The Mythical Man-Month: 8 key ideas
The core ideas of The Mythical Man-Month, each in under a minute, written in our own words. Want the full depth? Get the book.
- IDEA 1 OF 8
People and months are not interchangeable
Brooks, who managed IBM's huge OS/360 project, argues that the man-month, measuring work in people multiplied by time, is a dangerous myth.
Software work cannot simply be split among more people to finish faster.
- IDEA 2 OF 8
More people can slow a project
Brooks observed that adding programmers to a late project usually delays it further.
Newcomers need training, and the number of communication paths grows rapidly. Time spent bringing newcomers up to speed and holding more discussions can outweigh the extra hands.
- IDEA 3 OF 8
Communication costs grow fast
Brooks explains that as teams grow, the time spent coordinating increases sharply.
Large teams can spend more time talking than building. Every new member adds links with all the others, so the web of talking grows quickly.
- IDEA 4 OF 8
Teams should be like surgeons
Brooks proposes a surgical team model, where one chief programmer does the core work, supported by specialists.
This keeps design consistent. The idea is that one mind guides the work, while others handle supporting tasks and free that mind.
- IDEA 5 OF 8
Unity of design matters
Brooks argues that conceptual integrity, a consistent design from a single vision, is the most important quality of a system.
Too many cooks spoil it. Features that do not fit the main idea make the whole product feel confused.
- IDEA 6 OF 8
Second systems get bloated
Brooks warns of the second-system effect, when designers add every idea they held back from their first system.
The result is often overloaded. Because the designer feels freer, restraint is lost and the product becomes heavy.
- IDEA 7 OF 8
Expect to rebuild
Brooks argues that the first version of a system will likely need to be discarded.
Planning for this saves trouble. The first attempt teaches lessons that only come from building, so it should be seen as practice.
- IDEA 8 OF 8
There is no magic fix
In a later essay, Brooks argues that no single technique will dramatically improve software productivity.
Progress comes from many small gains. Each tool or method helps a little, yet none removes the basic difficulty of building software.
Get one idea like this every daySwipe through ideas, go deeper on any of them, and save the ones that matter. Free.
Read The Mythical Man-Month in the app