This interview was provided courtesy of the Computer History Museum.
Women in Software
Humphrey: One thing before I forget. I keep saying “he” but there are lots of very good women in software. The team lead at Boeing was a woman, as was the technical lead. The problem is that not enough women go into software. We have that image of a tech-y culture, sort of where everybody is glued to a screen night and day, so lots of women don’t see it as a fun job. I have four daughters, and they like working with people, and I think a lot of women don’t see what a marvelous people-intensive field this is, and the TSP really helps with that. It’s all about people and how do you lead and motivate them and how do you negotiate and mediate. Women are often darned good at that and we need more of that talent in software.
Booch: Yeah, you’re right. Let's talk about issues of inertia. Why is it so hard to change organizations and change people?
The Change Problem
Humphrey: I've struggled with that. One of the things I keep running into--and I ran into it at IBM--when IBM was at the top of the heap, you literally couldn't get them to look at things differently. This was IBM. We were winning. Why do you want to change things? We're ahead. The conclusion I reached is, one, when organizations are at the top of the heap and they're enormously successful, it's extremely hard to get through. No one has to change--unless you run into someone who is a strategic thinker and is looking out for opportunities and ways to improve--and very few do.
There are very few who, like Art Anderson of IBM, say, “there's always room for improvement.” Most of them are basically working on what they're doing, trying to improve costs and profits and the competition, but they're rather narrowly focused on the paths they're on. So when they're winning, they don't have any pressure to change it at all.
And then the other
side of that is when you get into harder times, now they get into trouble. The
difficulty there is, they can't afford to change. They're
under financial pressure. They've got high-priority stuff they've got to do. They're
in survival mode. And so that's what we run into. To get people to change when
they don't have to is what we've got to deal with. When they really have to
change, it's extremely hard to get their attention. Now we occasionally run into
people like that--for instance, these foreign countries. They're moving, not
because they have to change, but because they see an opportunity, and they're
after a strategic change. Well, that takes vision, and there aren't a whole lot
of people with the kind of vision to do that. And that's what
And so these great
big organizations, it's very hard to move them. So the lack of pressure, the
lack of something requiring change, means it's very hard to do it. And when you
really have to do it, software improvement is not on the priority list, because
they've got to survive. Take the case of General Motors. What are you going to
do there? You get in there, you go into
Booch: So as you think of companies that are on the top of the heap now, tell me where you think Google might fit into this. What advice would you offer to them?
Humphrey: My reaction is that any company like that--certainly Google would be one of them--they really ought to sit down, look at their technology, what's going on, what works and what doesn't. Why can't we deliver products on schedule that work exactly the way we want them to work? I haven't looked at their stuff in great detail, but that's certainly true. You look at that at RIM and the BlackBerry. They've got all kinds of software they come out with and quite frankly, if you look at it, they've got a hell of a lot of defects in their software. It doesn’t hit you and I too much, but it costs them a bundle. That's true in a lot of these places, and I see enormous staffs of people fixing and responding and that sort of thing.
So looking around and
saying, “What's going on, and what are the technologies that really show
promise and how could we improve and grow our business?” and that sort of
thing, there ought to be at least some effort like that in these companies, and
it ought to be from the very top of the business. The senior executives ought
to be actually looking around. What are the choices? What are the alternatives?
Because winning companies, by and large, ultimately lose because they're
blindsided. Somebody comes at them that they didn't expect. They're just not
ready. All of a sudden, there they are and they're out in the wind. That's a
real problem. Now General Motors wasn't blindsided by
What people keep
forgetting is that the rate of change is accelerating. According to the Economist (Sept. 19, 2009), 24 firms
dropped off of the Fortune 500 list, on average, every year from 1956 to 1981.
From 1982 to 2006, that number jumped to 40 firms a year that got pushed out of
their top spot in
Booch: When you're on the top, people like to push you off of the top because they want to be on the top.
Oh yeah. Well, they're going to. They're going to work at it. There's enormous
motivation to do it. And the thing that's interesting is, in our business, in
the software-connected business, getting into this business is really
remarkably cheap. All the open source stuff, the very low
cost hardware, the low cost of communication. People in
it be fair to characterize it, your view relative to the
I'm not sure. I think
But I think
So we've got to stay
awake here. If we don't really stay out front and lead the world
technologically, and with our methods and our skills--and we've got the
opportunity to do it--but if we do what the