Of course you should’ve known Heinz and Kraft would merge. Of course you should’ve known the Shanghai market would tank. Of course you should’ve known Oliver Stone would make the Snowden movie.
But, of course, you didn’t. None of us did. It just made perfect sense to us after we read the news.
That’s why it’s absolutely imperative that we all automate our code lifecycles.
An Uncertain World
Much of the narrative surrounding DevOps and continuous software delivery makes it sound as if C-suite leaders are omniscient strategists—and IT is the only thing standing in the way between them and market domination. According to this narrative, IT must move faster so the business can execute its awesome plans without technological friction.
This narrative is occasionally accurate. But the deeper truth is that business leaders are human beings operating with incomplete and/or inaccurate information. So they make lots of decisions that circumstances require them to make.
Of course, with Big Data and advanced analytics, we make much better predictions than ever before. We figure out that the customer who just watched “The Big Lebowski” will probably like “Fargo.” And we can stagger the lights on Eighth Avenue more precisely to get Jersey people out of town faster.
But we still can’t predict what Buffett or Bezos will do next. Nor can we be sure how next year’s Presidential election will affect re-admission penalties for healthcare providers. So we have to be ready to change course. Quickly.
Which is exactly what code lifecycle automation lets us do.
Baby, It’s Code Outside
Code, after all, is increasingly how we do things. So if we’re doing things one way and tomorrow’s news requires that we do them some other way, we have to change our code.
But you can’t make that change quickly if you’re stuck with a slow, manual process for capturing new requirements, mapping fulfillment of those requirements to the right team, putting the code they write through the right paces, and releasing it into production.
In fact, given the volatility of digital markets, code that takes too long to release can be obsolete before it gets there.
Or, as may more often be the case, a slow software lifecycle won’t let your LOB executives find out how wrong they were soon enough for them to change course before they do too much damage.
Faster, Smaller Mistakes
Your company may be an exception. Your executives may never get blindsided by a shift in market dynamics or an unexpected regulatory mandate. In that case, a small dose of Agile and a few DevOps best practices will be sufficient to make sure you deliver market-aligned software.
But the rest of us need to make sure we can rescue our business from code that efficiently, reliably, and securely does the terribly wrong thing we told it to do. And then change it again when we read tomorrow’s news.
So we are going to automate.
