Tag Archives: Identity

OpenID needs a killer-app

openid_big_logo_textThe OpenID community is buzzing with board elections coming up next week. With Facebook and Google drawing attention with their simultaneous recycling of old news (MySpace’s PR person should have been fired long ago for being so bluntly left out), there is growing concern in that community for the future relevance of an open, rather than a commercially controlled, identity. Dave Winer thinks the commercialists will kill it by over-complexity. Chris Messina believes that better usability and branding could jumpstart OpenID. 

Personally I see no reason right now why Facebook won’t pull it off. The main reason is that they have a full turnkey system in-place. As a publisher, I don’t need to adopt an OpenID library, access a few Contacts APIs (standards are still only making baby steps) and then integrate some form of postback to an aggregator. Facebook gives me the whole monty, on a very large network provider, and a simple WordPress plugin can do all that work.

That’s where an open community lags behind a concentrated commercial effort – tying it all together to a killer-app. An open content platform? well, that was boringly nice, until someone connected it to a killer-app need of an anyone-can-edit-encyclopedia and suddenly everyone’s using wikis. Firefox did not become popular because it’s open, but rather because its openness allowed open features, such as greasemonkeying and tons of extensions. Where is the OpenID equivalent? 

Personally I think blog widgets, and in particular commenting platforms, can be exactly it. The blogosphere is decentralized enough for an open product to compete fairly against Facebook’s push. There are already commercial products making use of it, such as Disqus. Now just decentralize that too, and let every OpenID user assemble their own Disqus page from an OpenID-based commenting plug-in. If you don’t like that, find another potential killer-app (David Recordon’s browser-based identity has good potential too), just don’t assume that an open technology alone makes any difference to anyone beyond the techies.

Bootstrapping Social Search

As a followup on Brynn’s review of Delver, I’ve had an interesting exchange with Lachlan Hardy, where Lachlan expressed his disapproval of Delver’s crawling and unifying the social graph (content alone seems ok). My response is in this thread.

The important issue is that socially-connected search requires a comprehensive and unified social graph, which can be quite difficult to achieve. When users conduct their first search, they would expect all of their friends, friends of friends, and their respective content to be pre-indexed, for such a service to be of any use.

Skipping that part makes it impossible to bootstrap, and would be like a web search engine that includes only websites that opted-in to be included in the index, or like a FriendFeed version that shows no public profiles, and if you want to follow someone you must create their consolidated profile yourself. These can be regarded as far more privacy-observing services, but will probably never bootstrap as their real-life counterparts did. It’s all about keeping the balance right.