June 10th, 2007

planet

the three phases

The act of making a whole new airplane can, from a business/engineering perspective, be broken into three parts. "Advanced Design" is what we use to refer to the really big fuzzy weird part... you're doing research, coming up with ideas, talking with management and customers a lot and developing concepts. This is fun for a while, but if you're really in advanced design, 90% of your ideas end up in a file cabinet somewhere without being a real airplane. Management gets bored, pulls ideas out, has you work on them, then has you put them back.

Then there's Experimental, which is where I am. Experimental is what happens when upper management takes a file from the Advanced Design world and hands it "over the wall", so to speak, assigns an official program manager, approves a plan to make an airplane, and has an airplane made. It's a big deal to make it to this point. We work "go" programs, which means that unless something goes wrong (a major supplier drops out, the economy tanks) this will definitely be an airplane. We're given hard-set dates... next year you will have a flying airplane. the year after that, you will certify. three months after that you will deliver to the first customer. and we get there.

After the first few customer deliveries, you're officially in Sustaining. Some people love working sustaining. Most of us don't. The phrase "rotting away on sustaining" is common. It's where we start the newbies who don't know what an airplane is. You work little projects here and there, and fix everybody's problems. The due dates are no longer looked at in terms of years, but weeks. You hope you can hand off to somebody else and go work experimental again, where you have more flexibility if something needs changed, there's less repitition, more working with lots of different groups, that sort of thing.

I didn't know about all this in college because I think school is all Experimental. There's definitely no sustaining. Once you turn in a paper, if it's sort of mostly okay, the teacher gives you a passing grade and you never worry about it again. There are no customers bugging you to make it better. You're not releasing patches or revisions. It's just there. so we get out into the real world and nobody knows how to handle sustaining... the act of planning to make something better, even though the exciting turnpoints and due dates are past.

and the worst airplane to work on is the one where they don't really plan for sustaining. they stop paying attention and thinking about strategy after the big dates. they try to avoid responsibility and fail to listen to lessons learned from other sustaining groups. they tell people they can go to another program, then string them along for months or years. they let the same experimental people try to learn how to handle sustaining, without entertaining any new ideas about what needs to be changed or admitting that we're in a different phase. They don't embrace the idea that sustaining is just as important as everything we did before the big exciting milestone dates, they underthink it, they don't look at it as its own science.

and for the record, this is basically what I think went wrong with American Iraq war policy.