[rw]

The no-one-noticing-good-submissions is seemingly a huge problem here and I think working on fixing that will make HN much better. I emailed PG a few months ago with some ideas I had about how to quantitatively measure submission and discussion quality, but he did not respond. I'll forgive him though, he's been making babies and startups :)

[fub4r]

Most submissions are probably good but I guess only a small percentage of users click the new link, and then they can only read a small percentage of submissions and up vote only an even smaller percentage of those. It would be easy for a great submission to fall through the cracks.

[sirsean]

That's definitely true, and I'm sure it happens a lot.

What's weird to me, then, is the frequency of crappy submissions getting enough upvotes to hit the front page while the inevitable good submissions slip through. (This, obviously, is bigger than my own submissions, which I'm not particularly interested in seeing on the front page given that I've already seen them -- I'd like to see all the good submissions, and less of the crap.)

[lacker]

Maybe the algorithm should guarantee every new submission gets at least X front-page views for some value of X.