Hacker News new | past | comments | ask | show | jobs | submit login

This is cool - I like reading about these sorts of open platforms having success. It always reminds me of a thought I've had bopping around in the back of my head for awhile: do we really need to develop new technology to emulate what social networks have become? It seems to me that the core of social networks are: (a) a profile/posts/wall that is a glorified RSS feed, (b) a home/feed/stream that is a glorified (and proprietary) RSS reader, and (c) a (proprietary) messaging system that could be filled by email or XMPP. (in fact, your facebook messenger IS an email account[0].)

So do we really need new technologies like GNU Social/Twister to accomplish this, or can we just repackage the above tools to look like facebook/twitter/g+/ello, and have decentralized social platforms. The biggest hurdle I can imagine is non technical users not knowing what in the hell to do with an XML page when trying to follow a person, so I'd propose that a new link type similar to <a href=emailto:address> be established -- perhaps something like <a href=follow:myfeed.xml> -- which would open the feed in the follower's default RSS reader. And I could sort my feed into circles/groups, which are really just folders in a feed. I think this could be prettied up to look the same as and be just about as easy as twitter, with no new technologies needed. A service could host each of these components, and thus look just like a full social network as we've come to know them.

The fact that it's so often referred to as "microblogging" makes me wonder why this sort of set up isn't discussed more often (unless I'm missing it). It would also then be trivial to host your own, because really all you'd be doing is hosting a blog, and using an RSS reader. It seems so obvious to me that I feel like I might be overlooking something.

[0] https://www.facebook.com/help/224049364288051




There are two pieces missing from what you describe: one is ease of deployment, and within that I can lump operating costs. This is the less difficult problem to overcome.

The second, much more important problem: discovery. This is the only user-focused service Twitter actually provides (the ability to find the accounts you want to follow). The 'network effects' people love to throw out as excuses for using shitty services like Facebook is not as big a deal as people around here pretend. Email works just fine and there's no 'network effect' preventing me from running my own mail server. Discovery is so important that just about everyone reading this post would rather have a lock on discovery (and the accordant ad revenue) than promote any kind of federated system. Discovery is search, and search is hard, and searching across a federated service is damn-near impossible, which is why there's no email whitepages any more. Much simpler (and more profitable) to hash everyone's email addresses, or let them pick a cute username -- whatever, so long as you hold the keys and can sell ad space on the results pages.

Nothing like what you're describing is possible any more. There are no more federated open standards; the IETF is merely propped up to rubberstamp whatever Google or Microsoft wants to do next.


Well, it's the same reason have both chose to write our comments here instead of self-hosting it on our blogs, I suppose.

I would like to add the spam problem to your explanation. A distributed system needs a distributed solution for spam, which is much harder.

But we also need to add a little complexity to the description. There is a difference between decentralized and distributed systems, with email the former and usenet or irc the latter.

Both discoverability and search are trivial with distributed systems, but also offers the possibility of competing networks. It is interesting to note that the open protocols actually faired worse in those cases, as it lends itself to competititon, and a someone with access to deep pockets can leverage that against you.


There are several ideas on how to simulate social network experience with classical w3c tech. Linkeddata group (one with Tim Berners-Lee) work on SoLiD - "for developers who plan to build social linked data servers and applications". As you said, there is plenty of ways to make online social experience with original WWW methods, and it'll be distributed, and ad free, you just need to make a "pretty wrapper" for common folks.

If more developers acknowledged possibilities of semantic web (or "Linked Data" approach as it been re-branded) not just for reducing entropy of the web, but also to provide solid services for ordinary users, it wouldn't be so hard to develop critical mass to be widely used.

[0] https://github.com/linkeddata/SoLiD


I'm not an expert, but doesn't Diaspora roughly offer these services? Even the search mentioned below seems to work reasonably (although maybe they're cheating somehow and it's still a little bit centralised -- again, i'm not clued up).




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: