Two HN Announcements

by Sam Altman9/29/2015

The HN community feels like it owns HN, and we like it that way. HN has become an important institution in the tech community, and though it was initially developed for YC founders it’s clearly evolved into much more than that.

We’ve always felt that the best way for HN to benefit YC is simply for it to maximally benefit the community, which mostly means keeping the story and discussion quality as high as possible. We read it ourselves, so we want that as much as anyone.

In that spirit, we have a couple of announcements to make: an organizational one from Sam, and a moderation one from Dan.

Making HN autonomous within YC (Sam)

We’re going to factor out Hacker News into its own autonomous unit of YC. It has de facto been like that, but it feels like a good idea to make it official. Going forward, HN will no longer formally be part of the investment branch of YC, but will be its own separate thing.

Everyone at YC knows that it’s vital for HN to have full editorial independence, and we have absolute trust in Dan’s decision-making in product, engineering, and moderation. Dan will report directly to me, though I don’t plan to be very involved–other than as an enthusiastic user (who would, however, prefer that it be easier to read on a phone) and someone who’s always happy to bounce around ideas. We’re also setting it up so that Dan will have the option of reporting directly to the YC Board of Overseers instead if he ever decides to.

Dan is an incredibly talented person who really understands the art and science of HN; I’m excited to see what he has planned for the future.

Modnesty I: More Community Moderation (Dan)

HN’s approach to moderation has always been three-pronged: software automation where possible, human intervention where necessary, and as much community moderation as users can provide. Our long-term vision for HN is to have the site be as self-regulating as possible, and we’ve been working on a project code-named Modnesty (for ‘moderation amnesty’) to develop new ways to do that. Today we’re releasing the first experiment from this project. The idea is to let the community make the final call on decisions made by HN moderators and software.

Currently, when an account is banned, a software filter trips, or enough users flag a post, the post goes [dead], meaning only users with ‘showdead’ turned on in their profile can see it. The trouble is that some posts end up [dead] when they shouldn’t be. Banned accounts sometimes post good comments, software filters sometimes have false positives, and users sometimes flag things unfairly.

Today’s new feature lets users rescue [dead] posts on a case by case basis. Beside the ‘flag’ link, you’ll see a ‘vouch’ link to click when a post should not be [dead]. When enough users vouch for a post, the software will unkill it. Think of vouches as the inverse of flags: a flag says that a post shouldn’t be on HN; a vouch says it should.

We’ll review all vouched posts to make sure that they don’t violate the HN guidelines. If we notice abusive vouches, we’ll take away vouching rights (again, by analogy with flagging), so please vouch responsibly! Only rescue civil, substantive contributions to the site.

You need a little karma (currently 30) to see flag links, and the same threshold applies here. If you don’t have 30 karma but would like to participate, email hn@ycombinator.com and we’ll try to help.

I called this feature an experiment above. We’d like everyone to understand that when we say ‘experiment’, we really mean it. It’s important to us to be able to roll out new ideas and drop the ones that turn out lame. HN’s simplicity is its core design value, and we shudder at the thought of only adding features and never removing them. The freedom to retract bad experiments (like we did recently [1]) enables us to try more things, which benefits HN most in the long run. So if Modnesty turns out to have unintended bad consequences–which I hope won’t happen, especially since we’ve been testing it for a while–we’ll withdraw it. As always, whether it turns out bad or good is mostly a function of your feedback, so please be generous with it!

1. See my edit of https://news.ycombinator.com/item?id=10223645.

Author

  • Sam Altman

    Sam Altman is the CEO of OpenAI. He was the president of YC from 2014-2019. He studied computer science at Stanford, and while there, worked in the AI lab.