Do you mean âIt isnât clear to me why reducing value driftâeither in favour or against some EA ideasâis a good thing universallyâ?
Usuallyâand in my comment aboveâthe term value drift is used to refer to something more like drifting away from EA values as a whole, rather than shifting oneâs focus between different EA values/âideas/âcauses, which I think is obviously often good and probably more often good than bad (i.e., people probably update in good directions at least somewhat more often than in bad directions).
I think value drift away from EA values as a whole is usually bad, but even thatâs obviously not always bad (at least when you consider cases where a person was very focused on EA explicitly and then they move towards pursuing similar goals but with less focus on EA specifically). And indeed, I note above that âincreas[ing] your engagement with & retention in EA, & mitigat[ing] value drift âwould often be a good thing, though itâs unclear how often, and it may also often be a bad thingâ.
Do you mean âIt isnât clear to me why reducing value driftâeither in favour or against some EA ideasâis a good thing universallyâ?
Usuallyâand in my comment aboveâthe term value drift is used to refer to something more like drifting away from EA values as a whole, rather than shifting oneâs focus between different EA values/âideas/âcauses, which I think is obviously often good and probably more often good than bad (i.e., people probably update in good directions at least somewhat more often than in bad directions).
I think value drift away from EA values as a whole is usually bad, but even thatâs obviously not always bad (at least when you consider cases where a person was very focused on EA explicitly and then they move towards pursuing similar goals but with less focus on EA specifically). And indeed, I note above that âincreas[ing] your engagement with & retention in EA, & mitigat[ing] value drift âwould often be a good thing, though itâs unclear how often, and it may also often be a bad thingâ.