Data Portability
Dec 5th, 2007 by Aswath
With the rise of social networks, a sizable number of people have expressed concern that user generated content is being locked up in hands of a few who benefit enormously. To protect this, these networks have hindered attempts by users to port the data from one network to another, confirming the concerns expressed by the first group. In a recent post, Abhishek Tiwari echoes these concerns and suggests that these networks should enable web-service access to these data. To a large extent this is the idea behind OAuth and OpenSocial from Google. Bill Wishon while commenting on this post observes that large networks like Amazon and Facebook have no incentive to share the data. So it appears to me that an alternate solution would be for users to save a copy of the data they generate in these sites for later use. During the days of typewriters, we used carbon paper to make a copy as we typed; email clients routinely save a copy of the mail we send out. So why not the browsers make a copy of the data that we generate at these sites? To elaborate further, I am suggesting that a plug-in for the browser that will scrape the local screen, collect the data as it is being generated and uploaded to the social networking site. Thus collected data can be stored either locally or if further sharing is required, in a user-controlled server residing on the public Internet. Not surprisingly, I am thinking of EnThinnai here. Since EnThinnai provides controlled access to the stored data, users can decide which other third parties can access which portion of the stored data and when.
Hi Aswath,
I agree that local storage of important personal data is one attractive option, I’d take it one step further and suggest that as data storage gets smaller, cheaper and more ubiquitous that we might even store this data on a personal device such as a cell phone or USB drive. That way we can have the information with us at all times in case we want to log in to our social networks while we’re away from home. These devices can adopt some form of biometric authentication to prevent access to this data by the wrong person if lost.
With the right client side integration websites and applications could ask the user’s profile manager for access to certain information and you could allow or deny the request for one time, a week or until you change your mind. Much like a personal firewall works today for managing connections to and from your machine.
Cheers,
~>Bill
Just a quick clarification… in your post you make it sound like OAuth is a Google initiative, like OpenSocial. This is not accurate. While Google has been productively involved in the development of the protocol, it is not under their direct guidance, in the way that OpenSocial is.
This is important, as OAuth is a collective effort by a number of providers, including AOL, Yahoo, Flickr, Twitter, Ma.gnolia, Hueniverse and others.
It may not have been what you meant, but that’s the way it currently reads.
As for your point, I think the browser has something to do with this — but even still, the problem really will be putting all that data to good use and efficiently accessing what you want, rather than just storing it forever.
Chris, thanks for the clarification. It was my bad sentence construction. As you suspected, I know OAuth is not solely from Google.
I was suggesting that, an EnThinnai user can share that data with the identified friends.