|How to Excel
in today's world
I am working on creating links to studies that show what does and does not work for software development in particular and managing
employees in general. If you know of any studies, please let us know. We are looking for studies and research, not opinions.
Click here to let me know of research done on productivity
I have always strived very hard to have the people working for me not get interrupted. There have been
numerous studies showing that interrupting programmers has a giant cost. Now there is a study showing that this
holds for almost every job (makes sense). Take a look at:
Hiring Smart People
Ok, management book after management book makes the claim that the best programmers are between 10X and 25X more productive than the worst
programmers. And this claim does make sense from what I have seen in practice. In fact, I think there is more to it than that because the
poor programmers cannot create as good a product no matter how much time they have.
But the question I have is - has anyone done an actual study to back this up? If so, could you please point me at it?
IBM's Santa Teresa Study
The special needs of the computer programmer in terms of working space, furniture design, access to terminals and conference
rooms, and overall working environment led IBM to construct a facility intended to enhance programmer productivity in a
development environment. That facility is the Santa Teresa Laboratory in San Jose, California, designed by MBT Associates
of San Francisco. This essay discusses the programmer's needs, how they were perceived, and the process by which they led to
unique design concepts, as well as the architectural philosophy underlying the design process.
Click here to read the study
(thank you to Michael Fogeborg for sending me the link.)
Offices vs. Cubicles
Now here we have studies. IBM did the Santa Thersa study and Tom DeMarco did a study. Both found offices providing a 2.5X productivity improvement
over cubicles. (If anyone can point me to an online copy of these studies I can link to - I would appreciate it.)
I am tracking down links to the following:
And having had to suffer through working in cubicles at one job, I learned first-hand just how bad they hammer productivity. And I think, based on
personal experience, that 2.5X is way undercounting it. Cubicles are one of the stupidiest ideas ever imposed on employees by management.
- Mythical Man Month, Brooks, 1995
- McConnell, Steve. "Code Complete". Microsoft Press. 1993. pg 550-555
- Boehm, Barry W. "Improving Software Productivity". IEEE Computer, September, 1987. pg 43-57
- Weinburg, Gerald M. The Psychology of Computer Programming. Von Nostrad Reinhold. NY. 1971
- DeMarco, Tom and Timothy Lister. Peopleware: Productive Projects and Teams. Dorset House. NY. 1987
- McCue, Gerald M "IBM's Santa Teresa Laboratory - Architectual Design for Program Development". IBM Systems Journal 17 No. 1. 1978. pg
- Assessment and Control of Software Risks" by Caspers Jones, chapter 10
- "Design for Research: Principles of Laboratory Architecture", by Susan Braybrooke (John Wiley & Sons, 1986)
- "High Technology Workplaces: Integrating Technology, Management, and Design for Productive Work Environments" by Pierre Goumain (Van Nostrand Reinhold, 1989)
- "Programming Productivity - Issues for the Eighties" by Caspers Jones (IEEE Press, 1986). This one includes a study by the Dean of Architecture at Harvard.
- "Programming Productivity" (1986) and "Applied Software Measurement" (1991) by Caspers Jones.
So here's the challange. Can anyone give me the name of a productive, successful software company that does put it's employees in cubicles? If you
can give me one, I will list it here. Because I think all of the successful ones do use offices. They are smart enough to know what works. (I do not
count companies like U.S. West and J.D. Edwards because while they are profitable and successful, their productivity is low, their employee morale
sucks, and their turnover is atrocious.)
The real question should be - companies spend close to $150,000/year on programmers counting salary, equipment, hiring, training, etc. and then they
save $200/month using cubicles instead of offices. Let's see - hire twice as many programmers at 150K/year because of the cubicle productivity hit
to save 2.4K/year per programmer. Even the accountants should see the problem with that.
Ok, I have heard from 5 different people that a study was done showing that there is a 20% increase in productivity from giving programmers a larger
monitor. It makes sense. But I have not been able to track down the actual study. If you know where it is, could you point me to it?
This is a link to a report that lists an unreferenced study as well as numerous opinions and rough estimates from several companies.
It's a better source than 1 or 2 individual opinions - but it is not as strong a study as I would hope has been done.
The Case For Large Screen Displays
Thank you to the following people for pointing me at sources: