Raw

Elias Torres has announced WP-APP, a drop-in bit of PHP. (Thanks for the nod!)

Heh, I was just going to say "Cool!" but it's turned into a ramble...

APP, like the Atom Format has emerged from the human-readable content world, its main application domain being traditional CMSs. It's based squarely on what we know of good practices for the web. The initial gain is getting away from the straitjacket of RPC over HTTP, but because of the foundations its potential goes a lot further.

A tired misconception on the Semantic Web vision is that it aims to create something entirely new rather than supplementing (or rather filling in missing bits) of the current web. Ok, if you view the SemWeb as purely the RDF/OWL bits, then there may be some truth in the belief that its an ocean-boiling exercise.

But there's a whole spectrum in between what we've got now and that artificial vision. I've got a proper write-up tabled (in the English sense) for an article next month, but I think APP is an aspect of what I've been calling the content model layer. Basically there seems to be an emerging trend from the current document-orientation towards a more general data-orientated view of the same material.

Independently of the Semantic Web initiative, the machine-readability of stuff on the Web is being increased. The metadata of Atom is essentially resource description, microformats are bringing the description of real-world things (i.e. pure data) to the web. More resource description. The internal models of things like these can each be taken as domain-specific schemas/ontologies (Atom being a generalised vocabulary for human-readable, web-delivered content). In a sense this is a narrow approach, there isn't an underlying fully-generalised formal model like that of RDF. But the significance is that it's refactoring the material we have on the web now into a data-oriented conceptual framework. That's transparently equivalent to putting the stuff on the Semantic Web, Semantic Web tools can use the data more-or-less directly.

Sure, to get the full benefit of the Semantic Web we need a broader range of data online in addition to human hypertexts and social-networking specifics like calendars and address books. But retooling to support this stuff in a way which maximises the amount of benefit computers can offer is a big step forward. With systems like APP, the Semantic Web is getting a huge boost independently of Semantic Web initiatives.

I guess the take-home is that you don't have to think about the Semantic Web to contribute to it. For instance, as Elias put it, you can:

Simply drop app.php into your root wordpress install directory and have some Atom/APP fun.