HN Explorer

Ask HN: Herd effects on HN?

4 points by christofd ↗ HN
I'm seeing a pattern here (I gotta start timing this):

Many comments start out with an even number of slowly incoming VOTES, and once one or few of those comments hit a critical mass in up-votes they gain rank much quicker: there is noticeable ACCELERATION in the uptake. I suspect we could have a herd effect going on, where people are not really reading the comments, but confidently up-voting what other people have up-voted.

Also, many stories sit around for a while (at the top), before anybody comments. Then traffic in COMMENTS trickles in slowly, and after a critical mass of comments, again, the uptake ACCELERATES. Here I suspect people are waiting for other people to claim the space.

Any thoughts?

5 comments

[ 2.9 ms ] story [ 23.7 ms ] thread
harpastum
I think this is simply a function of the perceived value of a story. If it doesn't have many upvotes, readers are unlikely to click on it, and therefore unlikely to upvote themselves.

Likewise, if there has been no commentary on an article, a reader is much more likely to comment on the article's site or not at all, rather than try to force a discussion.

An interesting question branching off of this is the percentage of upvotes related to the number of votes for a story. That is, how does the ratio of visits to upvotes change as a story gains traction. In my own personal experience, I don't think there's much of a difference. The only way to know for sure is to get the stats from PG.

christofd
To clarify: I don't mean upvotes on the story but upvotes on comments (which is the 1. point).

>Likewise, if there has been no commentary on an article, a >reader is much more likely to comment on the article's site >or not at all

Unfortunately, this doesn't make sense logically.

Comparing the speed of upvotes on the story vs. speed of upvotes on comments over time is interesting.

HalcyonMuse
I, for one, do not notice the number of votes next to a story. As long as it's in the top 150, if it looks interesting, I'll read it, if not, I'll skip it.
christofd
Again, I meant upvotes on comments, not on stories. I should have been more clear. Anyway, there are probably social effects on the voting of stories going on - I'm just not that clear on the other algorithms in place that HN uses to determine the rank of stories (that's why I left this out for now).
bendtheblock
I think this is probably exacerbated by the comments with most votes being at the top of the page (but that's also the value of the system). For stories with lots of comments, I'm sure many people only read the top few and are hence more likely to further vote up an already highly voted comment.

I guess the solution is to find someway of giving emphasis to new comments that have not yet had chance to be voted up. I like the simplicity of the current voting system though.