All of this stuff is built upon the premise that decentralization is a good thing, but I have yet to see much supporting evidence. I do however see a ton of evidence to the contrary.<p>(I see centralization as more like clustering than a unifying of everything.)<p>Decentralization happens in the real world out of necessity, not because it's preferred. In fact, we centralize everything that is convenient to centralize because of its efficiency. And decentralization tends to occur at the boundaries of cooperation, where the problems lie.<p>So why would I want enforced decentralization when doing so would be inconvenient or inefficient?
Isn’t the issue with these that ultimately it relies on someone to make good tools around, and whoever ends up owning those tools ends up owning and re-centralizing all of these decentralized entities? (e.g. Coinbase owning everyone’s wallets)<p>One example they posted was that Bob is a music lover. So he keeps his music on his web node. I don’t know how Bob is going to do that if it’s not painlessly easy. And it needs to be in some sort of schema-defined format, and to be honored by all decentralized apps. So Bob is going to put it in something called “MusicBase”, which now centralizes and owns all of the data + controls the schema.<p>A second aspect is that nothing seems to really stop the web from being decentralized. e.g. Scraping, APIs, and integrations. It’s just that orgs have learned that lock-in is advantageous and therefore found their way down that path.<p>Am I wrong about this?
All I can say is... I've seen it all before now? on "web2", nothing ever prevented someone using cert/key based auth, and there were libraries to abstract over proprietary APIs, but if you didn't want to go that way, there is also <a href="https://remotestorage.io/" rel="nofollow">https://remotestorage.io/</a> et al<p>biggest hugest problem though: no one cares, go ask your mom about the difference between web1, 2, and 3
We’re going to keep doing Web++ until we learn to stop worrying and love the mess that is online identity. The fact is that the distributed data model will never be clean because, in aggregate, people don’t really care and don’t neatly fit into the structures engineers want to impose on data. This is why experience started winning the app wars.
Interesting! Seems to have a fair bit of crossover with what the Fission team is doing. Yesterday I stumbled upon a web page re: a presentation that a key Fission dev/founder (former Ethereum Core Dev) will be making in late September[1].<p>I ended up spending some time playing with Fission Drive[2] and looking at their Guide[3], and just generally reading their dev[4] and marketing materials[5].<p>Anyway, looking at the Web5 site, it seem to strike some of the same notes.<p>I'm not affiliated with or participating in Fission or the Web5 projects in any way, but am working as part of a different team developing an open protocol for decentralized storage, which is focused on incentivized data durability.<p>[1] <a href="https://www.thestrangeloop.com/2022/a-distributed-file-system-for-secure-p2p-applications.html" rel="nofollow">https://www.thestrangeloop.com/2022/a-distributed-file-syste...</a><p>[2] <a href="https://drive.fission.codes/" rel="nofollow">https://drive.fission.codes/</a><p>[3] <a href="https://guide.fission.codes/" rel="nofollow">https://guide.fission.codes/</a><p>[4] <a href="https://github.com/fission-suite" rel="nofollow">https://github.com/fission-suite</a><p>[5] <a href="https://fission.codes/" rel="nofollow">https://fission.codes/</a>
This idea depends on two non-starters:<p>1. A completely centralized data authority (in this case, Bitcoin).
2. User's running their own infrastructure.<p>"Put everything in one giant bucket, and make users 100% responsible for managing that bucket without any room for error" is not a workable path forward.<p>Developers may love this type of approach, and appreciate the advantages it would offer, but it's too much to ask of my 70 year old mom or 11 year old nephew.<p>An easier approach would be Passkeys and localStorage transferability between domains, or export feature on sites.
There's a lot of infrastructure around DNS already, it seems to be an under-discussed tool for self-soverign identity. It does depend on ICANN and domain registrars but there is lots of precedent for transferring your domain between registrars or changing the provider you use to serve content at the domain. This is what IndieAuth[1] is built on and I'd love to see more discussion around it.<p>[1]: <a href="https://indieauth.net/" rel="nofollow">https://indieauth.net/</a>
It takes a lot to look past the branding, but I think the vision is interesting. I think there _will_ be more personal ownership of data in the future, but not across all services. Decentralized, permissionless self-custody data only makes sense for narrow use cases.
no one from TBD or even Jack can clearly state what the problem is, beyond some conspiracy laden, techno babble. with square stock going to 0, you'd think they would be worried about other things. But that's Jack as a CEO for you.
<a href="https://mobile.twitter.com/CoinDesk/status/1535333277740965892" rel="nofollow">https://mobile.twitter.com/CoinDesk/status/15353332777409658...</a><p>> Web3 critic @jack and his team at @TBD54566975 are out with their alternative vision for a decentralized internet: Web5.