3 ms·
> We can only guess why users flag things. I'm sure you've considered this feature, but since flagging is such a heavyweight (in terms of ranking impact) activ
by ryandrake 2y ago
> We can only guess why users flag things.
I'm sure you've considered this feature, but since flagging is such a heavyweight (in terms of ranking impact) activity, shouldn't a flagger need to at least articulate (via text or a drop-down menu) why they are flagging it? Even vague choices like "Site guideline violation," "Spam," "Astroturfing," "Political flame bait" might offer some of the missing insight.
- dang 2y agoI've always resisted that, but could be persuaded. Not sure we'd get the true reasons though.
- ajb 2y agoWhat are your concerns with that? If it's overhead for the flagger, they could be presented with a single multi choice question, eg: What most deserves flagging: a)item title b) item content c) hn discussion d) something else With different users presented with different questions to build up a picture. Flags could be recorded even if the question isn't answered, but with you reserving the right to weight ones without answers less.
- dang 2y agoMainly the bureaucratic nature of it. Preserving HN's minimalism has been a priority because it's the kind of thing that's so easily lost; like a frog boiling itself.
- ryandrake 2y agoYea, I get the hesitance to invest in this. Nobody is going to select/admit reasons like "Because flagging is a mega-downvote and I don't like this story getting attention," even if it's the real reason. We don't know how accurate or actionable Slashdot's drop-down choice moderation was, either. Even if people are honest about their intentions, what is actionable? Not much I guess.