Hacker News new | ask | show | jobs
by chrischen2 5864 days ago
I think all this decentralized social networking is unnecessary. All we really need is an easy way to switch social networks. Something like a list of UUIDs as your friends list which can be imported into your social network of choice, and exported when you want to leave.

Even email is centralized.

The only problem would be getting the current industry leader, facebook, to willing adopt such a thing. They probably won't so this would be more of a consideration for the next generation of social networks.

2 comments

"Even email is centralized"?

I can start my own email server (or use one of millions of others) and send/receive to/from users on any other server... clearly decentralized (where is the center?)

I meant in practice. You could, but do you? Does anyone? What happens when your computer shuts down for the night?

Email is centralized in that individuals usually register with a provider, rather than use their own machines. I think social networking can adopt a similar model so all we really need is an easy way to pick up our social data and go to a new provider. I mean who I consider to be my friends is my data, so why not have a systematic definition of it that can be carried with me to every site, instead of recreating those connections using whatever interface is on each of the social sites I use.

It's still decentralized. Not one company has access to all e-mail, as with Facebook.
Of course it is, I'm not saying you're wrong. Although usually one company has access to all your email. As far as the user is concerned it's still centralized. That is, our data is still stored in a centralized place. The protocols for these open social networks however call for decentralizing all the way to the user level, and that's not how email is (though it could be).
I think you've picked a poor analogy with regards to email. It's actually the perfect analogy for a distributed social network. With email you can host your own, use your companies, use a free provider (hotmail, gmail, etc) or use your ISP.

It's not possible for one company/person to get across all your email because you can have it somewhere else. It's not possible to store Facebook info outside of facebook.

I imageine a distributed social network being a woven fabric where some people host their own, some use their companies (think bands using their labels server), a lot of people use free ad-based services (think hotmail or yahoo diaspora). However, no one company would ever have all your data, and couldn't hold it to ransom, assuming some type of export facility is built into the design. For I can import all my emails into a different server, but I can't export all my facebook data. Once switching costs are reduced to negligible, then competition for features, usability and, yes, privacy, wins out. Same way as plenty of people decamped hotmail and yahoo for a shiny new gmail account when they had better featurs and an import facility.

Sorry, what was my analogy? I'm saying distributed social networking should work like email and you just described it. However what I don't think it should be is decentrAlizing down to the user level. Another problem with standardizing the system is that new or special features will not be cross compatible, however that's another issue for another thread.
Email is centralized yes but its also somewhat open. You can send an email to any provider and receive from any provider. You cannot send a message from say facebook to twitter and get a response back.
Email is also pretty featureless. An open standard will always innovate slower than a proprietary site.
How about an open nonstandard? To me, there seems to be a lot of open-source innovation these days. To take one example out of dozens, where's the proprietary site that has a more useful proprietary programming language than Python or Perl? I agree with what I imagine your point to be: that standardizing Python up front would have made it very slow to innovate. But being decentralized hasn't made it innovate more slowly.

The issue with email is that innovations have to be adopted independently everywhere for them to be widely useful. But that doesn't have to be the case. You can structure a decentralized system so that any user of the system can deploy innovations on it, that the other users can immediately benefit from, without requiring them to be standardized.

Email is also pretty featureless.

  *Google Wave enters, stage left.*
Wave's stuff isn't part of the standard right? So if I were using some other mail service/client I could only use the intersection of both services. Please correct me if I am wrong, I haven't had much exposure to wave (except a brief stint in the private beta where I realized nobody is using it, and therefore it was useless).