In my last column (Part 9 of this series), I suggested that any time an article series gets close to double digits, the author is either pretentious enough to think his readers are actually interested in that subject that many times in a row, or he’s just too boneheaded to come up with a new topic. Readers are left to speculate as to which of those two cases are at work here.
Nevertheless, commentary from the field has made it clear that despite the dangers of traversing into double-digit territory, one more article on multiparadigmatic design seems necessary to try to tie all the individual elements together—to demonstrate how to use each of these different paradigms and choose among them for a real-world problem. “Real-world,” for our purposes, means non-trivial enough to hint at how the approach might be useful for problems that aren’t as simple as those chosen to solve in a magazine article.
Trying to create such a problem turns out to be more difficult than it might appear; either the idea is too complicated, with too many distractions in it to get a clear view of the solutions used; or the idea is too simple, allowing for too little variation in its implementation to illustrate how the different paradigms might each be used to solve it. Fortunately, to get started we can make use of some work that has already been done—in the form of Dave Thomas’ “Code Katas.”
No comments:
Post a Comment