4 ms·
I agree, i would be very interested to see the effect of hiding usernames (an ID that lasted only for the thread), I bet it would distribute points significantl
by polshaw 14y ago
I agree, i would be very interested to see the effect of hiding usernames (an ID that lasted only for the thread), I bet it would distribute points significantly differently.
There is also the timing aspect however; habitual commenter are more likely to be in early to a thread.
- stephengillie 14y agoAnd it's much easier to get upvotes when you're one of the first posters in a comments thread -- others are more likely to upvote you instead of writing their own comment. If you see upvotes (U) as a product of the number of people viewing HN at a specific point in time (V) and the percentage of people who will upvote the post (Q) multiplied by the number of points in time (T) during which the post is viewable, we find U=VQT. We can increase U by posting when V is highest, by posting comments when the most people are visiting HN. We can increase U by increasing the % of people who like the post, which usually means writing a high-quality comment. We can increase U by increasing T, the amount of time the post is available, simply by posting it earlier.
- brudgers 14y agoIt's also easier to get downvotes as the first poster. Thus comment quality is still the long term determinate. Particularly, now that snark and meanness are less acceptable, more thoughtful comments are required.