You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Instead of discarding non-matched sites entirely (which could exclude useful results too), I tried downranking them all with a
$downrank
statement. This works for the $discard instruction, as shown in some of the example goggles and mentioned here. But based on some experimentation, this doesn't really seem to work when it's the downrank instruction being used this way. As far as I can tell, this instruction is just being silently ignored.
Is there a generic downrank feature that I'm just not seeing the effect of in my tests? Or is it not implemented? If it's not yet implemented, please consider this a feature request for a catch-all generic downrank.
The text was updated successfully, but these errors were encountered:
On this goggle, I tried the generic $downrank goggle, and the results were the same whether the line was:
$downrank
$downrank=10
! $downrank (i.e. no downranking)
I tried all these by editing, resubmitting, and doing the search again (the keyword I was searching for was 'hurdle model'). And in every case, the results were exactly identical to each other. Even if it doesn't surface new results, I'd at least expect the already existing boosted results in the search to get moved higher relative to the downranked ones, if there was downranking going on, but there's no change in their positions.
Instead of discarding non-matched sites entirely (which could exclude useful results too), I tried downranking them all with a
statement. This works for the
$discard
instruction, as shown in some of the example goggles and mentioned here. But based on some experimentation, this doesn't really seem to work when it's thedownrank
instruction being used this way. As far as I can tell, this instruction is just being silently ignored.Is there a generic
downrank
feature that I'm just not seeing the effect of in my tests? Or is it not implemented? If it's not yet implemented, please consider this a feature request for a catch-all genericdownrank
.The text was updated successfully, but these errors were encountered: