Tree Parable

The Parable of Inventing a NewTree?

I assembled a team of several different specialized programmers to take on a real break-through project. I had an expert in one of our target OSs, a general C programmer who was also a domain expert and a high level compiler writer. I made up this parable about how we were going to break the rules of convention:

A molecular biologist hires an architect and a carpenter to add a room onto his house and goes away for a week to a scientific conference. She returns to find the a half built room with the builder and architect screaming at each other at the tops of their lungs. The molecular biologist asks the architect what is wrong. The architect explains that the design requires that oak must be used through-out to maintain the proper texture. The molecular biologist then asks the builder what the problem is. The builder explains that there is no way to nail such hard wood in the way the architect designed it. The molecular biologist responded that there was no problem. She would just design a NewTree!

It is amazing to see what happens when you mix bright engineers from different disciplines together. One of them is always inventing a NewTree?, at least a new tree from one of the other's world view.

See also: ParablePattern

-- ScottElliott

The moral is unclear. Since literally designing a NewTree to meet arbitrary requirements is still far beyond the state of the art, it seems that the intended moral is that the suggestion is ludicrous, and so the moral is that, when you mix really different kinds of people together, at least one of them will come up with a stupid outlandish idea.

However, since the counterpart of the analogy is a compiler person, the counterpart of the NewTree would be a new language, and creating a new small domain-specific language is actually often an extremely good approach, even the best approach -- especially when you've got a person with the correct background to do it right.

You might have meant it either way. This parable needs disambiguation -- and if it disambiguates in the direction of "creating a new language is ridiculous", then I strongly disagree; ItDepends on the details of the project and its context.

The fact that it was a "real breakthrough project" seems to not play a part in the parable, yet one would think that it should.

My point, if some what befuddled, was to open everyone up to accepting paradigm changing possibilities from the other team members and to encourage the engineers to feel safe offering those ideas. There are times when what is in the text book will not get you there. Companies are so risk adverse that it can take some real effort to get engineers to get engineers into a risk taking frame of mind.

I guess the generic geek story would have been the KobayashiMaru no-win scenario. When the engineers and sr. management all agree that the number say it can't be done, then the team needs to "change the condition of the test," in the language of the movie. -- ScottElliott

Strongly agree. I've found ways to do so many times myself, and those tend to be the things I remember most proudly in retrospect; it contrasts rather nicely with just giving up on something as "impossible".