The karma of the submitter has no effect on how a story is ranked. Nor I suspect does it affect whether people upvote the story, except when the submitter is one of maybe 100 HN users who are well enough known that their username as submitter makes people pay attention to the story. Plus if we noticed anyone consistently rewriting titles we'd ask them to stop and eventually ban them if they didn't.
Ok, that's a good point. I assumed poster quality mattered because posts which are highly rated are mostly from posters with high karam. Confusing cause with effect probably.
That said, the quality of the titles matters a lot, and getting some up votes quickly also seems to mater a lot. Users could be bad actors and intentionally do bad titles. They also can try to avoid collect early votes through timing their posts during the weekday morning west coast time. I suspect that hacker news traffic peaks wednesday at lunch time PST.
If you were trying to hose a competitor by posting e.g. their blog posts at a slow time like 4 am pacific, they'd have to help you out by publishing their blog posts at that time. In my experience no one does that. Companies wait to publish their posts at good times, and then the links are quickly submitted to HN, either by the company or someone else. So a bad guy would rarely have an opportunity to submit an as-yet unsubmitted link at a slow time.
Fred Wilson posts then. I and other well-meaning HNers on the east coast sometimes accidentally kill good posts of his by submitting them at a bad time.
If you can think of a way to make gravity a function of activity rather than time without making it easy to game, then this problem will go away.
Yeah, i think blog posts aren't such a big issue, since often important news gets reblogged and the posts come from there as well. It's more likely to be a problem with the initial announcement of a url as being life. Say, AirBnB.com changes their name to rent.com, for example.
It'd be interesting to see what the split is between people upvoting under 'new' versus duplicate story submissions. I'm betting that the dupe submissions are much more frequent since highly on-topic posts from lesser known sources will often just vanish entirely.
For instance, I submitted a story about WalmartLabs from mercurynews that didn't go anywhere. A few hours later, a new submission on it from a different source zoomed up to the front page within minutes.
Honestly, I think this is just the nature of the social beast that we are working with. Whether things get "picked up" or not are often based on a lot of arbitrary, impossible-to-track variables such as "how many people interested in WalmartLabs happened to be looking at /newest within 20 minutes of your post". While there are definitely instances of causality that can even be extrapolated into trends, luck is always a key factor in these services
The timing submissions has a huge effect. Posting at a busy time makes a story far more likely to get buried. Posting at 6:oo am Saturday morning is more likely to provide enough time for a story to gain traction than 10am Monday.
What I thought this was going to be about was a killfile system for blacklisting websites for users who are logged in and are interesting in that sort of thing. Kind of like Slashdot's system that let you blacklist editors you didn't like (like Jon Katz).
Is this the sort of thing that you would ever consider implementing?
I shy away from this sort of thing because it fragments the community. I don't want HN to turn into a mere customized feed reader, where each user sees something different.
