[neuah]

I think using citation count as a metric for scientific importance by definition incorporates a ton of hindsight bias. Less cited papers are sometimes poorly executed research, but often they also just represent dead ends and uninteresting avenues that would've eventually needed to be explored. they don't form the foundation for lots of new research, but collectively they are important for directing things in the right direction. In that sense the citation count undervalues their collective worth. I don't buy the idea that we could get rid of the vast majority of them and the high impact work would be done just the same.