Some preliminary remarks: It is slightly odd to give an introduction into a NLG shell system which generates German text, and to do so in English. Its justification stems from the observation that virtually every native speaker of German with an interest in NLG can be counted on to read English (hopefully my crude English as well) rather fluently. Moreover, many, but by no means all, phenomena addressed by Suregen-II seem to be relevant in English medical documents as well. Therefore, I hope to do both sides justice instead of serving neither.
I would like to direct the reader with only little attention spare to the last section (an elaborated example) as I hope it illustrates best what Suregen-2 is all about.
An important confession is still due: This introduction scratches only the surface of many NLG related problems and, even worse, does not address some of them (lexical choice, for instance, or the generation of referential expressions). This will be alleviated in the next version, I simply wanted to set a starting point here.