The document metaphor paved the way for the mass adoption of personal computing, providing a set of concepts and language that most users were already familiar with. This metaphor is so pervasive it’s really hard for us to step aside from it now, roughly 30 years later. Even on the web, where we’re being weaned off of Documents little by little is still primarily built around the transmission of documents at its core.

Early on, Documents solved the problem of user adoption and translation of work processes to computers and networks. But since then, work process and users have evolved, and the problems preserved become more and more onerous. It’s time to take of the training wheels and ride the bike.

The biggest problem with Documents is managing the copies. The web grew out of hypertext and networking as a means of solving this problem, providing URLs as a way of safely referencing remote documents, rather than having to have copies locally available. This grand vision has been partially successful in reducing the need for copies. But some copies are made to preserve various states of documents over time, resulting in version control issues that are solvable, but introduce a level of complexity that is difficult for most users to manage.

In practice, documents are created, forwarded via e-mail, editing, forwarded again, creating a sea of mostly-redundant documents riddled with small changes, some irrelevant, and some of critical importance. The e-mail trail leaves little or no meta-data to distinguish one copy or set of changes from another. Software developers have used source-code control systems to manage copies and versions with great success, but these systems are difficult for non-programmers to use.

Another problem with Documents is a multitude of formats and media types, and lack of accepted standards for them. This means that many documents are only usable for users on specific computers with specific software packages installed. Once again, the web has partially solved this problem with standard formats and rendering definitions, though it has been a constant battle to preserve a universal standard, with competing corporations intentionally departing from standards to stake out their territory. This is a particular problem with media such as sound and video.

What does the world look like without documents? For the most part, it’s already here: Wiki’s, Blogs, and search engines. Sharing information via a wiki is really the most effective solution. Search engines solve issues about cataloguing and largely eliminate the need for taxonomy.

Web applications such as wikis and blog software help with the format problem by relying on web standards, and can also streamline version and change management, as well as attribution. Information can be copied from these sites for convienence or off-line reference, but there’s not any doubt about where to find the authoritative version.

What does this mean for Microsoft Office, OOXML, Open Office and Google Docs? It’s means they’re a waste of time, a large vestigial organ of computing. The key problems going forward are Internet Standards, better connectivity, security auditing, support a diverse set of devices, and backup and archival of information on the web.