Comments, observations and thoughts from two bloggers on applied statistics, higher education and epidemiology. Joseph is an associate professor. Mark is a professional statistician and former math teacher.
Tuesday, May 17, 2011
I can think of some exceptions
But I suspect Donald Knuth was excluding books like Light in August and Huckleberry Finn when he said this.
Nope, writing software is infinitely harder than books. Software has the fool paradigm against it - nothing is foolproof because fools are ingenious. I worked on some award winning projects and there was hardly a day or week that went by when some customer tried to do something stupid with the product. And because customer is always right, we had to modify the product (or the documentation warning you not to do this or that) to accommodate that particular action. Writing, when it's done and finished and never comes back to the door. Furthermore, writing a book is a what, a two or three person project, the author editor and possibly and illustrator. Take a peek at the acknowledgements of and commercial software package - hundreds of authors, tens of QA, and then there's marketing...
Nope, writing software is infinitely harder than books. Software has the fool paradigm against it - nothing is foolproof because fools are ingenious. I worked on some award winning projects and there was hardly a day or week that went by when some customer tried to do something stupid with the product. And because customer is always right, we had to modify the product (or the documentation warning you not to do this or that) to accommodate that particular action.
ReplyDeleteWriting, when it's done and finished and never comes back to the door. Furthermore, writing a book is a what, a two or three person project, the author editor and possibly and illustrator. Take a peek at the acknowledgements of and commercial software package - hundreds of authors, tens of QA, and then there's marketing...