Summary: I analyse a 1994 Personal Information Management program and compare its goals to what we want in in a similar tool in 2004. I discover the requirements are basically the same.

The blogosphere is mostly a synchronous give-and-take of content. People largely comment on and link to things that other people are commenting on and linking to. It's a circular flow of information, with a particular point in time always at the epicentre. It's why 99% of weblogs are primarily ordered chronologically - with the most recent post at the top of the page.

When I'm looking for information to quench my insatiable thirst for knowledge, I often use the Web in an asynchronous manner. That is, I like to read historical web documents and compare them to current blogosphere memes. The Wayback Machine is my friend in this regard. Why, just last week I discovered a gem of historical Web documentation: the Electronic Proceedings of the Second World Wide Web Conference '94: Mosaic and the Web. This is a record of all the presentations made to the 2nd annual WWW conference back in 1994. I intend to browse through most of the presentations in due course, but for now I want to tell you about the first one that tickled my Interest gland.

It's funny how a tool developed 10 years ago can still accurately describe the requirements of the 2004-era Web. That's exactly the case with PAINT, a "tool for individualizing the Web". Here's the executive summary:

The increasing complexity of navigating the Internet is becoming one of the fundamental obstacles to its effective use. This is due to the nature of the Internet, principally, a disorganized collection of both sites and site documents whose exponential growth rate rapidly is outstripping any user's ability to master it. There are two ways to deal with this complexity: reorganize the structure of the Internet or give each user the ability to organize an individual perspective of the Internet. Although the former would produce more global benefit, the latter is both easier to accomplish and potentially more beneficial to any individual or group of users.

Our approach, therefore, is to create a navigation tool which copes with Internet complexity at the individual, rather than the organizational, level. This tool, PAINT (Personalized, Adaptive Internet Navigation Tool), allows the user to impose a hierarchical organization on Internet sites and documents of interest by creating categories under which to group sites. Such categorization can be used not only by an individual user, but also can be shared among groups of users with similar interests. PAINT will also provide local automatic classification based on user parameters and user behavior. That is, PAINT will record visited locations and categorize them according to past use. The user is then free to examine the automated organization, modify it, and make it a personalized view of the Internet. In our report, we will describe the PAINT tool, its use, and some preliminary investigations of local, automatic categorization.

This webpage, even though nearly 10 years old, still in a nutshell describes what we're looking for in a PIM (Personal Information Management) appliance circa 2004. You can get all fancy and talk about wanting agents to gather data automatically, or using Bayesian filters, or latent semantic indexing. But really it still boils down to this: we want a tool that (in the words of Paint) individualizes the Web

Take the following sentence from the first paragraph in that 1994 webpage. It outlines the central problem - complexity - and the two general solutions. In red type, I've added how these two solutions are (generally speaking) being approached now:

There are two ways to deal with this complexity: reorganize the structure of the Internet (2004 = the Semantic Web) or give each user the ability to organize an individual perspective of the Internet (2004 = bootstrapping; eg what tech bloggers are now trying to do with their weblog taxonomies).

The program PAINT was designed to take the second approach. Paint wanted to put the user at the centre of their own personal Web:

This tool, PAINT (Personalized, Adaptive Internet Navigation Tool), allows the user to impose a hierarchical organization on Internet sites and documents of interest by creating categories under which to group sites.

The key things to note: PAINT enables people to create a hierarchical organization for their information, by grouping items into categories. Hmm, sound familiar?

PAINT circa 1994 was first of all an extension of the Mosaic web browser's hotlist facility. Hotlists were the equivalent of Favourites in the modern IE browser, or Bookmarks in Netscape. But at the time, hotlists could not be organized into folders. You just had the one list of documents and websites. So it could be argued that PAINT was simply a description of what IE Favorites or Netscape Bookmarks became a couple of years later - a hierarchical set of folders with which to store website URLs.

But I think PAINT's goals were deeper than that. Look at your usage of Favorites or Bookmarks today - do you use them as a way to categorize information you find on the Web? Do you organize your information into a hierarchy using the folders available to you? If you're like me, once upon a time you made an effort to do all this, but it long ago fell by the wayside. With the advent of RSS and Google I hardly ever use my IE Favorites anymore! And yet we still have this over-riding need to organize our information on the Web...

But obviously I can only take a comparison of PIM requirements then (1994) and now (2004) so far. What's different now? For a start we've had an exponential increase in the amount of data and information on the Web, thanks in part to having weblog tools that allow anyone (technical or no) to publish on the Web. But perhaps more fundamentally, information on the Web is now published as "microcontent". Information exists in "chunks", and each chunk of information is defined with a permalink. True, we haven't yet reached the stage where individual paragraphs or even sentences are given permalinks - but maybe that world of data isn't too far off.

So, could PAINT - or more likely a PAINT boosted with 2004-era technologies - be used to help us build weblog taxonomies based on categorizing our content hierachically? Well yes, but we're already building such tools. Dave Winer has developed a product called Channel Z which categorizes weblog posts into categories created by the author. k-collector allows bloggers to create and post to categories in a shared directory. And some clever bloggers (eg Paul Ford, Erik Benson, Bill Seitz) have created their own automated back-linking categorizing extravaganzas. So we're moving towards the goals that PAINT (and others I'm sure) defined back in 1994, and that visionaries such as Ted Nelson and Vannevar Bush defined decades before that. We haven't got there yet though. Most of us still muddle our way manually organizing our Web content. PIM Nirvana hasn't yet been developed. But with initiatives such as Chandler taking over the mantle from PAINT as the next big thing of PIMs, the circle of Web innovation continues and the dream lives on. Everybody wants to control and be at the centre of their information environment - will we ever succeed?