Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Blueprint of a distributed social network on IPFS, and its problems

beyermatthias.de

Post reply on HN
Print view View original

21–30 of 96 posts

Re: Blueprint of a distributed social network on IPFS, and its problems

#21

Post by fake-name »

hodl wrote: post #15
fake-name wrote: post #12

If you're thinking about building a distributed social network, you HAVE to have answers to the question: How do you moderate it? (Or how do the users moderate what they consume, if you will) Basically, think of the most annoying person you can imagine, give them infinite free time, and a uncontrollable urge to do EVERYTHING they can think of to make your life miserable on said social network (doxxing, abusive messag…

Users mod their feeds as per RSS you just unsubscribe from idiots. Its a bit like the internet itself. Lots of shoddy sites, but no one needs to visit them. Search spam is the exception but let a Google handle that.

That would be fine, but a bunch of RSS feeds is in no way a social network.

A group of people who all subscribe to each other could be a minimalist network, but how would things like conversations work?

Re: Blueprint of a distributed social network on IPFS, and its problems

#22

Post by rolleiflex »

It seems there's a lot of people working on this area, just looking at this thread. I did something similar to this in 2013 and I got to around 20k users: https://www.theverge.com/2013/11/27/5150758/aether-aims-to-b...

Two things that need to be solved:

a) The structure of moderation. No moderation does not work. Been there, done that, it gets very ugly very fast. But what can be done is to make moderation more visible, and allow people to choose who they want to be moderated by.

b) Scaling. 1-100 users are fun, 100-1000 are exciting, 1000-10000 are a little nerve racking, and after 15k users, you notice the reality of how nodes and networks behave has diverged from your 'reasonable' assumptions. Around 20000 or so, I stopped maintaining it. Scaling distributed networks is much different than scaling regular networks where you can add more hardware and servers and sharding, it requires new behaviours from each node as the system grows. The software that works at the level of a million users in a decentralised system will be much different than the one that works with a billion users.

Funny enough, I've recently quit my day job and I'm focusing on the second version of it, so the memories from 2013 are pretty fresh.

Re: Blueprint of a distributed social network on IPFS, and its problems

#23

Post by Gibbon1 »

fake-name wrote: post #12

If you're thinking about building a distributed social network, you HAVE to have answers to the question: How do you moderate it? (Or how do the users moderate what they consume, if you will) Basically, think of the most annoying person you can imagine, give them infinite free time, and a uncontrollable urge to do EVERYTHING they can think of to make your life miserable on said social network (doxxing, abusive messag…

> Moderation and control have to be designed into the network from the outset.

I wonder about the ability to set up privacy canaries to detect information leaks. Because with stalkers and the like what you are looking for is to make sure that they don't have side channel access to you. So it's not just being able to block them, but block people they associate with. And importantly have ways to detect when they've gotten shared info they aren't supposed to.

Re: Blueprint of a distributed social network on IPFS, and its problems

#24

Post by wjnc »

rolleiflex wrote: post #22

It seems there's a lot of people working on this area, just looking at this thread. I did something similar to this in 2013 and I got to around 20k users: https://www.theverge.com/2013/11/27/5150758/aether-aims-to-b... Two things that need to be solved: a) The structure of moderation. No moderation does not work. Been there, done that, it gets very ugly very fast. But what can be done is to make moderation more visib…

Robert Nozicks' "Anarchy, State and Utopia" (1974) describes how a system of multiple concurring "governments" could work. Quite a stretch from social networks but first thing that came to my mind when reading your comment. Moderation is a form of social network government, right? One could subscribe to a government / moderation scheme and the governments form meta-governments to decide which content from competing governments are fit for their users. Moderation in two (or N) levels.

Re: Blueprint of a distributed social network on IPFS, and its problems

#25

Post by rolleiflex »

wjnc wrote: post #24
rolleiflex wrote: post #22

It seems there's a lot of people working on this area, just looking at this thread. I did something similar to this in 2013 and I got to around 20k users: https://www.theverge.com/2013/11/27/5150758/aether-aims-to-b... Two things that need to be solved: a) The structure of moderation. No moderation does not work. Been there, done that, it gets very ugly very fast. But what can be done is to make moderation more visib…

Robert Nozicks' "Anarchy, State and Utopia" (1974) describes how a system of multiple concurring "governments" could work. Quite a stretch from social networks but first thing that came to my mind when reading your comment. Moderation is a form of social network government, right? One could subscribe to a government / moderation scheme and the governments form meta-governments to decide which content from competing g…

Interesting! I'm actually doing something pretty close to what he is proposing with v2. Meta-'governments' layered on top of each other.

Re: Blueprint of a distributed social network on IPFS, and its problems

#26

Post by throwaway2041 »

evolve2k wrote: post #18
throwaway127831 wrote: post #7

Earlier quoted context omitted.

Because not all of my friends and family are willing to pay a subscription fee :-/

Would you pay to be a power user who can invite and manage say a dozen family & friends? In this model it might be possible to assist uptake by having a class of power user who can manage their group over onto an ad free network. Example: I have a private github account with repos I share with various friends who are not paying github users but are nonetheless now engaged in the github platform.

That's not a terrible idea. You might also give some users free access if they're net-beneficial to the network

Re: Blueprint of a distributed social network on IPFS, and its problems

#27

Post by ianopolous »

People might be interested in the social network on IPFS a few friends and I are working on in our spare time, Peergos [1] [2]. It also happens to be the first social network on IPFS (We started before IPFS itself existed).

Everything is encrypted, including metadata and we try not to expose your follow list to the network. It's inherently multi-device, independent of DNS and even has streaming E2E encrypted videos playable in the browser (streaming is disabled by default at the moment because it only works in Chrome) and even hides file metadata like size, name and directory structure. It's still pre-alpha though.

[1] https://peergos.github.io/book

[2] https://github.com/Peergos/Peergos

Re: Blueprint of a distributed social network on IPFS, and its problems

#28

Post by RSSify »

fake-name wrote: post #21
hodl wrote: post #15

Earlier quoted context omitted.

Users mod their feeds as per RSS you just unsubscribe from idiots. Its a bit like the internet itself. Lots of shoddy sites, but no one needs to visit them. Search spam is the exception but let a Google handle that.

That would be fine, but a bunch of RSS feeds is in no way a social network. A group of people who all subscribe to each other could be a minimalist network, but how would things like conversations work?

RSS feeds are enough for a social network if you only care for conversations between people you follow.

If you want to support conversations with people you don't follow then you need to add support to your RSS feed for comments or pingbacks.

Re: Blueprint of a distributed social network on IPFS, and its problems

#29

Post by hateful »

I've been keeping an eye on peer-to-peer/federated social networks for a while. And though I left Facebook a couple of months ago, none of these can replace it until they do one simple thing: Allow one user to post to another users "feed". This is what Google+ was missing when it came out. This is what Twitter is missing. This is the reason why Facebook is so popular, IMO.

I see something on the Internets which I think my wife would like, but I also want her/our friends to see it and I post it on her wall/timeline. I don't want to post it on my feed, that won't help, and I don't want to private message it to her, that also won't help.

Any of these social networks add this feature - then I will champion them to the end of the Internet.

And perhaps add support for events. But that can be an adjacent but compatible application.

Re: Blueprint of a distributed social network on IPFS, and its problems

#30

Post by pjc50 »

RSSify wrote: post #28
fake-name wrote: post #21

Earlier quoted context omitted.

That would be fine, but a bunch of RSS feeds is in no way a social network. A group of people who all subscribe to each other could be a minimalist network, but how would things like conversations work?

RSS feeds are enough for a social network if you only care for conversations between people you follow. If you want to support conversations with people you don't follow then you need to add support to your RSS feed for comments or pingbacks.

.. and as soon as you allow conversations with people you don't already know all the badness turns up and you have to build an anti-spam and anti-abuse system.
Post reply on HN