If software was simpler, we could afford to offer some formal guarantees of correctness. Model check protocols, verify pre and post conditions à la Dafny, etc.
There's too much change for the sake of change.
+1 to this. Like a lot of issues, I think the root is ideological, but this one in particular very clearly manifests organizationally.
The companies building everyday software are ever bigger— full of software engineers, designers and various kinds of managers who are asked to justify their generous salaries. At an individual level I'm sure there's all sorts of cases, but at a general level there's often almost no other option but to introduce change for the sake of change.
He said it's often just some new marketing exec wants to put something on their resume, and they have certain metrics that they target that don't necessarily align with long term profits of the company.
I'm sure software has a similar problem.
+1 to this. Like a lot of issues, I think the root is ideological, but this one in particular very clearly manifests organizationally.
The companies building everyday software are ever bigger— full of software engineers, designers and various kinds of managers who are asked to justify their generous salaries. At an individual level I'm sure there's all sorts of cases, but at a general level there's almost no other option but to introduce change for the sake of change.
At a general level, I believe there are other options - changes/features need to meet some level of usage or it is scrapped out of recognition that supporting all these features make bugs more likely, performance likely to degrade, increase difficulty of adding features, make the product more difficult to use, etc.