The reasons for this are actually kinda fascinating (at least in a “root cause analysis of an engineering disaster” way).
The way performance evaluation works at Google very heavily takes into account what things your (and if you are a manager, your team) have delivered recently. Maintenance doesn’t really count, so if you want to get a good performance review (and promotions, not be first to go when it’s redundancy time etc) you have to be doing “new”. You can get away with ongoing evolutionary development of a thing if it’s a product or if it’s visible and important, but the unsexy, unglamorous things like looking after small corners of the developer ecosystem that (until very recently) weren’t strategic priorities is a very good way to kill your career.
The reasons for this are actually kinda fascinating (at least in a “root cause analysis of an engineering disaster” way).
The way performance evaluation works at Google very heavily takes into account what things your (and if you are a manager, your team) have delivered recently. Maintenance doesn’t really count, so if you want to get a good performance review (and promotions, not be first to go when it’s redundancy time etc) you have to be doing “new”. You can get away with ongoing evolutionary development of a thing if it’s a product or if it’s visible and important, but the unsexy, unglamorous things like looking after small corners of the developer ecosystem that (until very recently) weren’t strategic priorities is a very good way to kill your career.