> Until the code base becomes so intractable, such a dinosaur of brittle features and run-on code that nothing significant can be done without exhaustive debugging.
Yes. And that gets at the point of OP. "Let me replace subsystem A so we can free up B resources per year" is a big, if it works. To throw it on the backlog for "when we get time" is self destructive in the long run.
Yes. And that gets at the point of OP. "Let me replace subsystem A so we can free up B resources per year" is a big, if it works. To throw it on the backlog for "when we get time" is self destructive in the long run.