Personally I'd just like to see _any_ kind of federated social network system that gets some market share. Kind of like how email is federated. Doesn't really matter to me the feature set. My thought is, if one could get to some kind of adoption, folks might see how much better things could be. You'd get to select your own "onramp" to the social network, follow people the same way you do on email (I'm following sfblah@hackernews or whatever), etc.
My personal wishlist for a decentralized social network
181–190 of 271 posts
Re: My personal wishlist for a decentralized social network
#182Looks like there's a typo in the link for Urbit, should be https://urbit.org if anyone wants to take a look. Personally it seems like vaporware to me, there's not much based on the lofty promises of a new OS/language/social network. Probably the only reason anyone has even heard of it is because it's the pet project of Curtis Yarvin aka Mencius Moldbug, known (somewhat infamously, a few of his talks at conferences ha…
> Personally it seems like vaporware to me, there's not much based on the lofty promises of a new OS/language/social network.
> not much based on the lofty promises of a new OS/language/social network.
You can literally set an urbit up and join a not-dead distributed social network right now, this second, in probably less than 3 minutes — it's explicitly not vaporware if people are using it to build thriving communities.
Re: My personal wishlist for a decentralized social network
#183Personally I'd just like to see _any_ kind of federated social network system that gets some market share. Kind of like how email is federated. Doesn't really matter to me the feature set. My thought is, if one could get to some kind of adoption, folks might see how much better things could be. You'd get to select your own "onramp" to the social network, follow people the same way you do on email (I'm following sfbla…
If you don't require a feature set, wouldn't the Internet or RSS feeds already qualify?
What does it need to have on top of RSS feeds?
Re: My personal wishlist for a decentralized social network
#184My personal wish list:
* Privacy. This means actual privacy and not some bullshit substitution, such as anonymity. If privacy were really important nothing would ever be anonymous, because you would always want to know who you are speaking to or otherwise deny access to the conversation.
* Portability. Decentralization is too preoccupied with fixed addressing. If I move my laptop from one building to another the network has either failed to resolve my device or considers it a different/new user. You can have both portability and decentralization.
* Services. There is so much more to a valid social experience than texts, which is 70s era computation.
* Control. End users should have the ability to deny any output, or else this entire conversation about decentralization is moot.
* Turing Complete. The application should be able to build an instance of itself that can then perform an end-to-end self-assessment via test automation upon that instance.
Re: My personal wishlist for a decentralized social network
#185Earlier quoted context omitted.
And there has never been the incentive. People complain about walled gardens, but unless you dedicated to FOSS, any commercial venture (with a few exceptions) produces apps and tools that feed the master and excludes other parties. It might even be profitable to start a venture that allows complete easy self hosting of content. Wasn’t Berners-Lee working on something like this with his pods?
> And there has never been the incentive. For FAANG. Instead of improving standards such as XMPP and RSS, they actively ditched them to create walled gardens..
Yeah, absolutely agree.
Re: My personal wishlist for a decentralized social network
#186It's absolutely staggering to me that people believe everyone will self-host the decentralized social network, without thinking through the economics of the incentive layer, let alone how to achieve consensus/verification on compute and storage layers. The first successful decentralized social network will be using the decentralized compute and storage systems being developed in the crypto space.
You're right, if it weren't prohibitively expensive to store social media data on Ethereum (for example) then that's exactly what all the decentralized developers (including myself) would already be doing.
I checked the Gas price on ETH and quickly concluded that's not even remotely sensible to use it for social media networks, and the bigger problem is that if something became hugely popular that could drive up ETH gas prices too, so building on that infrastructure seems risky from a cost perspective.
Re: My personal wishlist for a decentralized social network
#187Aether (https://getaether.net) has some of these features:
- Distributed content delivery. Every Aether node hosts the content. There is no centralized server hosting the data.
- Customizable moderation. You're free to ignore a moderator that who is being unfair and you don't see the results of their moderation. What they delete is only hidden to people who trust the moderator.
Re: My personal wishlist for a decentralized social network
#188Looks like there's a typo in the link for Urbit, should be https://urbit.org if anyone wants to take a look. Personally it seems like vaporware to me, there's not much based on the lofty promises of a new OS/language/social network. Probably the only reason anyone has even heard of it is because it's the pet project of Curtis Yarvin aka Mencius Moldbug, known (somewhat infamously, a few of his talks at conferences ha…
> Personally it seems like vaporware to me, there's not much based on the lofty promises of a new OS/language/social network. > not much based on the lofty promises of a new OS/language/social network. You can literally set an urbit up and join a not-dead distributed social network right now, this second, in probably less than 3 minutes — it's explicitly not vaporware if people are using it to build thriving communit…
Where are you getting an Urbit ID in 3 minutes? I am not super familiar with Urbit. But my understanding is that even getting a temporary Urbit ID (a commet) takes hours.
Re: My personal wishlist for a decentralized social network
#189What I would prefer is a set of protocols that are interact in a well defined way with each others, like the Email or Usenet protocols, then you can write different clients for these and decentralisation is just a matter of running different servers clients that don't necessarily interact with each other. Something like the CapeProto RPC framework makes it relatively easy to declare the interface of a service.
I'm curious if Skynet or something similar could be used for that? I don't pretend to understand any of the modern blockchain frameworks, but a peer-to-peer distributed and encrypted system seems to fit the bill (possibly non-blockchain related, I'm not sure). It was trending yesterday: https://news.ycombinator.com/item?id=25710151
SkyFeed is a decentralized social network built on top of Skynet: https://skyfeed.hns.siasky.net
Re: My personal wishlist for a decentralized social network
#190My big wish, I want to truely own my own data. If I decide to delete something, I would hope there's not a hundred copies replicated in a hundred caches somewhere. When I delete my own data, I want it truely gone. If someones node is offline, I hope there's nothing of mine left on it.
This is difficult to balance with having data be available when you are offline.
If your data is local to a user's device (laptop, desktop, tablet, phone), then it's not online 24/7. So, you need some way for other people to see it.
If you require people to use a server, there's either problems onboarding and trusting that server operator. Or you make people host their own server and nobody uses it b/c that's too technical.