HN needs to post new stories on the front page to give them a chance | Hacker News Reader

HN needs to post new stories on the front page to give them a chance

blog.styleguidance.com

42 pointsvaksel35 comments

It really is a problem. Right now we're at a moment in HN history in which is possible to get a submission on the FP by dumb luck, but it's better to have a couple friends take a look and vote right away (not endorsing spamming--just saying that drawing attention to your submission during that critical few minute window makes a big difference).

As HN grows, however, this will change: it will be absolutely required for you to get your friends to upvote in that critical window to have any chance of hitting the FP. That's not a good road to go down, so I think it's definitely time to adjust the way it works.

I've created a poll to find out how common it is to ask friends to upvote submissions. If it really is common behavior, then maybe something does need to be done about this.

http://news.ycombinator.com/item?id=1075931

There's another option that should be added: People with multiple accounts.

I've submitted 18 articles, and 9 have made it to the homepage (http://news.ycombinator.com/submitted?id=ironkeith). I am certainly not a power user, and I don't have a circle of friends who quickly upvote my submissions to game the system. So far as I can tell, people use the new page, and interesting articles make the home page.

That said, I also think it's really important to ensure that the title you submit is properly phrased. The articles that have received the most votes have inevitably been the ones I put some time into rewording to appeal to HN users.

the stuff of yours that got voted up, is what you found online at popular sites. The reason it bypasses the death, is that other people try to submit it too, at which point it gets 1 extra upvote.

I'm more speaking for the people who write their own stuff(sivers, asmartbear etc). Who spend 2-3 hours putting an interesting article together, only to have it die in the new bin.

For example, I get asked all the time how I got my site to 35K visits in 2 months. And I'd love to share. But I always put it off because I know there is a 90% chance it'll never get off the new page.

Why should I spend 3-5 hours putting that article together with graphs and data, if I know for a fact that there is 99% chance that it'll die. I got a site to run, I can't throw away my time like that.

In fact the only articles of mine that actually made it to the front page, were those which I linked to from another relevant discussion. Those got 40-60 upvotes. But if I didn't link them from a relevant thread, they'd be stuck in the new page like all the other submissions.

For me it's a mixed bag, some of the stuff I write that I think will be for a very small set of people does surprisingly well, other stuff that I think is more important does much worse.

But just like making music, I write what I write because it helps me to put down my thoughts, how many others read it, find it useful and/or comment on it is really not that big a deal to me, though I'm always very happy when there is some kind of interaction around something I write. Usually I learn as much from the feedback and the comments as I did from doing whatever it was that I wrote about in the first place.

oh yeah, it's def something you need to do for yourself...but it's all about priorities.

If I knew the article would have a chance on HN, I'd probably put a little bit more priority on it.

That's true, I supposed I'd never considered what it would be like from a publisher's perspective. Does HN push enough traffic to make it worthwhile to create content directly targeted at its users? Knowing what to expect for traffic would help me evaluate if its worth it to invest 2-3 hours at a x% chance of hitting the homepage.

So far as the value of 'x%' goes, there are certainly ways you could game the current system in your favor: linkbait the title to appeal to a very specific demo, get a few friends lined up for a quick upvote... it doesn't seem like it would be too much trouble. I often see very recently submitted articles with 3-4 upvotes at or near the top of the homepage. If the only reason you're writing is for traffic, there certainly appears to be a lot of opportunity. Am I wrong?

it's not really about points or traffic. Mostly it's just knowing that the people who asked me to do the article actually get to see it.

If it never makes it to the front page, where 99% of the people view the stories, then all that effort was for naught. And I might as well save my time.

i wish the big stories would get removed sooner so that more content could filter in. maybe it could just be personal: stories i've seen already are removed.

either way, i feel like stories get big and hang around even though they're getting more and more upvotes simply for being on the front page. when i page down it's just the big stories from a couple days ago that i already read.

if more new content filtered in that would be better.

also, categories, or some way to filter the stories would be hot. again, i want to see more content.

i do look at new, but the problem is that the step down is from 50+ comments to 0 comments, lots of filtering to no filtering.

i don't want to read unfiltered, uncommented articles anymore than the next person, nor do i want good new articles to be bombed by a lack of views or the first vote being negative.

perhaps we can develop a small group of people dedicated to the task, either for extra or separate karma, or simply to be part of a group doing a good job. the IBM bug-finding black suits comes to minds. this could also be a good task for n00bs to grow their karma faster, though i suspect we'd also have to work in more experienced folks in order to pass on appropriate hacker news culture.

/i wish the big stories would get removed sooner so that more content could filter in. maybe it could just be personal: stories i've seen already are removed./

I think this is a great idea. As someone who visits multiple times a day and see most of the same content (like a lot of other people I'm sure) it would be awesome to be able to hide stories to create room for others.

I agree. Though I am interested if there's a discussion going on, and if it has grown since I last visited it.

Perhaps a 'window' at the bottom of the page of the 5 newest articles would be enough to bump more stories?

I think even a single one would work(as long as it's at the top like reddit)

Submissions aren't really flying here, you get one story every 2-3 minutes. Plenty of time for a few people to like a story and vote it up.

It averages to about 1 in 5 minutes, more during the day, less during the night.

Agreed. Especially with the speed at which the 'new' page scrolls by the chances of making it to the homepage are relatively slim (unless of course you want to game the system but that's not ethical).

A potential way of displaying this would be to simply interleave the home and the new page.

I definitely agree about the speed things are going down on the new page. We got a post for the review of our website (http://news.ycombinator.com/item?id=1075533) and it was out of the page in less than a hour.

I think that a better way to handle it would be to allow users to whitelist/blacklist users, and set up a separate "incoming" section, sorted by recency. Users that you already know, you put on the whitelist and all of their links stay on the page. Blacklisted users never go to the "incoming" section.

You could also put "degrees of trust" on the whitelist and have links from users that are on the whitelist of your whitelisted users.

Agreed, something like this:

http://news.ycombinator.com/item?id=919917

Maybe an optional two column layout for members with upcoming on the right. Maybe a smaller font...

Upcoming should be a function of recent votes. Frontpage order can then be a fx of votes (+), recent votes (+), and time decay (-). This has the benefit of creating the desired functionality without altering the interface.

I've been a huge fan of the 'hot or not' style quick voting for a while now. So much so, that my friend and I included it in a little test project we did a while back: http://beta.popularo.com/

The goal was to circulate new content to all users and at the same time reduce the chances of gaming the system (we were frustrated by the digg "power user" submissions).

Needless to say, it sits dormant now but we're both glad reddit and digg have implemented similar features.

Your site is getting caught in Safari's suspected malware filter. There's a out-of-place link to "Best Philadelphia Wedding Photographers" in the footer which might be related.

I'll let my friend know. It's his domain and hosted on his server. This was the first time I've visited the site in about a year and I happen to have been in Chrome when checking if it was still live, so it worked fine. I doubt he'll care or make any changes since I'm pretty sure nobody visits it anymore. Thanks for the heads up though!

I think that Reddit's solution with a new entry at the top is good. Right now, there is a bias towards the preference of people who visit the "new" page.

I think the current system works fine. People who want to find new stories just have to click the "new" link.

Usually 1 upvote is enough for a story less than an hour hold to get a shot.

The fact that not a single story got an upvote in a 27 minute period, shows that people just don't click the new link. And even when they do, they just don't vote it up.

No, it shows that people didn't find any of those stories worthy of upvoting.

Really?

I'd say the Sivers and the asmartbear posts would hit the front page, if they were featured there even for a second.

The asmartbear post on what you need to do at a tradeshow is now at 2 points, and no where near the front page.

The Sivers post on building intimacy is now at 3 points, and is also nowhere near the front page

There's only so much room on the front page. If you featured something new automatically on the front page, then something else falls off. No doubt people would then complain about that.

You could make the front page longer, of course, but then I'm sure some set of people would complain about that.

No, it means that the small subset of the people visiting the new page in that period didn't find any of those stories worthy of upvoting. But as the relationship between people submitting links and people visiting the new page slowly changes the chance of stuff getting lost gets bigger and bigger.

Mahmud pointed out an excellent example of that happening and I have seen plenty of others.

Go visit the last page of the 'new' section and have a look at all the stuff that was lost. It's really a pity.

And plenty of 'fluff' does get upvoted.

Even though one of my sub was on that screenshot 27 minute period of time, I agree with you.

HN only needs 2 votes to get something on the front page. Not all interesting topic will hit the FP, and not everything I consider interesting is also true for others.

Its not perfect system, mostly because, in this case, perfect is subjective.

For instance I absolutely dislike ~90% of Tech Crunch crap posted here, but I have to live with the fact that others find Tech Crunch drama interesting.

You need 2 votes in the first 3-4 minutes or so. After that the number starts increasing. This is just pulling things out of thin air, but from the looks of it, the system increases the amounts every period.

i.e.

   First 2 minutes? 1 upvote.
   4 minutes: 2 upvotes
   6 minutes: 3 upvotes
   8 minutes: 4 upvotes

etc

I don't think it works like that. From what I've observed, it depends far more on the state of the current front page stories. At night, when most of the stories are older, a single upvote on a New story an hour old can put it on the front page. In the daytime, you usually need that first upvote in 15 minutes or so.

But it's not a set time or point scale, it's based on what's currently on the front page, how old those stories are, and how many points they have.

The one thing I do agree with is that after a certain timeframe, I do wish even the most popular stories decayed more quickly.

agreed. i peruse the new section almost exclusively, and i upvote what i view as worthy of my upvote. people use the system as it is intended. we're just a harder bunch to please, in general.

Alternate ideas:

(1) Make it so that each user only sees a deterministic subset of 'new' (unless they try really hard to enable 'all'). Essentially, this means every 'new' article is seen by a random panel of users, and stays on the 'new' page longer. Tinkering with the assignment function could achieve any balance between submissions/users/views/duration-on-first-page that is preferred.

(2) Segment 'new' by number-of-votes-already-received. Make the default 'new' only 1-vote stories. Only allow viewing and voting on 2-vote stories by people who have already cast a vote on a 1-vote story, and so forth. (That is, force people to pay more attention to those less-reviewed, before piling-on to already-popular stories.)

Reply on news.ycombinator.com