5 ms·
No, that doesn't sound like web3 at all. Web3 runs on different rails than Web2. Web3 typically runs on IPFS which is like an upgraded HTTPS where instead of j
by Sargos 5y ago
No, that doesn't sound like web3 at all. Web3 runs on different rails than Web2.
Web3 typically runs on IPFS which is like an upgraded HTTPS where instead of just one computer hosting a file on the internet it's instead lots of different people around the world. It's pretty much impossible to censor a file on IPFS. IPFS also allows easy offline web apps and a built-in CDN.
Web3 also uses a decentralized execution and state system where instead of your data being in one companies' servers like Facebook it's instead hosted on thousands of people's copy of the database. The important state data is impossible to censor.
In Web3 the infrastructure is distributed and no one controls it. You can run an app without having to trust anyone.
- wizzwizz4 5y ago> Web3 typically runs on IPFS which is like an upgraded HTTPS So, there's one centralised portal, like with Google AMP? (Perhaps this is cynical.)
- wait_a_minute 5y agoFiles get uploaded and then distributed across nodes in the entire network. They are accessible via gateways which could be centralized, but you can also access the files directly as well from a browser that supports it. Brave supports it. So as a piece of infrastructure for hosting content and websites, it's decentralized and as more nodes come online it becomes more and more fault tolerant. https://filecoin.io/filecoin.pdf https://filecoin.io/filecoin.pdf https://docs.ipfs.io/concepts/ https://docs.ipfs.io/concepts/
- wizzwizz4 5y agoDidn't know Brave supported IPFS. Perhaps that will work, then.
- klyrs 5y agoSounds great against censorship, but what about surveillance & privacy?
- wait_a_minute 5y agoYou can build dapps with privacy in mind. It's an OSS-friendly world where you can verify the contents of the smart contracts you are interacting with and see what the contents of a website are if the website is deployed to IPFS. It's not perfect since it's an entirely new way of building web apps, but it's better than proprietary systems. You can create new wallet addresses, for example, when you want and sign in with those. Or your dapps can use https://magic.link/ https://magic.link/ and users can just use private email addresses if they want. Generally, for surveillance I think the legacy systems we're all still using are much worse. Firstly because we don't always know what they're doing, and secondly because these companies can say one thing and change their mind next week. Like how Apple was supposedly a privacy champion, and now they're leading the charge for building on-device surveillance tooling.
- klyrs 5y agoSorry, but "private email" is an oxymoron. This system sounds open, but also indelible. Can you make a self-destructing file? Sounds like your pseudonymous data can tracked, fingerprinted, and correlated to your real name, to me. And no rush, the data ain't going anywhere. Privacy's gotta be engineered. Today's strongest crypto will be plain in 100 years.
- wait_a_minute 5y agoI agree with you that privacy has to be engineered. Please consider getting involved in some Web 3 projects! I'm not an expert yet but I don't see why self-destructing file sharing couldn't be built in the Web 3 world. You could also build hybrid apps, where most of the app is a dapp and the sensitive feature like self-destructive file sharing are housed on proprietary systems guarded by a company with a reputation for security/privacy. I think that if today's strongest cryptography becomes plain in 100 years, I'll be okay with that since I'll be gone and the future generations will have figured out more advanced cryptography :) Or society will have moved away from privacy because we're all linked to each other with neural laces and everyone would know everyone's deepest thoughts.
- jazzyjackson 5y agoEvery IPFS node advertises the content it is making available on the network. As soon as a hash is deemed an illegal number, it would be trivial for an ISP to block nodes providing that hash. Maybe no one controls the end points, but Verizon-AT&T-Comcast still own the wire. From the docs: “information about which nodes are retrieving and/or reproviding which CIDs is publicly available.” https://docs.ipfs.io/concepts/privacy-and-encryption/ https://docs.ipfs.io/concepts/privacy-and-encryption/