What's the problem with the distributed social networks? If you don't want a single entity to be in control of all people's data, then distributed is the only way to go.

Of course the problem with the existing distributed social networks, like Diaspora, Friendica and all those others, is that there's not a lot of money behind them, so they're mostly amateur projects. And there's certainly a lot that can still be improved about them in terms of features.

What I personally don't like, is how you currently need different centralised sites and apps to follow various things an people. Just Facebook isn't enough; you also need Twitter. And Instagram. And Reddit. And then there's every possible blog.

I'd like a system that acts as my own portal to all those systems, and allows me to follow people on the Fediverse, Facebook, Instagram, RSS feeds, you name it; and is easy to expand to support new things, but also gives me more control over my feed; all of those social networks have a tendency to include all sorts of nonsense in my feed that I'm not interested in. More powerful filters would be nice.

And I wouldn't mind paying for that either, as long as it really has the capability to support nearly everything. The internet has grown to only support free, and therefore ad-monetised, possibly selling your data. But I somehow still pay EUR 22 per month for TV that I rarely watch. Something that helps me manage my feed well and puts me in control, should be worth something, shouldn't it?

Looking through this thread I realized that people have different problems with FB: - centralisation - privacy - it's too open - it's advertising driven (perhaps ties into privacy)

It seems to me that no one can define exactly how Facebook is evil or agree which is the thing they want to get rid of in Facebook. Without that it might be hard to change Facebook to the best or create an alternative.

Yeah, the rest of FAANG (plus Microsoft) have maybe one type of evil. Facebook has a broad range.

Apple: monopolist, flimsy products, not supportive of self maintenance, the iPhone battery thing

Amazon: exploits workers, doesn't pay taxes

Netflix: makes people lazy, doesn't hire interns

Google: privacy, can't escape it (browser, mobile OS, leading search engine, leading email), bad documentation, kills products that people love

Facebook: privacy, attacks every other tech company, no vision except growth, created zynga then dropped it after they got the user base, largest open plan office in the world, crappy ads, allows only the voice of politicians they like on the platform, charity is another platform for growth into developing countries.

Not everyone might find say, Netflix or Amazon evil, but FB's is such a broad range that someone will find it a little evil.

It’s not an active thread, whole books have been written about Facebook’s misdeeds.

On Facebook, you have to pay extra for the privilege of getting your posts in front of the people who have already said they want to see them, your subscribers.

I think early FB (maybe ~2007 iirc) was fine. You posted status updates, photos, and browsed your feed in chronological order with no silly prioritising of different posts or stuffing ads and videos between them. The ads were unobtrusive vertical banner ads on the rhs I think and I don’t recall them being targeted.

At the end of the day every tech platform that’s free needs to make money somehow and I don’t see how we escape a race to the bottom unless users pay some fee to avoid it. Unfortunately.

And we’ve now gone way beyond just “pay a fee and everything will be fine” because FB et al are now huge data brokers, they have lobbyists, they have morphed so much that changing the model wouldn’t fix it (imo). Maybe a new set of less evil startups can gain traction in a paid model but I don’t know how..

It would be a tiny scrapbook app with the following features:

A little interface for creating little "posts" of pictures and text.

A list of friends to sync new post id's with.

A little interface to browse new posts of friends, add likes and comments.

Additional equally trivial features for creating shared scrapbooks for for events, clubs, topics, businesses, etc.

Each person's list of friend/group ids and posts would be held in the cloud with end-to-end encryption. The encryption would mean complete privacy for individuals, with only permissioned sharing to others.

Given the needs of a series of small posts, this is a trivial amount of data.

The app could be free with a small hosting fee of $1/month or something.

There would be no ads. No surveillance. No massive corporation doing groundbreaking work in AI in order to surveil and serve ads, promote engagement, manipulate users, then making a show of screwing up half hearted attempts to censor bad stuff that it had actively promoted due to its engagement properties.

The app's authors and cloud providers would simply have the incentive to make sharing scrap book posts fun for its users.

Let's call it "The Scrapbook". Wait ... drop the "the". It's just "Scrapbook". It's clean.

I think there are a few key requirements:

1. End-to-end encryption, like Signal. Without this, there's nothing to prevent the new platform from turning into another Facebook. (Meet the new boss, same as the old boss.) But with E2E, the platform can't analyze the posts that you write or read, so they can't build a creepy tracking profile for you like some East German secret policeman. And because modern E2E protocols also include integrity protection (eg HMAC), the platform also cannot inject new posts into your feed.

2. This one is going to be unpopular, at least at first. But users need to pay. Because the platform needs to work for the users. Not the advertisers. Not the government. Not anybody else. On the bright side, they wouldn't have to pay very much. Even Facebook makes only maybe $100 per user/year, and they spend a ton on "R&D", and they still make a healthy profit. Their cost of goods sold is a lot lower.

3. Ideally the service should be run by a public benefit corporation, with the public benefit for its users clearly defined in its charter. That way, the company doesn't get to maximize revenue at the expense of users.

4. Some sort of decentralization or federation would be good. So if users are unhappy with one provider, they can easily switch to another one. Like with email. Tired of Gmail? No problem, just switch to Fastmail or Microsoft or whatever.

> Is it worth building something that isn't advertising driven to fill this niche? Is it possible for such a best to gain traction in the current environment?

I sure hope so. I've spent the past year or so building a prototype. Link is in my profile if you're interested.

Great ideas!

But if someone builds this please worry about point four down the road. The other stuff is good enough!

I constantly see people just building 4 and then it never goes anywhere and it’s too complicated.

Hey, thanks!

My app is called Circles, and it will be launching on the App Store next month.

It’s built on Matrix to take advantage of their support for E2E encryption, so #4 about federation won’t be too difficult. At least not technically. And I’ll happily federate with anyone who agrees not to do ads or tracking.

The company is currently a Delaware LLC but will be converting to a public benefit corporation as soon as we get the thing off the ground.

https://github.com/KombuchaPrivacy/circles-ios

I think what Facebook is fundamentally missing is interoperability. Imagine if Gmail users could only email other Gmail users. If Facebook allowed other services to be fully interoperable then competition and choice would solve many of the other problems.

Unfortunately Facebook is so dominant that it's not in their financial interest to allow that, so the only way I can see it happening is if the government steps in and forces them to.

I think the existing social networks are about as good as you are going to get. They are good enough for the majority, and the majority is strongly driven by price. A free service is going to win over any service you have to pay for.

Free leads to advertising. Advertising lead to algorithms tuned for engagement. Engagement leads to the garbage we have.

- Interoperability: FaceBook / Twitter and all other social network take their advantage from the network effect / the winner takes all, not because it is "good", "private", "secure", "beautiful", "user friendly", but because they have the critical mass. A decentralized social network must be like bittorent. You have many client, choose the one that suit you the best, and share data with other whatever their client.

- Add-free Business model: Big question in how do you monetize the system, making it viable. Maybe the people contribute in storage, storing the data of people they interact with to enable availability, and contribute depending of their possibility.

- Data ownership (a lot of concern for some tech people, but for the majority, I think they don't really care / understand the problem)

I’ll give this a try:

Focus on users photos and text posts.

Make it hard to share news and articles. Perhaps a critical thinking “captcha”. Maybe you have to write >50 words on why you’re sharing posted with the link? Maybe limit to two articles a week.

Perhaps your friends could have an anonymous way to indicate that the stuff you’re posting is politically offensive or just obnoxious.

And obviously no filtering. Everything is chronological. (User directed filtering of what they’re shown would be encouraged though.)

> lets us keep up with our family and friends, have productive discussions, share pictures, invite each other to parties

E-mail, with a better interface?

Seriously, this all can be done with E-mail and some locally stored data?

It is just a matter of implementing a UI (Web/Phone/Native whatever) and a messaging protocol, on top of existing Email infrastructure.

Delta Chat [1] is a good example of such an interface spanning all devices and platforms. Works suprisingly well!

[1] https://github.com/deltachat

Yes, found an old thread about this. It appears that the issues raised in that thread are not that hard to solve..

https://news.ycombinator.com/item?id=19216827

> have productive discussions, share pictures, invite each other to parties, and not make the world darker?

You are fighting against the very core of human nature. Some of us like to keep it nice and good, some of us just want to watch the world burn. No software is going to fix this.

> I think the distributed answers have failed, sadly, mostly due to barrier to entry.

Any new non-distributed social network would face an even higher barrier to entry. Without support for the ActivityPub protocol, the social network would not be interoperable with existing fediverse accounts, including Mastodon's nearly 1 million active users. Instead, you would start from zero.

Mastodon stats: https://fediverse.party/en/mastodon

If you want to create a new social network, supporting ActivityPub would be the strongest boost you could give it from day one.

https://www.w3.org/TR/activitypub/

No ads, same funding model as Wikipedia: non-profit, nagging for donations.

The platform needs to have zero motivation to keep the user on it: in my opinion it is the profit-motivated encouragement of addictive UX that is the root of Facebook’s evil.

A small community space supporting a couple hundred users at the most.

User accounts are owned and controlled by the users by means of a private key. (And thus the same identity or multiple linked identities can work across different "facebooks")

Entire social graph, data store, and all relationships and actions are public-ledgered, fully transparent, and available for download, archiving, and re-homing.

Moderation is also fully transparent, performed by volunteers who are part of the community on a rotating basis, and heavily aided by bayesian filtering.

Hardware and software maintenance costs are covered by voluntary payments and effort investment.

Is there anything I missed?

> Entire social graph, data store, and all relationships and actions are public-ledgered, fully transparent, and available for download, archiving, and re-homing.

This sounds like a privacy disaster. I wouldn't want data mining companies to find out about who I'm connected to. In my opinion, connections should be private to anyone but yourself (which is probably going to be hard to perform in a distributed manner).

You can make all those things private only to the extent that you trust each individual participant. The larger the group, the more likely a leak can happen. The other defense is limiting what you share with the network, and selectively encrypt.

There's a lot of benefit to be gained from sharing the social graph (or at least statistics based on it, which, however, can sometimes be reversed unexpectedly)

At the moment, we have the exact opposite of the situation you're asking for--anyone who's got a database and 10 bucks can see everything, while those for whom the data is the most relevant and potentially beneficial have very limited access... so even full transparency is an improvement, IMO.

The evil comes from optimizing the profits when all they have is data.

Technically, the only major cost of Facebook is storage space, so they could just function as Dropbox. Free plan up to a point, then charge for storage.

They could run some ads too, you know. Not targeted and not blended with content.

There are plenty of ways, so long as you don’t look for maximum profit.

Making it free maximises the number of users, and then AI-ing the shit out of user’s data and selling the results to gullible advertisers is definitely the ultimate money making machine...

I would personally paid a subscription. I actually did for a few years, to livejournal.com before Russians bought the company.

Unfortunately, I don't believe it's possible in the current environment. I already have a web hosting, would only paid if the hypothetical service would integrate with mainstream ones like FB/twitter. Otherwise it's nothing to pay for, social networks are useless without being able to communicate with other people.

If anyone is interested in getting together, brain storming and working towards a startup in this space - I am interested.

I particularly like "I'd like a system that acts as my own portal to all those systems, and allows me to follow people on the Fediverse, Facebook, Instagram, RSS feeds, you name it; and is easy to expand to support new things, but also gives me more control over my feed" by mcv here.

You can drop comments here or respond to me at sridhar AT toonclip DOT com

Well, you can't be everything for everyone.

IMO, the big problem with Facebook is the message-board style feed. I want my events in a separate app; and I don't want to get sucked into some online chat every time I need to open the app.

Same goes for groups.

Another problem is moderation. HN is awesome because the moderation is so good. Facebook lets hapespeach go all over the place.

If you try to tackle this, start small and grow organically.

A small monthly/yearly subscription; the vast majority of users can be served for a couple bucks a month. The outliers (those who upload significant amounts of media) can be made to pay more.

Alternatively, a platform where non-commercial usage is free but businesses or those who promote commercial products have to pay which subsidizes the free tier.

A nonprofit akin to the Wikipedia foundation running the platform funded by donations*. No advertising and no tracking. Facebook is mostly good (even tough I don't use it). It's funding it the rotten part.

*You could also have some tax scheme akin to a TV license to get public funding without public influence.

>I think the distributed answers have failed, sadly, mostly due to barrier to entry.

I don't understand this? I've been using Mastodon for a few years now, and the Mastodon network itself has been continually growing since it's founding with over 1 million monthly active users at present.

I think there are two issues: 1) Can you get your technically illiterate grandfather on there? 2) Is it easy to have a good experience without someone guiding you?

The distributed answers seem to fail both of those metrics?

I maintain notes about related social media questions and ideas that may help: https://github.com/joelparkerhenderson/social-network-plan

Perhaps we need a non-profit centralized social netwok. just like wikipedia. The goal of the organisation is not to maximize profits as long as it is self sustainable. All data is end to end encrypted.

It would look like the world before Google Reader was killed. We'd all still be blogging and reading RSS feeds in a mostly distributed fashion.

iMessages and its native Apple Photos and Shared Album features. No need for reshares, reactions are supported for messages, groups are supported, and something can't easily go viral (you have to make an effort to share the content outside of your iMessage group).

It's already the #1 social platform for teens.

HN with image support would be enough.

That's... just a subreddit.

Yeah was about to say reddit is good enough. The old one.

Slightly relevant xkcd https://xkcd.com/2481/

Like 4chan