Innovation through constraint

by Sean McGrath

October 2, 2006 —

This morning I had a thought which was roughly speaking, fifty percent interesting and fifty percent horrifying. Actually no. That latter fifty percent should be split into twenty five percent for horror and twenty five percent for depression. Actually no, that twenty five percent for depression should be split...Maybe this would be easier if I just explained what the thought was? I think so. Then the complex mix of emotions I experienced should hopefully be self explanatory.

I was thinking about innovation, how it arises, where it comes from, how to bottle it and sell it, that sort of thing. I was looking back at my own career in computing to date and thinking of examples where I witnessed, or took part in, the production of something innovative.

A common thread runs through all my memories of innovation events. In all cases I can think of, there were serious engineering constraints afoot. For example, I remember a time when the system I was working on had to work in 1K of RAM. (Yes, you read that correctly.) I remember a time when a system had to fit a myriad of user options intuitively into an 80 column black and white screen. I remember a system...etc. etc. All the examples I can think of included this sort of constraint.

I see two ways to rationalize this. The first via philolinguistics. It could be that the very definition of innovation in the world of computing includes the idea of suceeding in the face of some constraints. After all, this is (for the most part) a real-world engineering discipline right? You know the drill: "time, cost, quality. Pick two." We eat and drink constraints in engineering. It is what makes engineering, engineering. Put another way, if there are no constraints, it ain't engineering. I can buy that.

The second way to rationalize it is the apparent truism that necessity is the mother of invention. Ingenuity appears to be boosted by the sort of strictures that plague our daily engineering lives. At least it is in the case of those engineers that posess that most precious of all engineering personaility traits - a "can do" problem-solving mentality.

So it would appear that constraints are good. This leads, of course, to the curious phenomenon whereby the very things that cause innovation to happen are the things that the innovators themselves moan about the most. I sure wish I had a supercomputer under my desk. I sure wish I had 4 gig of RAM, twice as many pixels, a faster hard disk, a quicker internet connection...

And now comes my horrible, tentative conclusion. The best way to foster innovation in an engineering crew is to give them a set of constraints. If they are not moaning about the constraints, keep adding them until they start moaning. Then you have created a wellspring for innovation to happen.

I can literally feel you cringing at this point. I know, I am cringing too. Let's see if we can nuance this sweeping conclusion into something that is both more palatable and more realistic. Thinking back again to my experiences over the years of true innovation and the constraints that were either causal or co-occurent with said innovation, I cannot think of a single example where *time* was the critical constraint.

I think this is an interesting fact. Generally, if you were to ask management to identify a constraint to be applied to an engineering project the four letter word that results is the word "time". "Tell them we need it by next week.". Groan. Cringe.

I have seen teams do great things under time pressure. I am not saying that deadlines are not a good thing. However, I do believe that if this is the over-riding constraint you place on a software engineering effort, the results will not be good.

Far better, if possible, to add constraints in other ways relating to operating environment, speed, ease of use, capacity etc. These are things engineering teams relish. These are interesting challenges. These are the really powerful innovation vectors. Time challenges are rarely truly relished in my experience.

And this is why my emotions are so mixed about all of this.

I think I have talked myself out of a faster computer, a bigger hard disk, a faster internet connection...

Groan.

ITworld.com