Something worth pointing out is that # and @ got their start when the service was still rather young and most users were technologically affluent. Disseminating structs throughout the system, I think, will be much harder now, though I’m sure you’ve considered this. Maybe if you got Ashton Kutcher to adopt new microsyntax…

I’m very curious about how you plan to evaluate the success of proposals. I’d love to get into a deeper discussion on that sometime.

You mention adoption as the critical factor in success, and I agree. But there are many barriers to adoption, not just going against what people do. The history of technology is that people change what they do when a technology allows them to do something they could not do before.

Though I think there are a number of reasons for the failure of many (most?) microformats, one certainly has to be the fact that they are all complex, one-off efforts with few underlying technical themes. As an implementer of any application that might consume a microformat, I have to weigh the implementation cost with the value. Unfortunately, it hasn’t been demonstrated that most microformats provide enough value to end users to justify that work.

On the other hand (and not discounting the nightmare that it can be), XML makes it easy for applications to consume arbitrary data. Interpreting that data is certainly another problem that is complex on its own (e.g. understanding an XML Schema document or a WSDL), but at least getting an application to read the data and turn it into an abstract representation is trivial. Once applications can easily consume data–and the easier the better–one significant barrier to adoption is removed and the opportunity for users to find value opens up.

The ability to embed arbitrary, structured data in message streams, on Twitter et al., IM, SMS, email, and beyond, is unquestionably useful, and the way this is done has to be easy and useful to both application providers and end users. This is exactly the reason why I and several others have proposed the Twitter Data specification (http://twitterdata.org), with the goal of providing a formal means for embedding arbitrary data in Twitter messages in a way that works nicely for people and machines, in a way that is searchable, and a way that is open enough to build higher-level constructs on top (e.g. semantic triples). Without this formality, it’s a free-for-all, and implementation of any particular micro-/nano-/pico- format is increasingly unlikely.

While most attention has been on the commercial application which is beginning to deliver on some practical use cases, the potential for social impact is huge. There is a big opportunity to consider this in the early stages of protocol, standards, and infrastructure and keep an eye to optimizing it for public benefit. This goes squarely back to the spirit with which the web was created and the greatest need the web can serve tomorrow.

I’m glad you’re on that board Chris. Keep up the good work/advocacy.