Wealth is built up over generations via systems that are extensible and reusable by those that come after. If we changed the standard size of screws because we thought it looked better aesthetically, we'd instantly make massive swaths of previous work unusable.

Exploratory change is needed to figure out where improvements can be found, and I generally enjoy building and working with novel technology much more than trying to interface with older systems, but there should be some sort of purpose for the change. Older systems also need a certain amount of pruning and destruction to remove accumulated cruft, so it's not like it's always best to use what's been accumulated.

The correct balance is hard. Achieving that balance requires intentional effort, not arbitrary change or arbitrary preservation, if only in some sort of testing/evaluation phase (sometimes it does make sense to just arbitrarily tweak stuff and see if it ends up working better, but you need to actually do some sort of intentional verification to see if it's worth keeping).