The web was originally a space for people to read/write documents and link them together…it’s mad when you see what has become. The idea of the web was a mash-up of other existing concepts and happened to be the implementation that took off. Almost by accident, the www has become this world-eating giant, and so fast!
This kind of explains why technically and through a user-interface point-of-view, there are so many problems with the majority of the sites out there. Browsers don’t render sites in a consistent way, site designers use all kinds of hacks or just ignore the audience and make monstrous designs… The web should be as simple and flexible as it can be for the user, like Berners-Lees’ WorldWideWeb browser. The technology should be hidden from the user, but allow for content (rather than the existing definition of a site) to be accessed through many methods and used for different purposes.
Content is what its all about. The navigation is a means of accessing relevant content, not strictly about perusing pages. The paradigm of the web being individual documents much like a book is dated – what we actually have is many documents, document fragments and media files; all of which has meta data and can be related to each other, regardless of location.
This flexibility relies on tools. While Dreamweaver and other wysiwyg editors churn out pages, the real issue is over what the site should contain. A good approach behind the scenes may be to follow this path and establish links between content elements and media files within a ‘page’ context that has various views of it. With more devices of different capabilities accessing the web now and in the future, a whole, current website may not be needed on a mobile phone for example but elements may. The question is how can a site’s structure appear different dependent on the device (user-agent) used?
This is where the traditional page model lets us down in this modern context. Pages can be roughly reformatted for multiple devices but this can be very costly (ie. time consuming) and the returns may not justify this. If the content knows it’s own relationships with other content and media, we could dynamically alter structure based on given parameters or the capabilities of the client device.
This is an area I’m looking into for CMS applications. It would be interesting to see how this altered model could be presented to a novice user. How could terms be used that adequately describe processes but not shoot above the user’s level of understanding…?