2 ms·
It seems that once something scrolls off the first page or two of 'new', or sits too long at a low count, there's no chance of it catching a flurry of new votes
by gojomo 17y ago
It seems that once something scrolls off the first page or two of 'new', or sits too long at a low count, there's no chance of it catching a flurry of new votes -- and even if it did, I don't think the rating function considers momentum enough for it to reach the front page.
(That is: if a 3-day-old article that was initially overlooked gets 4 votes today in rapid succession, while a brand-new article gets 3 votes today in the same period, the older article is still punished by averaging its new votes over the days when it wasn't even really being reviewed. At least that's how it seems to work.)
There needs to be a better way for items to get a second wind when the right readers show up.
A possibility would be to replace 'new' with 'newly worth a look'. Essentially, this would be the combination of new items (which have an implicit upvote from their submitter) mixed with older items that just got a fresh upvote, minus items that are already on the front page. That is, every upvote would bump an item back to the top 'worth a look', even if not to the 'top' rankings, so it has a chance again of a self-reinforcing rise.
Probably, once a user has upvoted or 'passed' (hypothetical new option) on an item, it would no longer appear in 'newly worth a look'.