[PREV - WEDDING_PARTY]    [TOP]

CROSSING_BROOKS

                                             June 06, 2022

On Frederick P. Brooks, "The Mythical Man-Month" (1975)
and the expanded "anniversary edition" from 1995.            He also published
                                                             "The Design of
This book is very well-regarded, with                        Design" (2010),
some of it's principles and slogans now                      which earns a
firmly entrenched in the culture of                          blue tab.
software development, in particular
"Brooks Law" (adding programmers to a                                 TABS
late project makes it later).  Various
other phrases ("There is no silver
bullet.") are very familiar even if         Though, you can exaggerate how
their source is forgotten.                  well know things like this
                                            are-- recently I had a
Nevertheless, iconoclastic skeptic          conversation with a manager
that I am, I want to remind that            and a designer, neither of
even Brooks can be questioned.              whom had heard of Brooks Law.

From the additional chapters in the
1995 edition:

      "Much more is known today about software engineering than
      was known in 1975.  Which of the assertions in the original
      1975 edition have been supported by data and experience?
      Which have been disproved? Which have been obsoleted by a
      changing world? ... "

      "Most of these propositions are operationally testable.
      my hope in putting them forth in stark form is to focus
      readers' thoughts, measurements, and comments."


Brooks goes on to summarize his
original points, with a list of      And that earns my
"assertions I believed to be true,   seal-of-approval,     The highly coveted
facts and rules-of-thumb-type        not that Brooks       "seal of doom".
generalizations of experience".      needs it: he could
                                     easily pose as a
                                     management guru
                                     who knows all, but
                                     he refuses to
                                     overreach.

Brooks was also someone who kept up
on the literature, and occasionally
uses published data, such as John
Harr's data on debugging rates
found in several different software    Brooks also frequently alludes to other
projects at AT&T.                      people's ideas, even when they're
                                       somewhat different from his own--
                                       another sign of intellectual integrity.
   I find this particularly
   interesting, because as
   far as I can tell that sort
   of data-based approach
   toward software engineering
   has disappeared.




--------
[NEXT - EVOLVING_BROOKS]