http://www.daniel-lemire.com/blog/archives/2007/11/29/storytelling-and-research-papers/ Daniel Lemire's blog is about life in academia, research in Computer Science, wondering how we can reconcile fast databases and algorithms with the informal and asemantic nature of the world around us. It is broadcasted from Montreal (Canada). Fri, 21 Nov 2008 18:13:48 +0000 http://wordpress.org/?v=2.6.3 http://www.daniel-lemire.com/blog/archives/2007/11/29/storytelling-and-research-papers/#comment-49599 Peter Turney Thu, 29 Nov 2007 19:47:51 +0000 http://www.daniel-lemire.com/blog/archives/2007/11/29/storytelling-and-research-papers/#comment-49599 Stories generally rely on the power of surprise, withholding important information as long as possible, to maximize dramatic effect. I believe that a good scientific paper should do the exact opposite: the most important information should be given as soon as possible; for example, in the title or in the abstract. Very early in the paper, you should give the reader a general framework for understanding what will follow. You should seek to maximize clarity and understanding. Stories, on the other hand, thrive on mystery, ambiguity, and vague allusions. Stories generally rely on the power of surprise, withholding important information as long as possible, to maximize dramatic effect. I believe that a good scientific paper should do the exact opposite: the most important information should be given as soon as possible; for example, in the title or in the abstract. Very early in the paper, you should give the reader a general framework for understanding what will follow. You should seek to maximize clarity and understanding. Stories, on the other hand, thrive on mystery, ambiguity, and vague allusions.

]]>