When “good enough” just doesn’t cut it

My father’s younger brother once was a decently ranked squash player back in India.  As the story goes, someone once asked him why he didn’t continue with squash and try to get to #1.  His response was, and I paraphrase from my memory, “It’s not that hard to get from #100 to #10, but to get to #1 from #10 is very hard”.  In other words, he was happy being good enough. That’s true across the board in many fields for many people.  On the flip side you have people like Michael Jordan, who was never satisfied with being anything other than the best.  We all have to make our choices in life, whether to be good enough, or be the best, or at least make the effort to be the best.  There is nothing wrong with being good enough in most cases, but it’s been something I’ve been thinking about with regards to software development, science, and business.  Sometimes “good enough” just doesn’t cut it.  I’ve been in places where the bar for excellence is one of “good enough”, and you end up with stuff that works, and is even useful, but seems to lack that extra something.  On the other side, I’ve been in places where good enough just didn’t cut it.  There was a sense of pride in pushing the envelope, trying to better yourself.

Where this really frustrates me is scientific software, most of which falls into the “good enough” category.  I’ve very rarely seen software or software systems where one gets the impression that the “good enough just doesn’t cut it” approach was taken.  Perhaps it’s the nature of the beast.  Scientific software serves a purpose, i.e. getting a scientific result, and is not an end in itself, or at least not traditionally perceived as such.  I hope this changes, and I think it is.

Good enough just shouldn’t be that sometime.

Posted via email from Flashing Neurons!!!