In a way, one could argue that there is no such thing as "technical debt". This term usually denotes technical decisions which lead to functional software at a

In a way, one could argue that there is no such thing as "technical debt". This term usually denotes technical decisions which lead to functional software at a cost of now difficulty in future development, debugging, performance etc. But that cost only matters only when it's actually collected, e.g. when later development is actually difficult because of those decisions. Which might never come to pass, and activities to"improve" the codebase are wasted unless one is 100% certain that it will happen. It might be better to consider that practice "fixing non-functional bugs".