Fear and Fail.
This should be a longer post, but it won't be.
First, Fear is the mind-killer discusses ways to reduce fear in a software project. I like that the author points out that forcing us to to do the thing fear is one method of overcoming the problem(e.g. continuous integration), and the other way, which is very contradictory is by making failure less painful(e.g. unit testing, test driven development, etc.).
And on the same blog, linked from the same article is a nice discussion of applying the The Five Why's to a software project at a startup and how it lead to the evolution of a pretty stable, scalable system. In turn, that post mentions a paper: Holt-Winters Forcasting Applied to Poisson Processes in Real-Time[PDF] which sounds pretty interesting too(apparently its part of their gradual cluster roll-out solution on updates...).
First, Fear is the mind-killer discusses ways to reduce fear in a software project. I like that the author points out that forcing us to to do the thing fear is one method of overcoming the problem(e.g. continuous integration), and the other way, which is very contradictory is by making failure less painful(e.g. unit testing, test driven development, etc.).
And on the same blog, linked from the same article is a nice discussion of applying the The Five Why's to a software project at a startup and how it lead to the evolution of a pretty stable, scalable system. In turn, that post mentions a paper: Holt-Winters Forcasting Applied to Poisson Processes in Real-Time[PDF] which sounds pretty interesting too(apparently its part of their gradual cluster roll-out solution on updates...).