Files
Obsidian-Vault/Career/Resources/The Mythical Man Month/Ch 2. The Mythical Man Month.md
2025-10-25 20:11:21 +02:00

2.1 KiB

The reasons why estimations often go awry are multifaceted. Firstly, estimating techniques are often poor and can fail to accurately predict the time and resources required for a project. Furthermore, these techniques can confuse effort with progress and assume that men and months are interchangeable, which is not always the case.

Another reason why estimations can go wrong is due to uncertainty, which can lead to a lack of stubbornness in adhering to schedules. Additionally, schedules are often not monitored, and schedule slippage is often countered with increased manpower. This can result in a further decrease in productivity due to communication issues and the time it takes for new team members to become familiar with the project.

Programmers are often optimistic, perhaps due to their youth, and assume that there will be little implementation difficulty in their work.

Due to the absence of productivity and bug incidence figures, estimates can be inherently poor. However, in such cases, programmers should stick to their estimations rather than engaging in wishful thinking.

The man-month measure is only useful when no communication is needed between workers. However, communication is necessary for effective project management and can become increasingly difficult as the number of workers increases. The effort of communication must be factored into the estimation process, as training grows linearly with the number of workers and intercommunication grows exponentially.

System tests are the most challenging part of a project and are often done last, resulting in the greatest cost. A recommended schedule for a project would be to allocate one-third of the time to planning, one-sixth to coding, one-quarter to component testing, and one-quarter to system testing.

When faced with schedule slippage, it is important to ask whether the original estimate was too low locally or universally. The cost of adding more men is N times the training, and communication can become exponentially harder. Finally, the number of men required is dependent on the number of individual subtasks.