Tracking the tools that decentralize the media. tools process ideas resources eventsav

unmediated

 

August 08, 2005

When I look for the things which distinguish Web 2.0 applications from earlier ones, I don't see chromed-up DHTML interfaces, I see DOM hacking, REST, semantic HTML, microformats, end-to-end design and other things more associated with consensus than innovation. The movement is about orthodoxy, not radicalism.

In this sense, Web 2.0 is an indicator of the long-term shape of the web: what we are seeing is the rise of a canon.

The rise of this canon is fueled by increasing returns. As more sites use HTML semantically rather than presentationally, the benefit of doing so grows; ditto for REST over RPC, a common DOM for Javascript, and for features which allow systems to be extended at the endpoints. Though it has always been beneficial to stick closely to standards, it is now more beneficial than before, and in the future it will be fatal to not do so.

Microformats as retro-radicalism

For example, let's assume that microformats make sense for business. The h* family of microformats, like hReview and hCard, are a sort of HTML fundamentalism. The premise is that HTML already supports most of the semantics a new data format would need, and that reinvention is destructive.

I don't love this way of thinking, in the sense that it doesn't give me warm feelings. It's too confrontational for me. On a practical level, though, the benefits are overwhelming. If you use h*, you can easily take advantage of the Javascript DOM and REST, and these are powerful tools.

Microformats are a bleeding-edge idea at this point, and at the same time they are a retrenching. That is the tension driving the web towards orthodoxy both in the long term and in the 2.0 generation.