> you can only compare performance across a cohort I can't accept this premise. Optimally the persons responsible for managing engineering staff should be abl

> you can only compare performance across a cohort I can't accept this premise. Optimally the persons responsible for managing engineering staff should be able to independently determine whether the work being produced was of sufficient quality or not regardless of the cohort. At issue here, I believe is that this is a difficult thing to cultivate consistently in many corporations and so there's some desire to create standardized metrics for performance against which a cohort is measured. Regardless, most large tech firms have some kind of well defined rubric against which the engineers are measured.