Udo's Techblog

With much fanfare, Google announced OpenSocial, a set of API specs designed to be the new standard when it comes to the development of social networking sites and the gadgets that extend those sites. Because Google is the market leader when it comes to web apps, there was a huge window of opportunity to define how social sites can work for the next years. Sadly, there is so much conflict of interest for Google and its other partners, who all have their own social networking sites. And this is the reason, why OpenSocial ended up just being a way to write Gadgets for Orkut and its sisters when it could have been so much more.

Here is a quick run-down of issues OpenSocial could have solved easily:

Market Fragmentation
The social market is getting crowded with sites. Most people are not registered with one site exclusively, they maintain several profiles at several sites. None of them are connected. Profiles and friend lists have to be updated constantly, and maintaining this information at several places simultaneously is just not working out. The result: people tend to have dozens of accounts everywhere, and none of them are up-to-date. Accounts (and sites) are abandoned because the hassle is just too much.

Who Owns My Data? Who Stores My Data?
Currently, every social network keeps all of your data in its own store. That means, you don’t get to see what they’re actually storing, you don’t get to export and re-use your data, and when the site goes out of business everything is just lost. You end up entering your information completely anew with every new site you register an account at. Sites are actively preventing you from using your own data as you see fit.

Friends, Interests, Activities
A particularly nasty aspect of the fragmented market combined with no control over your data is the total absence of any convergence when it comes to maintain your relationships and affiliations. You have several completely unrelated networks, none of which contains the whole truth. Of course the time investment alone keeps users coming back to the individual sites, but please bear in mind that this repetitive maintenance is just wasted time. Maybe it’s time site owners looked a little more closely at the activities users are spending time on. How much time is spent performing repetitive, frustrating maintenance on your social network? And in contrast how much time is spent, you know, actually living your friendships and affiliations?

Persistence
We tend to invest huge amounts of time and effort into maintaining these giant datasets. We create personal history, original content, events, message feeds. Would we do that if we thought about how persistent this data is, if we knew that all those hours can (and do) go down the drain in a few months or years? Sure, lots of your stuff is only needed temporarily. But we have shifted so much from a world of permanency (books, letters, diaries) to a world where everything as a half life of only weeks. Maybe we don’t realize it explicitly, but don’t you think it would be nice to save a little bit of personal history? And if you’re not into that, at least you can understand the need to preserve all the original content you created? Or interactions with people you love? A good data model social networking could easily incorporate archiving mechanisms where you as a user can decide what should be kept, where it should be stored, and for how long it should be available.

A Federated Model
Obviously, some kind of cooperative network model would accomplish all of this, but don’t expect the large providers to support anything like that. The short term benefits of lock-in are just to great for them to ignore. That’s why independent organizations have to step up. Or maybe cornered competitors – yes, that’s you there, Facebook! For an example how federated data models can work, look at the Jabber/XMPP protocol.

A federated data model would enable users to concentrate on interacting with each other instead of spending so much time building all these separate networks that only get abandoned soon after. Likewise, such a set of protocols would enable sites to concentrate on providing interfaces for users to interact without having to worry too much about lock-in and drawing new users. A federation would also mean lowering the critical mass for small new sites, since they would be instantly connected to the collective.

It’s time to pull all of this together. It is time for users to actually profit from their data. It is time for site owners to stop reinventing the wheel. It is time to stop wasting so much time. A decentralized data federation is not only technically feasible, it is also the only prudent step forward. And if we ever dare to call something Web 3.0, this would be it.

Possibly Related Posts:

1. Social Networking More Important than Sex
2. Spock.com
3. The Top 10 Lies of Web 2.0
4. Aggreg8
5. MySpace Down: A Single Tear Slips Slowly Down My Cheek