The Personal Text Miner (sidebar)

How this happened ...

Back in 1989-90 one of my favororite catch-phrases was "No word-processor can process words". I tried to program a few tools for manipulating texts, among those one that would search for repetitive use of words - even parts of words, their morphological stem, that is - in a text. A few years later I realized that we need programs not just for writing but for reading also. And we need them badly.

Rather early during the first wave of CD-ROM publishing I got tired of bad interfaces. So I started to extract the texts from their program shells - and dug into them with different word-processors or editors of my own choice - thus getting to be much more effective in my work.

I started to think of this in connection with evolving methods of data mining and text mining. Those techniques would be excellent for the future reader, not just for large corporations or organisations to provide for their needs regarding informational logistics. I hinted at these ideas now and then in lectures and articles, and in 1999 I wanted to visualize this idea in a more concrete fashion, so I made these series of images to show what a future reading tool for individuals might look like.

Of course, this is a very anachronistic approach, depicting a futuristic computer tool with the help of Mac OS interface details. But maybe it is still better that way, to show something unfamiliar in a familiar setting.

The idea is mine, but it is founded on techniques that are either already possible or will be. This is not really an attempt to predict the future. The problem with predictions is that although things may be technically possible to do, they might still not be realized. So it is more the way to think about these matters that I want to convey, not the exact solution.

Close window to return or go back within this window