The Roadblocks to Distributed and Open Social Networking

My last few posts have been about changing the paradigm for social networking sites and how a distributed and open social networking scheme would avoid a lot of the issues we see today. The distributed scheme I propose solves a lot of problems but creates a few new problems of its own.

Standards Problems

One of the problems that any open and distributed platform has is that it requires good standards that are agreed to and followed by all participants. For something like this distributed scheme to be a reality someone will need to step up and take leadership of creating the standards. Standards and open APIs are slow to evolve and often suffer from not being able to adapt to changes. With the social networking space being so new and so rapidly changing the standards would need to be updated very quickly and done very well in order for participants to stick with them and not create lots of little proprietary extensions. Google has suffered some criticisms of its Open Social APIs based on the fact that the APIs are both not finished and don’t provide all the functionality that developers are looking for.

Transition Problems

Transitioning from one way of doing something to another way is always difficult. This particular transition would be made more difficult in that some of the changes would be forced on the individual end users. For example a key feature of what I propose is the idea that you can tag relationships with different attributes and then control distribution of your information to that group - some info to co-workers, other info to relatives, etc. This would require manual intervention- individuals going through each of their “friend” relationships and tagging them appropriately. There’s also the big issue of people who currently have multiple profiles needing/wanting to consolidate those profiles into one. While under my scheme a MySpace user can “friend” a Facebook user I can’t imagine that Facebook would be too thrilled with transferring all of someone’s friends to their MySpace profile.

Security and Spoofing

teadrinkerSecurity in my scheme relies on the identity of the profile viewer - if you’re going to show different information to a co-worker and to a relative you need to know who the viewer is. This would present a tempting target to some shady characters who could attempt to present themselves as you closest friend to gain acces to your most private info. Within the APIs, verifying that the request really does come from a particular viewer isn’t terribly difficult. But with the open nature of the scheme users and profile providers would need to take special care to ensure that rogue sites didn’t come along and masquerade as another site. When you get that request from Jessica Alba at Facebock to be your secret lover you’d need take steps to ensure that the request really was from the lovely Ms. Alba and notice that this request wasn’t exactly coming via Facebook.

Bandwidth Issues

Under this scheme there’s a lot of addition traffic flowing around the net and the bandwidth and processing costs that come along with that traffic. Unfortunately for the sites this additional traffic isn’t something they can monetize so it becomes a pure cost. The traffic is caused by all the remote lookups a site would need to do when seeking to present profile information that’s hosted on a different site. If my profile is at Facebook and yours is at MySpace and we’re friends, when you view my profile at MySpace it would require MySpace to obtain my info from Facebook before sending it along to you.

Business Issues

Probably the biggest issue with this scheme is that it’s going to be a difficult pill for the social networks to swallow. These sites benefit from being closed and proprietary. The fact that if you want to friend someone on Facebook you need a Facebook account drives people to register at Facebook. It would seem that its not in Facebook’s interest to allow a MySpace user to friend their users and thereby lose those additional users. It’s a tough sell to convince companies to spend a lot of money developing something that in the end may cost them more money. For the distributed scheme to work it will require that enough of a critical mass be reached that the established sites are forced to join in and support the effort.

Will it Ever Happen?

So will this ever become a reality? Likely not in this form. Though I can’t help but believe that some limited form of openness will eventually come to the social networks. In the old days we had online BBSes that allowed messaging to other users on the site. This evolved into messaging across sites within the network and into email that works across the whole Internet. Closed systems like all the current instant messaging systems are slowly showing signs of opening up to allow communication across the various providers. I think that as social networking matures the users will very naturally demand that they not be forced to maintain multiple profiles to keep up with all their relationships and slowly over time the social networks will be forced to evolve in a more open and distributed direction.