Amazing how the difference in the time of submission affects the voting so radically. Amazing. It's a "feature" of HN in so far as it's documented, but is it something that can be improved? The fact that something gets lots of votes at some stage shows that it's relevant and interesting. The fact that it didn't get votes first time around shows that it was missed.
How many relevant and interesting articles get missed because they are submitted once and sink without being noticed?
It’s a problem, but I’m not sure that it can be fixed.
One possible idea is to weight each vote using it’s time, instead of using the original time of the submission.
For example, if a submission was made 5 hous ago, and got 1 vote 2 hours ago and 3 votes 1 hour ago, then
Actual:
Value = (1+1+3)/(5+1)^1.4
Alternative:
Value = 1/(5+1)^1.4 + 1/(2+1)^1.4 + 3/(1+1)^1.4
This way an upvote of a old submission is not wasted.
The problem is that we may see Norvig’s spellchecker in the frontpage forever (It’s a good article, I upvoted it perhaps in a few repost, but it’s resubmitted every few months.)
How many relevant and interesting articles get missed because they are submitted once and sink without being noticed?
Is this a problem?
If so, can it be fixed?