If you think that you will be able to bypass regulations by using crypto-currencies and suffer no consequences, I think you're in for a surprise.

[1] https://en.wikipedia.org/wiki/Money_transmitter

Nothing I’m talking about requires me to get licensed to operate a money transmitting company.

I've been Googling the word SPA and I keep finding places to go for a holiday, but nothing relevant to what you're talking about.

SPA in this context would likely mean Single Page App. A Javascript module, to be run in a browser.

If you include a context word like "web" it pops right up.

I tried a bunch of additional keywords before asking. I also did think of "Single Page App" but I doubted how it could be relevant.

I agree that posters need to be careful about undefined acronyms, but, pro-tip, you can always check Wikipedia for overloaded terms to narrow down to the right one. Here it will give you a section for the meanings in the context of computing:

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

Edit: also, I suspect you’re being cheeky there, because when I Google “SPA”, the top results are local spas (even in private mode), and thus not “places to go for a holiday/vacation”, as if you only guessed what a Google search would give and had decided to make a joke about the ambiguity and most common lay meaning of (case-insensitive) SPA, rather than express an actual frustration at an honest attempt to look up the meaning of that term.

There's more than one technical meaning. One is web-related and another is cryptography-related. All possibly relevant.

Yes, the other one is a small stub article about a proprietary Microsoft protocol for authenticating email logins.

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

I guess that’s what led you to “keep googling” the same term over and over?

Who's maintaining the service instead of AWS in the web3 scenario?

There might be more, but the only one I checked out is https://akash.network

Well usually whoever deployed the web3 product is collecting fees for its use in some capacity so they have a vested interest in developing it.

Either that or they are holding a huge bag of some project-native token and so they have a vested interest in developing it because increased usage usually correlates to the price of said token increasing as well.

I think that just as AWS, Stripe and few other players cornered the infrastructure and payment market, as things scale and consolidate, few players will control the nodes and control the blockchain as well. So things won't change much.

For the ones that don't know:

SPA = Single Page Application

why it has to be SPA?