Hacker News Reader

Comment by windex | Hacker News Reader

Time for more service decentralization.

ok. done. I've written you a decentralized social media platform. why haven't you come and started using it?

For all you know, windex may already be using Mastodon or some other service that's part of the Fediverse.

Your question should instead be aimed at the millions of people who still use Twitter, but of course you know the answer already: They use Twitter because millions of other people also use it, i.e. network effects.

That’s his point, I believe. The problem isn’t decentralization, the problem is how do you undermine network effects.

With great time, and care. Discord managed to do this in only 4 years, so it is possible. The cynical side of me says that a distributed social network, by definition, doesn't have the ability to run large marketing campaigns like Discord.

Yeah, a better question might be: "How can I convince a VC that there's a buck to be made when the user can just take all their data and leave?"

"Look, many countries are enforcing privacy rights; the ad model isn't sustainable. You should start a social media network with a different model so that you can be the first mover who benefits from network effects when the privacy shoe finally drops".

Something like that, although I don't delude myself into thinking that's coming in the near future unless perhaps Apple makes it part of its new privacy-centric campaign. Even then I'm not sure Apple would do better with respect to free speech and moderation.

It's a little different. Discord launched amidst a power vacuum, at least with respect to a chat/community platform for video games. Xfire was the only real one, but it had been on the decline for years. A better analogy would be a decentralized Discord competitor trying to start right now and with the odds it has going up against Discord.

Discord got traction because it was substantially better than Teamspeak, and Skype leaks IPs letting jerks DDoS you. It turning into something of a social network was very unexpected.

One way is to compel (via legislation) all social media networks to speak a decentralized protocol. Another is to make Mastodon, etc a lot better because every one I've tried has been pretty abysmal. In other words, make a product that is compelling enough to get people to move over to it.

One way is to compel (via legislation) all social media networks to speak a decentralized protocol.

This seems both feasible and practical. Sure, it will have problems - from the point of view of ___ platform regulars, it would be like an 'eternal September' of how everything got worse once those awful people from Other Platform were able to cross-post but that would probably spur innovation in client-side filtering.

I like Mastodon (apart from the me-too-ism of calling Mastodon posts 'toots') but agree that Fediverse offerings are mostly like desktop Linux or cheap fusion - they'll be really great when one of them takes off. The sad fact is that there are lots of designs that are as good as or somewhat better than the big platforms, but none of them are categorically different.

Consider the evolution of social media:

Newspapers: Anyone can write to editor

Telephones: Anyone can talk to anyone

Usenet: Anyone can post in a hierarchy of topics

Web: Anyone can have a webpage

Google: Anyone can navigate the web

Blog: Anyone can have an opinion column

Youtube: Anyone can be on TV

Twitter: Anyone can amplify their random thoughts

Facebook: Anyone can publish a yearbook

GitHub: Anyone can be a tech star

Instagram: Anyone can publish a magazine cover

Snapchat: Anyone can look like a celebrity

LinkedIn: Anyone can be a professional

Tinder: Anyone can be a porn star

Bitcoin: Anyone can be rich

Crypto: Anyone can be a central banker

Periscope: Anyone can be be a TV reporter

Tiktok: Anyone can be iconic

Fediverse: Anyone can run a social network

Obviously I'm being very tongue-in-cheek here but what makes a platform stand out is providing the ability to do something different that was not previously available, and either socializing that or making it stupid easy to do so - ie 'Love XYZing? Do it here and get noticed.'

If your offering is 'YYYYY for ZZZZZ' (where Z might be anything from 'hermits' to 'everybody') then you'll get some rapid adoption because it's easy to understand but it will plateau pretty fast. The secret of success is making people say 'No way, how does that work' and being able to say 'easy just Do The Thing and press this magic button.' The Thing has to be something whose benefit is immediately comprehensible but whose availability is extremely limited or outright fictional.

>90% of social proposals offer new and improved buttons, but experience suggests people are quite tolerant of inferior buttons (network quality, uptime, etc) as long as Doing The Thing is still novel; you work on the button in between getting your first and second competitors. If many people hate or don't get The Thing You Do that's actually a plus because all the misfits will flock to your platform and give you their cult-like loyalty.

I think the mistake with Fediverse stuff is the pitch that 'you can have total control/do it all' - if I wanted that I would have become a sysadmin, but the reality is that everyone wants to use email but hardly anyone wants to run their own SMTP server, much as most people want to drive fast cars and look good doing it, but relatively few choose to be mechanics or take up the risks of competitive driving.

You need to have a benevolent dictator/Santa Claus/Mommy/Team figure(s) that people can coalesce around, who gives you both a way to Do The Thing and permission to use it - it's 50% technical innovation but also 50% validation. People like having a person they can connect with both for technical requests/gripes and to serve as surrogate parents/friends. A socially oriented product/platform with no personalities attached is a contradiction in terms; it might be fun to play with but you probably won't tell your friends about it.

  A: I have feelings about this
  B: Me also
  A: The world should know of my feeling
  B: Indeed
  A: I shall now Do The Thing
  B: WHAT
  A: Yeah check it out
  B: WANT
  A: Sign up like so
  B: I am momentarily fascinated, who made this
  A: Bro and Sis, they are super cool
  B: How I do a thing...
  A: Like this, it is the Bro and Sis way
  B: I feel my life changing around me
  A: Now you are family
  B: Best friend would love this

If your pitch (to anyone) is 'Look at This Thing' or 'That Thing but different' or 'Have you tried The Other Thing' you are ultimately soliciting someone else's attention and the results will be proportionate to how people allocate their limited time. What you want instead is to improve an existing moment in some unexpected way, and then convey a sense of permanence or continuity in relation to the improvement. The witness to/beneficiary of the improvement will then be motivated to trace The Thing back to its source, to whom they will offer publicity/ suggestions/ technical help/ money or whatever.

To conclude, the breakout path for the Fediverse is not to solve this or that problem of social with a nicer button; when you offer solutions to a problem some may appreciate or even love it, but the appreciation often comes with a 'but...'. Rather, it is to build Fediverse technology into some novel object, and use federation to solve the (inevitable) issues that arise as the novelty gains in popularity.

Been there, helped build an open source decentralized product. Rhetorical demand higher than actual demand.

Of course it's a ridiculously difficult nut to crack. But people like the idea more than the reality, and it's hard to blame them when the product typically has some serious (and likely unavoidable) tradeoffs.

Nowadays people target multiple platforms in parallel, streaming to YT/rumble/odysse at the same time for example.

People go to these social medias because there is a large audience, how is decentralization going to help?

The only people who are using decentralized tools are the ones that have been kicked out of these big social medias at first place.

> "The only people who are using decentralized tools are the ones that have been kicked out of these big social medias at first place."

Do you honestly believe that? Some of us use decentralized tools because we remember the Internet before centralization and corporate/government control over it started turning it into a cesspool of advertising and propaganda and see the huge corporate properties (Facebook especially) for the societal poison that they are.

Thank you. I too miss the days when I could find niche opinions and in depth knowledge via a simple search.

Reply on news.ycombinator.com