3 ms·
It would be nice to know who downvotes how often in relation to others (publicly or at the very least for the person herself) People could then compare their b
by dmichulke 11y ago
It would be nice to know who downvotes how often in relation to others (publicly or at the very least for the person herself)
People could then compare their behavior against the others and one could define a "too often" in terms of std-dev from the mean or a quantile, both within a specific time window (say last 365 days).
In this way one has at least a candidate set of abusers (for the "HN Mods") and everyone could compare his downvoting behavior and adjust.
Additionally, since it's a simple statistic, it should be easy to implement.
- InclinedPlane 11y agoPersonally I like the stackexchange model where downvotes have a cost.
- dmichulke 11y agoThat's also possible, maybe with increasing costs, but the problem is that it introduces state (so it's not as easy as a statistic to calculate). Constant costs could discourage downvoting and AFAIR, PG wanted to have his numbers being integers (so no rational costs either).
- Semiapies 11y agoThe anonymity of votes is something that acts against the worst aspects of groupthink and knee-jerk opinions in the HN crowd. I don't trying to literally penalize disagreement with the median voter here would help at all.
- dmichulke 11y agoThe idea would be not to penalize local quantity (votes or downvotes on a topic) but aggregated quantities. Chances are that if I downvote 10 times per day (with say an avg number of down votes of 2 per day) that I am doing something wrong or I five times as much as the average user. In both cases it's food for thought. Ratios of vote/downvote per user would probably also mean something. Finally, I'd prefer to have these statistics private for each user, so there is no way groupthink could get a hold.