Tags: * against 4 reviews * penalize bad reviewers * incentivize reviewers * review load proportional to submissions * improve interface for load management * improve interface for reviewer assignment * make COIs private * for tracks * improve matching
"Aiming for 4 reviews per paper given late reviews exacerbates the issue -- other community members
end up reviewing significantly more to make up for late reviews. It would be more effective and fair
to address the issue directly, working to balance review loads, require reviewing from those submitting
more, or in some way penalizing those who submit late reviews or do not submit at all and/or rewarding
on time reviews. The current lack of infrastructure support and robustness is a significant detractor.
The form to mark reviewer unavailability works inconsistently. It takes several emails to get a response
when asking for review reassignment, in some cases taking up to 3 weeks (the bulk of the review cycle).
The email of the AE to contact is not always exposed to the reviewer. There is no way to add COIs for
all cycles that are privateSEMICOLON currently COIs are listed publicly, which may prevent some from
being listed. Lack of tracks has in my experience led to assignments that were outside my area of expertise.
The need for multiple reviews led to pushback from AEs when I asked to be reassigned given their distance
from what I am comfortable reviewing. I strongly support tracks, particularly until automatic assignment
can be improved. "