I’d be interested in your brewing article, “In Defense of Walled-Gardens.” I am not for these large centralized services with user lock-in, and federation is an appealing concept, be it an old concept when we look at email. The benefit is openness and detachedness.

I can’t say how many times I’ve wanted to post a reply to a number forums topics, but am unable without registering and logging in. The issue is, I don’t want multiple accounts on services with redundant functionality. Social networks are a great example because they generally all provide the same features, but each requires usership to interact, they are all closed communities.

If you do write the article, I hope to see some analysis of the OpenMicroblogging specification, which imo is a decent open solution, and I think it keys on the main factor federated social system must rely on for “security” and that is the social relation opt-in. A user subscribing to another [possibly federated] user’s content is making that decision, essentially yielding authorization for a publisher to post to the subscriber’s inbox. If the publisher decides to exploit this fact, or is compromised, what is the exact security threat? I believe it varies between systems, especially when comparing OMB and Wave.

OMB is generally rendered in one specific way, while Wave will allow self-defined rendering with widgets/html/js/css. In that light, I don’t necessarily feel the architecture is a threat, but I do think the content possibilities need a sanitizing method at a more fundamental level in the overall system.

Federation is a long-standing, open solution, email proves that. When the concept of “walled gardens” expires and the focus shifts to federated solutions, I think we will have the right answers.

This is why I enjoy the best open+social community, the blogosphere.