OpenSourceSocialNetworking

From KnowledgeLab
Jump to: navigation, search

This page is intended to gather information about the need for and the potential of an open source alternative to mainstream social networking sites like Facebook, Myspace, Beebo, Flickr, and so on.


Issues

Some of the key issues explained. Just sketched out at the moment.

Privacy

Not good to have a single company having access to all the personal information of millions of users.

Data ownership

With sites like Facebook, it is difficult to transfer your personal data (posts, photos etc) to another site if you decide you don't like them.

Distributed Networks

Better to have many cooperating smaller servers than a single monolithic server farm owned by one company.

Open protocols

Are needed to enable social data to be shared freely between many sites.

Critiques of proprietary social networking sites

Why Facebook is evil...

  • Guardian.co.uk article about Facebook. Among other things, it points out that one of the companies who recently put a large amount of money into facebook has links with the CIA.
  • Blog post by someone explaining why they have deleted their Myspace account.

Open Source Alternatives

We.Riseup.net

This uses its newly developed software Crabgrass - indicated as a secure system for activists to collaborate and network.

OpenId

This is a protocol which defines a way for users to log into many websites using a single username and password. It not in itself a social networking system, but provides an essential building block for designing these systems. Becoming pretty widely adopted. Your primary login can be kept on any one of many servers which have implemented the system, so it doesn't depend on any single organisation.

NoseRub

This is a protocol for storing user profiles and contacts in a decentralised way using a network of small servers, together with a web application that implements that protocol to aggregate social information from the various sites that each user belongs to. E.g. if you post a picture on Flickr, or add a bookmark on del.icio.us, this information will appear on your NoseRub home page, and in the 'Social Stream' page of your contacts. See here for a demo site using it. Its strengths seem to be:

  • it's first and foremost a distributed protocol not an application running on a single server, so has the potential to form the basis of a truly distributed social network, rather than a fragmented collection of smaller sites.
  • it uses existing protocols and technologies, such as http, OpenId, and FOAF.
  • It integrates well with existing sites that people already use.
  • The code is relatively mature.
  • It deals properly with friendship links that cross between different NoseRub servers.

Weaknesses:

  • Not much attention seems to have been paid (at the moment at least) to privacy and security issues. The only real choice you have in this respect at the moment is whether to have your social data on the front page of the server you're using or not.
  • By itself, it won't replace sites like facebook - just offer a different way for people to aggregate information from sites they already use into a single place.
  • It doesn't yet deal with more complex relationships between people such as work/interest groups as opposed to personal contacts.

Hello World

(English wiki - the main site is in German.) This is a similar project to NoseRub, but uses a much tighter and more clearly defined privacy/security model. I've only had a quick look at the specs pages, but from this, its strengths seem to be that:

  • They have really thought through issues around privacy, security, and data ownership. All content is held on peoples' personal machines not on servers, with the servers in the network only holding minimal identification data, like a phone book.
  • Their model for relationships between users, and for profile information about a user, is more sophisticated than NoseRub's.

Weaknesses:

  • As far as I can see, it doesn't use existing protocols in the way NoseRub does - they've tried to build something pretty much from the ground up.
  • In particular, it doesn't look like you can access data on this network using a normal web browser - you need to run the HelloWorld app.
  • Because of this, it doesn't fit easily with the way people use the internet at the moment, so doesn't seem likely to be widely adopted in the near future.

Elgg

This is an application for running a social networking website similar to Facebook. Strengths:

  • Seems to be well developed, and has already been taken up by external organisations such as universities.

Weaknesses:

  • There is no facility for enabling links between users on different servers.

DaisyChain

Open source social networking project. Not much information as yet.

Other listings of projects which are working in this direction

Proprietary Sites and Services

This is a list of proprietary sites and services which are moving in a similar direction to the ones described above. Know thy enemy!

Cliqset

A proprietary web service provider, offering similar capabilities to the ones being developed above, but with all the social data (as far as I can tell) held on a single server run by them.