[dalke]

You liked the article so much you decided to submit it twice?

I read it after yesterday's link. I concluded that I really don't understand the urge to prospect for gold, at least not in the secret mines, ancient tales, and schlepping ore by pack and pack animal sort of way.

[dang]

> You liked the article so much you decided to submit it twice?

No, that was us. We asked the submitter to repost it as part of an ongoing experiment we're running to give good stories multiple chances at the front page. I wrote about this here: https://news.ycombinator.com/item?id=8790134.

[dalke]

The account submitted 9 links, including to high-quality print magazines (Outside Magazine, The Atlantic, Rolling Stone) within an hour. I wondered if the submitter had not remembered which had already been posted.

Rhetorically speaking, how do I distinguish between someone who was asked to re-post, and someone who is making multiple submissions for, say, self-promotional reason? This case didn't look like self-promotion, though others with a sparser posting history might. I tend to flag the more egregious self-promoters.

After reading your comments, I'll be more generous about upvoting /newest.

[dang]

> how do I distinguish between someone who was asked to re-post, and someone who is making multiple submissions for, say, self-promotional reason?

For the time being, you have to tell that by looking at the articles and the account's submission history. It's usually pretty easy to spot self-promotion. Keep in mind, though, that self-promotion per se isn't against the spirit of the site; uninteresting content is.

Also, we modified the FAQ to make it clear that when a story hasn't had much attention yet, a small number of reposts is ok. The point of all of this is try to mitigate the weaknesses of /newest as a mechanism for recognizing good stories.

[waterlesscloud]

One of the more frustrating aspects of submitting stories is finding out someone submitted the same story hours (or days) ago and it only garnered a few karma. In that situation, there's no hope of the story getting much attention even though multiple people found it interesting enough to submit. The karma is just spread out over too long a time for the algorithm to raise up the story.

I can imagine you need to defend against rings of people submitting the same spam story over time, but maybe if a few established users are submitting a story, its rank could be raised more than the karma/time ratio alone would indicate.

[dang]

I agree, and it's one reason why we want to make a better dupe detection system. We did look at changing that particular behavior but it turned out to be hard to do in isolation.

[weaksauce]

you could have a feature on the front page that takes 3 stories from the new feed an displays them at the bottom of the front page for users that are above some "wisdom karma threshold" similar to the downvote threshold. if you want to provide some checks and balances to the 3 stories you could include one that is flagkilled but is not distinguished as such to the user and see what the user does with that story.

You can also have stories with more text behind them as an indicator of quality. While long stories could be gamed, I think it's a decently helpful metric of quality as most spammers are lazy enough not to write a long article. not a perfect metric but something to consider.

[bmelton]

> Rhetorically speaking, how do I distinguish between someone who was asked to re-post, and someone who is making multiple submissions for, say, self-promotional reason?

Not trying to be crass, but why would we care? I upvote interesting content, and ignore uninteresting content.

Is there a reason that we should be concerned whether the submitter is benefitting in some way from his submission?

[dalke]

If the submission is "carpet cleaner in London", with a link to a carpet cleaning service in London, then I flag it without remorse. That's not something I want to see in HN. There are many other places for that type of post.

There are 232 links to questions in travel.stackexchange . Over 90% of them have an affiliation tracker id of /101 , although coming from different HN accounts. The questions include "Does condition 8115 mean you can't check your work email while travelling?" and "Can travel agents match automated ITA Travel Matrix prices?". These are only slightly less off topic than a carpet cleaning service advertisement.

For a while it seemed like I was seeing those links every day, to the point of annoyance. Just reading the travel-related title often meant that I knew that the topic would be uninteresting, and with the /101 at the end of the URL. My interpretation was that the operator of the account was more interested in the reputation gain on stackexchange rather than posting interesting links, and would switch accounts in order to get around HN flags.

There was another account a few years back which linked to an inane math trivia blog, perhaps several times per day. (I can't find examples now, but they were like "What's interesting about the number 120?") It wasn't really appropriate for HN, and not something I wanted to wade through every day. I assumed it was either to drive traffic, or a sort of way to spread a parent's love, metaphorically speaking.

For posts like that I usually point out in the comments that the topic isn't really that interesting, before bringing in the flag. I don't recall any responses, which leads me to think it's more to drive traffic than to find out what others might find interesting.

If submissions aren't flagged, and there are no mechanism put into place to remove off-topic or almost-off-topic posts, then there will be a lot more uninteresting content to ignore, like Photoshop Experts For Hire: Cheap!

Do you want that?

[ertdfgcb]

I like this idea, but if you decide to go ahead and make it a regular thing you should indicate that it's a repost somehow, or maybe have an "Editors Pick" tab with the chosen stories.

[dang]

We've bounced the first idea around for a while but I am loth to disrupt the look of the front page. Your second suggestion, though, sounds like it might solve the problem. Thanks!

I'd like to emphasize that our long-term intention is to have the community manage all of this.

[zhte415]

Be careful with Editors Pick. It can be great, or it could be the death of the site. Slashdot, for example. I like the way the community works, especially its diversity.