But imagine if you were to systematically note certain structural features of the text. If you are anything like us, your markings tend to be all over the place. Think about all the annotations that you put on your own pages as you read them. There are a number of ways to represent such structural information, and we can get towards one that works for a computer by working through a system that you might use on your own when you read.

Nor it would understand anything about how the internal components of those lines are connected to each other.

It's all just text.Ī computer program looking at the above passage from Oscar Wilde's The Ballad of Reading Gaol would, most likely, not even recognize those six lines as related in any way. Right now it just thinks the text up there is no different from the other lines of text on this page. For example, if we were to say to a computer, 'Hey! Find me the poem in this lesson!' It would have no idea what we were talking about: we have to find some way of telling the computer where it can find the poem. Computers tend to work in hierarchies and clear-cut structures, and, even then, they only know about those structures that someone has told them about. What elements of the text convey meaning? How do they do so?Ĭomputers have a hard with abstract concepts like this. When we read, we tend to skip to much more complicated understandings of a text:

Many of the methods that you will learn are simply sophisticated ways of counting words, whereas reading entails far more complicated processes of interpretation and analysis. But, as you read along, you may notice that they cannot do all that much. This book studies texts and the things that computers can do with them.

Text encoding initiative tei