[Nextgrid]

One big issue is that there’s nothing decentralised that currently exists that can rival the quality & user experience of mainstream social networks, and decentralisation comes with its own problems (I personally think the problem with mainstream social media is its ad-based business model and not centralisation).

Mastodon (which seems to be the biggest alternative being proposed) is still a joke, even the name and branding sounds awful IMO. And who in their right mind thought calling a post a “toot” (https://en.m.wiktionary.org/wiki/toot) was a good idea.

Besides the branding, decentralisation comes with its own issues like the lack of network-wide content moderation and agreement on what content is acceptable. There are solutions (more like hacks) around this where instance admins can choose not to federate with instances they don’t like the policies of, but it then causes problems for end-users where they can’t communicate with their peers on those banned instances despite all of them being on Mastodon. Good luck explaining to a non-technical person why they can’t talk to/see the posts of certain people despite them all being on Mastodon, and the solution is to spend time choosing an instance with policies you agree with and making sure your friends are on it or on a similar instance that’s not banned by yours, and then hoping the instances stay online without any kind of funding (there’s also no knowledge of whether they would scale to the size of mainstream social networks).

The solution IMO is not Mastodon or any of these fringe social networks. The main problem is the lack of an ethical business model in mainstream social media. The solution would be to vote with your wallets and fund a better Facebook alternative - it could even show the current social networks that there’s profit to be made treating their users with respect and make the situation better for everyone else too.

[r3bl]

I agree with you all the way until the middle of your third paragraph.

You're not gonna suggest Mastodon to someone, you're gonna point to a specific community (probably the same one you're a member of). Only one set of rules you need to worry about. Federation? You don't have to pay attention to it at all. It's a nice feature to have for sure, but it only becomes relevant once you don't have people to follow inside of your own instance. By recommending an instance, you're recommending a community, not the software behind that community.

There have been plenty of attempts of taking Facebook's crown (both VC-funded and user-funded), and they've all failed spectacularly. The reason for that is simple: people don't want a global network. Facebook was the first and last one to succeed. Nobody wants to be on the same network as their parents, so they indeed decentralize: they decentralize in group chats, Facebook groups, Discord servers, Slack servers, Twitter communities, Discourse instances, Mastodon instances, forums like HN, subreddits etc. Facebook and Google+ failed immediately simply by having a real-name policy. That's okay if you want to communicate with people around you, but terrible if you want to truly express yourself to a bunch of strangers. The younger you are, the bigger the odds that you belong in the latter. Nothing wrong with communicating with people around you, but that's not the group that drives your numbers up drastically.

Mastodon surely can't be the new Facebook simply because that's not what it aims to become. It aims to become the software of choice for the communities. The easier you make it to jump on board (and the less personal data you need to provide in order to do so), the bigger the odds that you'll be the home for a community.

[Nextgrid]

I am not sure that being a global network is a problem.

Instagram is a global network and seems to be doing fine (although the quality of the content has now declined).

Personally, a global network is what I want. I already have the solutions you mention (group chats, Slack/Discord instances, forums, etc) for specific communities. What’s missing is something like Facebook or Instagram where everyone is on it and I can just “add” them and get updates about them every so often.

If anything, the per-community problem is already solved thanks to Discourse, Slack/Discord, Reddit, group chats, etc. But a global network is what’s missing.

[r3bl]

You don't jump from nothing to a global network. You host tangentially-related communities in the middle. The more of them you host, the bigger your overall numbers are.

Instagram succeeded for that same reason: profiles set to private, no real name policy, people can't look you up in a search bar. It was easier to group up in small communities. The less that's the case, the crappier the content. Instagram was ruined the moment Facebook accounts were attached to Instagram accounts — it's just dying slowly, the same way Facebook is dying slowly.

The next "global" platform is going to be Discord. It started as a place to host gaming communities. People were subscribed to a few gaming communities, so it already made sense for them to join more communities that are available on the platform. Right now, it's no longer the place exclusively for gaming. Every subreddit has one, every Patreon supporter is a member of some secret one. It'll outlast both Instagram and Facebook for one reason only: no personal info what so ever. People can't find out anything about you by clicking on your username: not your real name, not your contact info, not even a list of other communities you are a part of. You join a community by being invited to one.

[themacguffinman]

On the contrary, I think you don't jump from nothing to tangentially-related communities. The reason why niche community subreddits are so successful is because reddit itself is a global, central network that congregates users all on its own with its popular default subs. You have it the wrong way round: niche communities benefit from a global platform that funnels users to them, niche subreddits do not drive the success of the global platform in the first place (although it may enhance it).

Discord and Mastodon could never be as successful as subreddits for this reason. It's pretty difficult to establish a niche community by setting up shop next to other niche communities, niches are full of passionate people who don't spread their passion across niches thinly. You want a global platform like reddit to expose your community to the masses to unlock the niches within it.

[jammygit]

Anecdotally, most of my irl friends don’t or else barely use Facebook at this point. Lots of profiles with no pictures posted in 5 years

[jakear]

To me, Discord is just a partial implementation of Slack. What does it have that Slack doesn’t?

[r3bl]

Feature-wise, pretty much nothing, it's just targeting a different set of users. Slack started as a solution for companies and pretty much stayed there, Discord started as a solution for gamers and expanded from there.

Slack is Discord for adults, but adults are never the ones who make or break social media. It's younger people who tend to be more invested in the platform they use. It's not a rational choice (nor was Facebook for my generation), but it got kickstarted out of necessity (as a substitute to the limited chat options within games) and right now, it's the convenience that drives it further. Why switch over when, unlike Facebook or Slack, everyone you know is already using Discord?

[AJ007]

So basically what we need is Discord/Slack, but as an open protocol.

The entire concept of having a gigantic centralized social network on top of a gigantic decentralized communication platform (the internet) seems like a bit of an anti-pattern to me. Email, Usenet, and IRC seem to have never gotten the successors they deserved. Centralized, closed, and operates by a for-profit organization does not count.

As a side note I’m surprised there isn’t more mention of message boards here. Their peak seems to have been 2001-2010 and then their importance faded. One can discuss the shortcomings of phpBB and vBulletin at length, but their basic function seemed to have worked extremely well.

[doute]

> The solution IMO is not Mastodon or any of these fringe social networks. The main problem is the lack of an ethical business model in mainstream social media.

That is certainly a problem and something should be addressed. But I don't think it is the root cause. The real tragedy is the lack of understanding that reality happens between the mainstream and the fringe.

Most of the technology we like was created in the second half of the 20th century. We have been standing on the shoulders of giant and eventually the giants got old and were replaced by self-serving large tech companies. But since most people are making money, or get things for free, we don't want to recognize that we are their servants. There is no longer any urgency to create something different, because being different means missing out.

The reason you have to go to esoteric solutions when talking about something like decentralization is because the Internet is no longer built for it. From authentication, to networks and even the state of ip addresses is less than great. That you can avoid these problem by going to the fringe likely comes from the idea that hackers still have influence. We think that if something doesn't work it can't be our own fault, it most be some conspiracy or inherent limitation, rather a lack in our own understanding and ability to organize.

> Besides the branding, decentralisation comes with its own issues like the lack of network-wide content moderation and agreement on what content is acceptable.

Decentralization built on individualism rarely works. Because it leaves the unorganized powerless. It has to be built upon common features used by different cliques were the participants have choice. Maybe most importantly there has to be separation between the platform and the activity.

[Nextgrid]

> because the Internet is no longer built for it

I’d say the Internet is no longer optimised for it because the use-case is no longer possible for the majority of people.

You can’t run a server on a phone due to power constraints and yet more and more people are using phones as their only computing device.

[doute]

That is part of it. But the important point is that it is not inherent. There nothing saying that you can't be protocols for that. It is just beyond what the Internet was design for many years ago, and the de facto Internet today include proprietary infrastructure run by large tech companies. But it is generally even worse then that because today smart phone don't have external ip addresses so you can't connect to them even after the initial centralized wake up. So it is an absolute regression as well.

[Nextgrid]

Mobile networks are moving towards IPv6 now, so phones are uniquely addressable again. Making them reachable from outside (if they aren’t already) is a trivial problem to solve should there be demand for it.

But I don’t see any major use-case for that - the main issue of battery life remains for any significant usage, and frankly I wouldn’t want to have something running in the background that would deplete my battery in an hour because someone happened to connect and watch a video hosted on it.

[EGreg]

We are building it.

https://qbix.com

https://qbix.com/blog

https://qbix.com/platform

How do you suggest we have people fund our project with their wallets?

Some links to videos of things you can build:

Videoconferencing https://youtu.be/QDqX7EN7ci4

Chat https://youtu.be/QJg3ZwKalmU

People https://youtu.be/ZRfOKuacdqI

Events https://youtu.be/RTFcFGZeCsw

Rides https://youtu.be/QDqX7EN7ci4

Payments: https://youtu.be/Z7Q7IzVv1VU

HTML: https://youtu.be/Z7Q7IzVv1VU

[kickscondor]

> One big issue is that there’s nothing decentralised that currently exists that can rival the quality & user experience of mainstream social networks.

The Web at large is the decentralized network. And, anyway, part of the reason the ‘quality’ isn’t there (and this is arguable, I wouldn’t call social media a ‘quality’ experience) is due to the fact that we are spending our time and resources there. If that time was spent on the Web, it would improve.