Colloquium at Stanford
The Unfinished Revolution


Memorandum

Date: Sat, 26 Feb 2000 14:39:14 -0800

From:   Eric Armstrong
eric.armstrong@eng.sun.com
Reply-To: unrev-II@onelist.com

To:     unrev-II@onelist.com

Subject:   Use of Case Studies in a DKR

In session 8, Rob Swigart gave a fascination account of how he produces small stories (vignettes) from future scenarios. (As a writer, I was enthralled by his concept of protagonist/antagonist plus guardian/tempter dynamics for a story.)

I confess that I couldn't see how to apply that notion to a DKR, however. The problem with "war stories" and case studies, in my mind, as always been that they are subject to so *many* different interpretations.

Suppose there were a story in the repository about how Ford made better cars, for examples. How on earth would I ever find it, if I was searching for ways to improve my software design process?

We mare almost certainly decades away from having the capacity to automatically index a story under all the many headings that might be appropriate. The story about improving cars, for example, might be about improving designs, overcoming management resistance, achieving worker acceptance, getting customers involved, moving to standardized interfaces, none of the above, all of the above, or any combination of the above and many more.

How could we do that?

Later in the Q/A session, though, Philip Gust leaped to the rescue. He pointed out that DKR contributors could use stories as *evidence* for generalizations they make. War stories and case studies would then provide a basis for supporting or disputing ideas.

That struck me as the perfect answer. It goes back to our local lawyer's (Pete Jacobs, right?) assertion that a DKR would be a valuable adjunct to the adjudicative decision making process, where case studies are used to support a variety of positions, and a decision is reached which adds to the body of case knowledge.

This was a big moment for me, because for the first time in my life I found value in case studies. It struck me that the problem I had with case studies (the lack of clearly defined general rules) was the same problem that people generally have with "outlines"...

In school, most people hated outlines because "they don't think that way". The problem was not with outlines, but with the tools they had at their disposal. Having to write out a structure on paper, and being unable to change it easily, required a tremendously organized mind.

But in a more "fluid" medium like an outline program, outlines are a real joy. The ability to easily rearrange things as your understanding grows lets you *find* a good organization, rather than requiring you to perceive it in advance.

Stories are like that. A DKR system like the one Philip alluded to would allow you to *discover* good generalizations as you went along. Both the story base and the index of generalizations would grow over time.

Such a system would evolve into something very valuable, indeed.

Sincerely,


Eric Armstrong
eric.armstrong@eng.sun.com