Analytics

Aside from her other sterling qualities (and she has a lot), one thing that my wife does that nobody else does: she has thea-conversations with me.

The name comes from a comment that my wife made to her sister about how we communicated, shortly after we were married. She said that we liked to talk about a range of things, and she said 'Doesn't your husband like to have theoretical conversations?' Her sister laughed and said that they did just fine not having theological, theoretical, or thea-anything conversations. For us -- for me -- they're critical. I have observed, and freely admit, that I get grumpy if I am deprived of my wife's sheer presence. But what puts the shine on it is when we talk about theoretical concepts. Now, these aren't deep, philosophical conversations -- in fact, sometimes they're nothing more than diatribes about how incredibly stupid the management of our various organizations, or sometimes the government, can be. But other times, we really do try to come up with some kind of explanation about why what they're trying to do is so stupid -- or, at least, what it is that they're trying to do with their stupid policies and plans. Because we do understand that even though what they're doing is stupid, the people themselves are not -- in fact, they're usually quite bright. So we try to think of how we'd do it differenty -- or at least, why it is that they're doing what they're doing the way that they're doing it. Why did they chose this way?

This evening, we talked a little bit about organizational quality standards, and how they just do not work with software companies. And I came up with a succinct way of putting it, which I will now inflict upon you.

You can get a small group of people to implement a complex procedure, whether they believe in it or not. You can get a broad group of people to implement a simple procedure, whether they believe in it or not. But you cannot get a broad group of people to implement a complex procedure unless they really believe in it.

Such is the case for organizational quality. In small doses, small groups, it can be seen to be effective. In large doses, its not possible. There are so many players, so many variables, that the only way to understand and validate the process is to track it at a global level, and to put a lot of effort (people, mantime) into it. Think global, act local. Keep the big picture, understand what the overarching goal is, but track and implement at the local level, keeping an eye on making each piece demonstratably better. Not just different, not just tracked within an inch of its life, but better. Understand how the pieces fit together, and how they affect the global goal. Otherwise, you're just assuming it works (a little bit works, a lot must work). You can't actually prove it. You can show is that a lot of effort goes into implementing standards (for which read: working on passing the audit) at the lower levels. You just can't prove that it bought you anything other than that. To prove it, you need to put lots of effort into it. Lots of non-product-producing effort. And no one is going to do that, are they? Not in our lean, mean world.

I really enjoy thinking and talking about this kind of thing. I'd like even more making it work in the real world.