Here is the discussion of what sort of corpus format we should be using and how to process the incoming texts into that format.

First we have to decide about the XML -format. What information do we want and how to represent it. Simplest way to define the XML -format is to represent it with DTD. Other representation options include W3C Schema and Relax NG. We use DTD here and transform into some of the schemas in the future.

DTD -discussion

Second target is to figure out how to convert the incoming raw original files into our XML -format. Below are listed the formats to be consisdered in the conversion process.

• Word document (.doc)
• Adobe document (.pdf)
• Web -documents (.html)
• Plain text (.txt)
• Other formats? (...)

Conversion

Filesystem

OpenOffice -format

Last modified: $Date: 2009-05-11 11:06:46 +0200 (man, 11 mai 2009) $, by $Author: boerre $