Diaspora Beyond its Growing Pains

20Sep10

Last week, the developers of “Diaspora”, an open source alternative to Facebook released the code of their application. Responses were mixed. Some found security issues, others critiqued the AGPL license, while many others merely expressed their excitement. Many people would like to see a project like Diaspora to be a success, if only because of its David and Goliath drama appeal. In this post I would like to try to look a bit further in time, and analyze Diaspora’s long term chances.

While the ideas behind Diaspora may have been lingering for a longer time, they gained momentum after a row over Facebook’s privacy policies (commentary by Dave Pell, Infographic). In response, four New York University students (New York Times profile) proposed to build an open source alternative that was inherently more privacy proof. Using crowd funding platform Kickstarter, they managed to raise $200.000 for the project, well above their $10.000 goal. Apparently there was a need to be met.

Diaspora borrows ideas from p2p file sharing and WordPress. Apart from being open source, the basic solution that the Diaspora founders propose for putting the control back to the user is to ‘decentralize’ the social network site. Each user sets up his own little diaspora server called a ‘seed’, which aggregates social status data collected from other networks. The user can decide how this data is shared across the diaspora network through ‘aspects’. This enables what I call ‘audience design’: sharing different streams with different groups. Moreover, since, as a user, I control my diaspora server I can take it down whenever I want to. As a figure of speech, I do not have to ask Facebook founder Mark Zuckerberg to get my data back.

There are obvious problems with this proposal. What has to happen when my server goes down? Will I disappear from my friend’s social graph? Do I have to hire space in the cloud, to make sure a regular back up of my social data is available? What is the point of building an open source alternative to Facebook, if it is reliant on social network data from, among others, Facebook and Twitter? While the developers found a clever way to solve the privacy problem with different technology, are they also capable to deliver the UX solutions? Can the Diaspora interface make privacy control understandable to naïve users? Whatever the anwers may be: the basic ideas behind Diaspora seem to be sound and promising.

So will the college four pull it off? At this stage it is not so bad that there are security issues, but it is important to deal with contributors that write patches in a good way. With the code out in the open the initial programming challenge changes to a community management challenge and this is quite a different game. In this sense Tomasz Stachewicz’s rant about the AGPL license, may seem insignificant, but is in fact highly relevant. Like the ‘normal’ GPL, the AGPL is viral. This means that derivative works from Diaspora need to be published under the same license. Indeed this may be too extreme. As AGPL is agressively non-commercial, commercial parties may unwilling to contribute to the broader Diaspora. This is quite a threat for building a sustainable developer community. Even ‘winner takes all’ competitor Facebook, does not work ‘alone’ (think Farmville). Choosing a licence is always political statement, and picking the AGPL makes is quite a strong one. But let us assume these are growing pains. What else?

If a reliable and safe version of Diaspora is build two interrelated challenges remain. The first is user adoption and the second is user experience (UX) design. As I wrote earlier, social media in general have a strong winner takes all nature. It makes no sense to adopt a social network site if your friends aren’t doing the same. There are three reasons why I believe Diaspora may become big in terms of user adoption. First, there is –even before its release – a strong core community which will drive its early adoption. I believe Diaspora has critical mass and this will grow if the network matures. Second, if Diaspora can aggregate data from other social networks it can grow without forcing people to switch or manage multiple networks at the same time. Third, I believe Facebook’s privacy policies will remain to cause the occasional stir. Mark Zuckerberg’s ideas about privacy have remained remarkably stable over the years and I believe people will not move much more in his direction.

However, the biggest difficulty Diaspora faces is to get the UX up to a level that is acceptable for people without a background in technology. Diaspora needs to become ‘simple and stupid’, and in particular privacy control is challenging in this sense. Moreover I believe it needs a competitive edge in functionality, which could be its ‘aspect’ system but this is a very difficult UX problem in itself. Facebook has a fairly good UX track record, while open source projects in general do not excel at this. They are certainly not known to deliver many innovative UX solutions. It is a matter of guessing why this is so. Possibly because the programmers, who drive open source projects, like ‘control’ over ‘simplicity’ or maybe good UX design needs a creative leap that is not easily crowdsourced. Still, if Diaspora aspires to become a ‘Facebook killer’ it needs to make sure to find a way to overcome this problem.

Advertisements


3 Responses to “Diaspora Beyond its Growing Pains”


  1. 1 Diaspora beyond its growing pains « @koenvanturnhout MacroBlog at The WordPress GPL Debate
  2. 2 Social News Needs a Nuanced ‘Like’, Quickly « @koenvanturnhout MacroBlog
  3. 3 A case for Privacy Coordination « @koenvanturnhout MacroBlog

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: