designed the Santa Teresa Laboratory, they took some highly unusual steps to ensure that their programmers could be productive: They studied programmers at work, and designed a space that enabled productivity.
Based on their scientific studies, they concluded that each person needed at least 100 square feet of private office space, with windows and doors, and at least 30 square feet of work surface. A flexible arrangement of furniture also helps.
This is described in an article by Gerald M. Mc
Cue in IBM Systems Journal, Volume 17, Number 1 (1978), pages 4 to 25. http://www.research.ibm.com/journal/sj/171/ibmsj1701C.pdf
(The paper is reprinted in SoftwareStateOfTheArt
; selected papers, edited by TomDeMarco
The C3 team is one of the most consistently productive teams I've ever observed. They work in a large open space with two big tables with computers studded all around. The reason that this works, I believe, is that they mostly work in pairs. When you're working with another person, the soft discussion between the two of you drowns out other distractions, and pairs are very discernibly more productive than two people working alone, especially over the longer term.
For related discussion, see also PairProgrammingErgonomics
. -- RonJeffries
I worked at the IBM facility in Boca Raton (birthplace of OS/2). It had the best offices I've ever seen. Despite all the excitement around "pit areas", I found that a quiet work environment was most conducive for thinking and working. Interestingly enough, each office held two programmers. Since you sat back to back, it was possible to do PairProgramming
or not. Your choice. That's why I'm a believer of PairProgramming
and not one of PitProgramming?
. -- RonPerrella