Ask HN: What do you think about decentralized websites/apps?
2 comments
I think I am more interested in “decentralized” in the sense of local-first, federated (not p2p), community-supported, resilient web. That does not require blockchain.
It’s also too bad that the mindshare for “Web3” pushed out “Web 3.0”, that is the (arguably failed) Semantic Web. Semantic Web would have allowed for decentralized structured data. The successor ideas include what powers ActivityPub (JSON-LD, effectively RDF for JSON without calling it RDF). I also find it interesting that Tim Berner-Lee’s new startup is trying to take Semantic Web’s idea to create a personal data technology where that data is controlled by the user.
The ideas of Web 3.0 does not require blockchains.
IPFS is interesting, but the critical use-cases I see are not for serving web. If you think about what it would take to recompile Linux distros or any number of web apps from scratch if Github or other forges go down, it becomes very difficult. Package distribution based on IPFS would be more resilient, while also bringing some interesting benefits. (Cost is borne by the people who care about the package; fetch from neighbors if they have it)
Following on that idea, I’m interested in building out a decentralized forge, one that is community supported rather than corporate-sponsored. By that, I mean the features are intended for a small or medium group, communicating with other similar groups over unreliable or slower networks. If capital is raised to help fund additional features, it would be raised kickstarter style with the resulting code made available for all.
It’s also too bad that the mindshare for “Web3” pushed out “Web 3.0”, that is the (arguably failed) Semantic Web. Semantic Web would have allowed for decentralized structured data. The successor ideas include what powers ActivityPub (JSON-LD, effectively RDF for JSON without calling it RDF). I also find it interesting that Tim Berner-Lee’s new startup is trying to take Semantic Web’s idea to create a personal data technology where that data is controlled by the user.
The ideas of Web 3.0 does not require blockchains.
IPFS is interesting, but the critical use-cases I see are not for serving web. If you think about what it would take to recompile Linux distros or any number of web apps from scratch if Github or other forges go down, it becomes very difficult. Package distribution based on IPFS would be more resilient, while also bringing some interesting benefits. (Cost is borne by the people who care about the package; fetch from neighbors if they have it)
Following on that idea, I’m interested in building out a decentralized forge, one that is community supported rather than corporate-sponsored. By that, I mean the features are intended for a small or medium group, communicating with other similar groups over unreliable or slower networks. If capital is raised to help fund additional features, it would be raised kickstarter style with the resulting code made available for all.
As for "Web3.0" being ignored: what do you expect from a movement led by the likes of Paris Hilton and Jimmy Kimmel?
Pro tip: only take technology advice from ugly people.
Pro tip: only take technology advice from ugly people.
I never heard of Jimmy Kimmel and I don’t know of either of their connection with the Semantic Web.
I do know Paris Hilton is a savvy businesswoman who created her own business empire while cultivating a celebrity image that makes her look dumber than she actually is.
I do know Paris Hilton is a savvy businesswoman who created her own business empire while cultivating a celebrity image that makes her look dumber than she actually is.
It is those kind of people who ignored "Web3.0" in the name of "Web3". Kimmel was famous for having a cringeworthy interview with Hilton about NFTs.
For me the "semantic web" was like one of those asian "gods of poverty" that show up in places like Touhou. After more than 10 years of being involved with it I finally met somebody who knows who to do interesting things with OWL and I think she thinks I'm the first person she met that cares.
For me the "semantic web" was like one of those asian "gods of poverty" that show up in places like Touhou. After more than 10 years of being involved with it I finally met somebody who knows who to do interesting things with OWL and I think she thinks I'm the first person she met that cares.
There are a handful of people who think there is some there there with Web3. For the rest of us "ownership" is at best "A TRS-80 with a coinslot attached".
If it cost spammers $50 to send me an email that might be a good thing but it's more like $30.71 to log into my computer in the morning, $17.44 in gas fees to check something into git on a good day, $771.41 if the network is busy.
I am sure it would have made Sam Bankman-Fried rich except that he's going to jail. You might think you're the wolf who is going to get rich at the expense of us sheep but odds are you are the one who will be left holding the bag.
I heard there are some people who trade loli pictures on Tor. I bet they'd be really interested in a way to publish that can't be censored.
If it cost spammers $50 to send me an email that might be a good thing but it's more like $30.71 to log into my computer in the morning, $17.44 in gas fees to check something into git on a good day, $771.41 if the network is busy.
I am sure it would have made Sam Bankman-Fried rich except that he's going to jail. You might think you're the wolf who is going to get rich at the expense of us sheep but odds are you are the one who will be left holding the bag.
I heard there are some people who trade loli pictures on Tor. I bet they'd be really interested in a way to publish that can't be censored.
What use-cases do you think would be uniquely suited to benefit from this? Are there any? Would you build one?
Or is existing Web2 infrastructure decentralized enough that there's little benefit to full decentralization?
E.g. Censorship-proof, ownership.