I worked at Wang for 13 years, from boom through bust. Quarter after quarter of 34% growth felt like it would go on forever, but of course it didn't. Wang was good to and for me - I morphed from technical writer to engineer, fresh out of school student with no idea what a job was to a valued employee. But living through the fall was sad. I left of my own accord in 1992, just months before the bankruptcy.
As Wang was thrashing, they signed up hook, line, and sinker for the "Quality Leadership Process" program (QLP). Every employee in the company had to take an initial short introductory course, and were supposed to follow up with a multi-month course that took something like 20% of your time for that period. Employees in the intensive course were supposed to complete some sort of project that would make a positive impact on the bottom line.... One of the more famous ones was putting up signs encouraging people to take the stairs instead of the elevator if they were only going up one or two floors in the Tower...
I escaped the long course with a well timed maternity leave, and deciding to leave the company before I could get snared into such a time sink. I often wondered what would have happened had I taken the course and suggested that one of the better ways of making a positive impact on the company would be to stop wasting employees' time in that program....
So one of my scars from Wang is that I shiver at the word "process."
I understand that it must be hard to manage engineers - but I struggle with applying process to art. If there are too many rules, the creative process gets thwarted. The key is to give engineers enough creative freedom in a problem to enjoy solving it, without so much freedom that there is chaos in a project. And that's a tough balance to find. Programs like QLP have a formulaic approach to how to do software/business. But when things get formulaic, they lose their art.
I'm told that when Dave Cutler interviewed people at DEC long ago, one of his trademark interview questions was whether software was a science or art. Had I ever interviewed with him, I would have said quite firmly that software was art.
Subscribe to:
Post Comments (Atom)
2 comments:
I've discovered that bad process = bureaucracy but good process = necessity (for larger efforts). The problem of course is that most process is bad process. :-)
I agree - process is definitely necessary, especially for larger efforts like Neil says. I think that it is critically important to develop the processes with their *audience* in mind, just as we do our software. So processes not only need to enable and assist good engineering, but also be human and humane (engineer as Michaelangelo).
One of the things I really like about Lotus is that it is committed to *human* software. In my job, I try very hard to apply the human aspect not only to our end product, but to how we build it, too :-)
Post a Comment