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

I don't think decentralised P2P is really the answer anyway: the answer is separating the content from the convenience.

Facebook killed blogs because when I want to find my friend I just searched for their name, and if I was their friend their content made it to my wall.

Because of Facebook's convenience my friends use it, and because it's a walled garden I am compelled to use it to.

My dream for this kind of thing would be something akin to DNS but for identity (pseudo or not), and an open API for content.

So, Adam knows nothing about computers and creates an account on Facebook, which creates an IdentityDNS entry for him on his behalf.

Belinda is also not that techy, but likes G+'s layout more, and so has an account on there instead. She also has Twitter.

Zer0cool is a massive nerd, and has their own domain and hand-written blog software. They went to some website somewhere and created their IdentityDNS entry, and has hooked up and validated their blog on their domain as their content.

Adam can search for Belinda and Zer0cool in Facebook, which in turn searches IdentityDNS. He can add them to his Facebook wall, and their posts will show up in some fashion (maybe Facebook shows its own posts normally and only shows titles of external blog posts, FB can do what it wants). Belinda can do much the same with G+.

Zer0cool has also added Adam and Belinda, and reads their posts through muttdentity, a mutt clone but for identity stuff.

OR something. I don't know. Effectively something that means that I can choose FB because I like how it displays photos, or because it has the best mobile app, or because it has lots of storage. Or I can run my own blog because I want that control. ETC.

I shouldn't have to use Facebook just because that's all my Dad can work out how to use.




http://indieweb.org allows independent blogs to aggregate discussion from centralized social networks. RSS used to enable centralized social media to aggregate indie blogs, but is less used these days.

See recent discussion on these topics at http://www.decentralizedweb.net


> My dream for this kind of thing would be something akin to DNS but for identity (pseudo or not), and an open API for content.

SDSI/SPKI had an interesting mechanism for getting around Zooko's Triangle (before, I think, the latter was formalised): identities are keys (and therefore globally unique), with each user assigning his own names to people (e.g. 'Bill' or 'Ted'); users can then refer to other people as 'my wife's Uncle Bill' and 'Microsoft's Bill' and not confuse them.

This is decentralised and federalised to the max, which is awesome.

It'd be great for something like a Facebook killer, in which when I search for 'Adam Smith' I probably want the Adam Smith I went to gradeschool with, as opposed to the 18th century Scottish economist.


Hum, For my point of view Facebook didn't killed blog. Blog is used not only to show pictures of yourself, but writing. Nobody write on Facebook, they only share things. It's totally different, but hey anyway you can make your blog what you when that they are. Multiple account kill the service. You have at less 5 social media on every newspaper website. I don't use anymore twitter and I hope people will do it too. It's really a waste of time using 10 social media. I even don't use Facebook too.


You can write or blog on Facebook but few people choose too. Techies seems to flock to medium nowadays... So there is some kind of theming around hosts that decides where you publish. I can see why, in many cases Facebook can be too private to share in the professional world.


This sounds really awesome. The thing you are missing, I think, is the way Facebook lets you post things for just your friends (people you have approved to follow you) to see. That's a negotiation between the origin of the content and the consumer of the content that a simple DNS-like system doesn't solve.


The problem is, each 'viewer' application (Facebook, G+, Twitter, mutt, etc.) has an incentive to add proprietary features which only work on their system. Not only does this differentiate them from the competition, but it's also an easier problem to solve than if they tried to make it distributed (e.g. Facebook can add features to Facebook much more easily than they can to email, since they store all of the Facebook data and broker all of the Facebook messages; their only problems are how to make it appealing to users and cost-effective to run).

Over time, the walled gardens return.

In fact, we've basically seen this happen before: many people ran their own standalone Web sites on their own domain. The "viewer" was the user agent, which could be your choice of browser, aggregator, etc.


I think the killer feature of Facebook over standalone websites and blogs is that with Facebook you can choose who is allowed to view what you post. Maybe I'm wrong, but I remember people being a little uncomfortable posting pictures of their kids on a blog for the whole world to see. With Facebook, only your friends are going to see those.


> with Facebook you can choose who is allowed to view what you post

Well that's their kool aid, which many people have been drinking.

I still follow the rule that you only post stuff online that you're comfortable sharing with the world: it's not a matter of if the databases will be leaked, but when.


> Maybe I'm wrong, but I remember people being a little uncomfortable posting pictures of their kids on a blog for the whole world to see. With Facebook, only your friends are going to see those.

Use a blog software that allows to enable viewing of some content only to users that authenticate (for example via OpenID) and are of course in the list of allowed users (that you carefully maintain).


> Because of Facebook's convenience my friends use it, and because it's a walled garden I am compelled to use it[,] to[o].

Rather: Because it't a walled garden I am comelled not to use it (because otherwise I would violate the principles that I stand for).




Join us for AI Startup School this June 16-17 in San Francisco!

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

Search: