3 ms·
I seem to remember first seeing a table like the one the author describes in "The Mythical Man Month" but my copy is currently at home. The data underlying tha
by smoyer 14y ago
I seem to remember first seeing a table like the one the author describes in "The Mythical Man Month" but my copy is currently at home. The data underlying that book was gathered from projects at IBM in the '60s and '70s and I don't really doubt that the underlying data was fairly represented.
The bigger question is whether improvements in processes and tools have obsoleted this data. TDD and automated regression testing would be one place where inter-phase defects could become less costly. On the other hand, projects that have uncorrected architectural errors can be completely functional and yet their maintenance costs never decrease.
If someone has a copy handy, please validate my memory otherwise I'll check when I get home this evening.
- alttag 14y agoI've just gone through it page-by-page, and didn't see a table like that. Using the 20th Anniversary Edition, I also looked through the summarized list of claims in each chapter, and the 20-year retrospective, and did not see the table or a section that might have been a textual version of the same data. Perhaps the closest bit I found was in the chapter "Plan One to Throw Away": "The total cost of maintaining a widely used program is typically 40 percent or more of the cost of developing it" (p 121). Similarly, another section quotes Capers Jones, "Focus on quality, and productivity will follow" (p 217, emphasis original). That, I think, is about as close as MMM gets to these claims.
- smoyer 14y agoI couldn't find it either but in addition to "Code Complete" this data is also shown in a chart in "The Software Project Survival Guide", also by Steve McConnell (p29). The attribution is "Researchers have found" but Boehm and Papaccio are listed in the references.