Ever tried to get a teenager to do more chores around the home? For those without this joy in their lives, I’ll let you in on a secret - it goes down like a bucket of sick.
You can sometimes cajole them, sometimes bribe them and even threaten them (We’ll take away your laptop!) But at best, this has mixed results. You’ll often get an uptick in throughput - the rubbish & recycling will exit the apartment more often. But quality will suffer, the cacophony of bins being banged, incessant grumbling and milk cartons being scattered about will lead to you questioning many of your life choices.
This is often the case in life generally and software development in particular. The old adage “Faster, better, cheaper - Pick any 2” still holds true. Interestingly this isn’t always bad news for those in the business. More often it's a problem for customers, be they other teams or actual customers.
For example, if you provide the means to produce faster - more people may be buying your tool. So in software development this might be a product like Claude Code, Cursor or Github Copilot. Making initial ‘coding’ quicker.
If we assume the business needs to maintain a certain level of quality to maintain market share or avoid prosecution & regulator scrutiny, then that increased ‘flow’ of features will need to be tested and fixed. Again a tool could help with this, but we need to be aware of the common method problem. The same approach used to create the code (with certain problems and biases) will likely test the code with the same biases and problem-blindness.
The same disgruntled teenager that can now “do” more will be less able to see the poor job they’ve done. Mind you, ChatGPT is always so much more polite than the average teenager. Maybe an AI dominated world would at least be more polite.
So ‘raising’ good software is becoming a more important job. Questioning whether the slick new app is reliable, accurate and performant will become a more involved, more challenging and more interesting job. As observers have noted, advances, many of them valuable, have a jagged edge. There are frequent gaps in the quality of applications and systems at the very time more people are able to produce more of them.
So when your bank replaces that old system that pays your mortgage every month, with a new application heavily built using AI assisted coding tools, also hope your bank hires skilled test engineers to investigate how well the systems actually work.
Comments
Post a Comment