“Software correctness is a lot like flossing”

One issue is that most managers don’t actually care about quality: their focus is feature delivery and and reducing the backlog. Bugs that you haven’t found aren’t in the backlog so they don’t matter yet. https://www.hillelwayne.com/post/flossing/ ">https://www.hillelwayne.com/post/flos...

I mean, they _do_ care. Managers just aren’t paid specifically to tackle quality. They are almost always measured based on what they deliver, not whether what they deliver turns out to be buggy later on.

Incentives and priorities matter and in most software companies those are both lined up to promote feature delivery or iteration speed (if you’re “agile” in the “startup-agile” sense).

A coder who spends too much time on anything that slows down the pace will be quietly pushed to lay off by almost everybody. Not just because it slows down their own dev pace but because it makes everybody feel bad about letting incentives guide their own priorities.

Ultimately, the main reason why the various code quality methodologies never seem to get adopted is that bug-free software generally doesn’t make that much more money, certainly not in proportion to how much more work they are in the short term

In the long term you switch jobs

The software industry is quite nicely set up to minimise software quality: it& #39;s impossible to have a healthy career without regular job hopping, everybody& #39;s incentives focus on short term delivery over quality, nobody has to experience the consequences of their own actions.