[locopati]

'I would love to hear ideas for ways we can fix this that are realistic'

Peer-to-peer social with discovery. A system where companies/people/groups could self-host their own server, companies could host for others, but all the systems could talk to one another. Data gets stored where you want it to - on your own system or on another (though it's portable even there, by design).

[SoftwareMaven]

I agree that this is probably the right answer. Unfortunately, it's a difficult business model to pursue and open source may not be able to solve it.

I look at this like I look at Linux versus Windows/OS X (as I type on my MBA next to my work Linux laptop): Open source may provide an alternative that, for a certain class of people, is every bit as good as the commercial mainstream; but for Joe Sixpack and Molly Mall, using such a solution doesn't even factor into their consciousness. It's not that it isn't good enough; it's that it doesn't even exist on their plane of existence.

I've spent a reasonable amount of time thinking about this (as a side-thread from a previous start-up I was involved with), and I have yet to come up with a combination that satisfies my preferences for data privacy and ownership with "normal people's" preferences for ease of use.

Unless a catastrophic failure occurs (Facebook crashes and loses everything), I don't think people will realize there might be better solutions.

[intended]

You don't always need Joe and Molly sixpack. You do need the leading edge of adopters.

Some solutions aren't for everyone.

[dlitz]

You guys should touch base with the https://tent.io guys.

[oscargrouch]

I think im working right now in a technology that hopefully will attack this sort of problems..

i have fried my brains out , and i think i got a answer..

i think the root of all evil that is happening with the web is in its core design.. a common thinking that reduce reason to client-server (with the weak part bean the client and we know now who are the strong server part now) and where the protocol is on the visual entities.. not in the thing that really should be given the real attention.. data and information itself..

The answer was not easy to come through, cause "its not there".. but i think ive got a good point here.. and i expect to show it soon.. with a running code

[regularfry]

You just described email.

[locopati]

Or DNS. Or probably any number of other foundational systems of the internet. At the bottom of it all, the basic concept of communicating data across distributed systems is the internet.

This is where the original post is spot on - the new players are not interesting in distributing, they want to collect and retain, but not share, except in their own limited and controlled ways.

To break out of that you need to break away from centralization. Yes, I don't know how you make this is self-supporting business, but then again, I doubt the Wikipedia folks did either and they went forward (open platform, freely available data), not to mention the folks who wrote the SMTP, NNTP, FTP, HTTP, and every other RFC we have built our global communications on top of.

[regularfry]

DNS isn't social, though, whereas email is. That's my point. Of your other examples, only NNTP comes close.

[locopati]

In your view, what defines social?

I'm seeing this as one point receives updates from many other points and announces updates of its own. In the case of DNS, it's domain record updates. In the case of email, it's messages. In the case of social, status updates, photos, events.

[regularfry]

If the messages are between humans, that helps. You can't build a community over just DNS.

[yawn]

Are you interested in building this?