Link to the Article

Mr Gouy has provided his own formatted article, not in this site's format, so here's a link to it: Misstating the Evidence for Agile and Iterative Development.

I have not compared the original sources to Craig's quotes, and have formed no opinion on whether he is appropriately quoting the documents he refers to, much less whether his interpretations are better or worse than Gouy's, nor whether interpretations are accidental or intentional mistakes. I have one of the original sources referred to here, and I'll take a look at it when I become bored with watching paint dry, and will comment here regarding what I find. I'm also inviting Craig to comment here if he cares to. (He has now done so: see below.)

Mr Gouy's comments may well be accurate, for all I know. As all here surely are aware, I'm a proponent of Agile methods, and my preference for them is based on nearly a half-century of software development practice, and on over ten years' observation of tens of teams actually trying Agile processes. Therefore, while I find Craig's book to be interesting and convincing, my confidence in Agile methods is due to direct experience with those methods, not with anything I learned from books.

In any case, here's Mr Gouy's article, and I hope that somehow it serves us all well.

Reply from Craig Larman

I dropped Craig a note asking him if he'd care to comment on what Isaac has said, and he sent me a brief reply from Oslo, Norway, where he's presently working with "agile offshore" projects. Here's what he said:

On your question about Isaac Gouy's investigations: first, I've been way too busy to spend time on the subject in any depth, but he was very kind to send me some corrections or clarifications on what I wrote in my last book. My view is that sotfware engineering shouldn't be religion and should indeed be based on evidence and statistics, and I appreciate Isaac's or anybody's improvements to whatever I wrote. (I hope Isaac may be a reviewer on my next book, because he seems thoughtful).

Without having the time to investigate in detail, I guess that any misstatements on my part that Isaac identified come from the fact that I wrote the last book while mostly on the road and I relied usually on second-hand references (e.g., on a paper on the web or in a book I had with me that referred to the source paper) rather than the original source references, and that second-hand summaries may not give the clearest picture. For example, I think I remember reading an example by James Martin that timeboxing resulted in X% improvement in productivity, but in the summary that I read James Martin didn't mention that the project-comparison study also involved different software tools in the non-timeboxed vs timeboxed comparison, which I learned later.

I suppose some differences just boil down to perspective or interpretation, which will always exist. There was no intention to misstate, and I appreciate any help to improve content.