I’d argue that research is to try to pass knowledge learned to someone else ( which is why professors teach and do the majority of research ). The first thing that goes when a deadline is coming up is... documentation... so I’d argue no, even a cobol program in a bank who no one knows about but is 100% stable has value.