“That’s a solved problem. So nobody knows how to solve it anymore.”

Once a problem is deemed “solved” interest in the problem plummets. “Solved” problems may not be fully solved, but sufficiently solved that the problem is no longer fashionable. Practical issues remain, but interest moves elsewhere.

The software written for the problem slowly decays. Of course the software itself doesn’t change, but the world around the software changes, and the effect is the same. More on that here.

I spent a couple hours this morning trying to build several different software packages that supposedly fully solve a problem that was fashionable not that long ago. But it doesn’t take long for links to break, for software to accumulate breaking changes, etc.

I’ve often gotten consulting jobs to re-solve a “solved” problem. The problem have have been settled for most cases, but my client is very interested in what others have deemed an unimportant corner case. One man’s corner case is the center of another man’s world.

The post Solved problems becoming unsolved first appeared on John D. Cook.