Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

A List of Hacker News's Undocumented Features and Behaviors

github.com

Post reply on HN
Print view View original

181–190 of 196 posts

Re: A List of Hacker News's Undocumented Features and Behaviors

#181

Post by sambroner »

dang wrote: post #177
sambroner wrote: post #137

Earlier quoted context omitted.

What kind of changes to the internals do you all do? And would you do a post on the architecture? I assume there's been lots of work to make the site more efficient/reduce cost. "No [few] internal pressures" sounds like a great recipe for an interesting design.

We've done a lot of work on Arc implementations in the last couple years, but the biggest work still hasn't rolled out yet. If it succeeds, it will buy us a lot more performance. But again, visible features wouldn't change, except that hopefully we'll be able to stop paginating large threads.

Great. Looking forward to it. Would love a write up if possible.

Re: A List of Hacker News's Undocumented Features and Behaviors

#182

Post by m_fayer »

dang wrote: post #175
m_fayer wrote: post #159

Earlier quoted context omitted.

A well designed official API would go a long way toward letting a thousand flowers bloom. I know there's web and mobile apps already out there, but IMO the average quality is not great. Which is not surprising, given the awkwardness of the existing firebase API, which probably keeps many developers away and eats too much of the efforts of those devs that do try their hand.

We're planning to make a new API that gives a simple JSON version of any HN URL. Would that count?

Sure, that would be great. Will there also be endpoints for voting/favoriting/commenting?

Re: A List of Hacker News's Undocumented Features and Behaviors

#184

Post by dang »

m_fayer wrote: post #182
dang wrote: post #175

Earlier quoted context omitted.

We're planning to make a new API that gives a simple JSON version of any HN URL. Would that count?

Sure, that would be great. Will there also be endpoints for voting/favoriting/commenting?

Good question. I'm not sure. Probably we'll produce a complete read API first and then consider a write one. Maybe with proper credentialing there would be some value in it, since people have written lots of scraping clients that do it anyway.

Re: A List of Hacker News's Undocumented Features and Behaviors

#185

Post by dang »

sambroner wrote: post #181
dang wrote: post #177

Earlier quoted context omitted.

We've done a lot of work on Arc implementations in the last couple years, but the biggest work still hasn't rolled out yet. If it succeeds, it will buy us a lot more performance. But again, visible features wouldn't change, except that hopefully we'll be able to stop paginating large threads.

Great. Looking forward to it. Would love a write up if possible.

What would make the most sense is to open-source it all. The trouble is that that takes a non-trivial amount of time.

Re: A List of Hacker News's Undocumented Features and Behaviors

#186

Post by pvg »

dang wrote: post #151
krapp wrote: post #135

Earlier quoted context omitted.

>I know people have strong feelings about downvoting, but those feelings are rooted in something other than discussion quality. How do you know? People have different opinions about what constitutes and contributes to quality. From what I've seen, a lot of people assume Reddit's standard should apply here, and they seem surprised to learn that it doesn't. Those people do seem to be concerned about discussion quality.…

I appreciate your passion on topics HN-related! Taking the last point first: HN downvotes aren't arbitrary and excessive. In most cases, not 99% but maybe 90 and certainly 80, it's easy to see why a comment has been downvoted—except when you agree with it on a topic that pushes your buttons, in which case you will always think the downvote was unfair, but then your opinion can't be trusted. (That applies to all us; w…

The emotional dynamic in downvoting is very strong.c

Have you lot thought much about ways of reducing that, without changing the effect votes have on ordering? I subscribe to the 'emotional dynamic' theory of this and it seems like it's empirically testable without fiddling with the way the site much.

Re: A List of Hacker News's Undocumented Features and Behaviors

#187

Post by Beskz »

I just turned 24, I’ve been a software developer since I was 18-19,and it’s so interesting to go back in time and see what was going on in the HN universe on specific days in the past, way before I even got interested in tech (I was in 6th or 7th grade when HN started). I really wish I would’ve discovered my passion for tech earlier.

Re: A List of Hacker News's Undocumented Features and Behaviors

#188

Post by dang »

pvg wrote: post #186
dang wrote: post #151

Earlier quoted context omitted.

I appreciate your passion on topics HN-related! Taking the last point first: HN downvotes aren't arbitrary and excessive. In most cases, not 99% but maybe 90 and certainly 80, it's easy to see why a comment has been downvoted—except when you agree with it on a topic that pushes your buttons, in which case you will always think the downvote was unfair, but then your opinion can't be trusted. (That applies to all us; w…

The emotional dynamic in downvoting is very strong. c Have you lot thought much about ways of reducing that, without changing the effect votes have on ordering? I subscribe to the 'emotional dynamic' theory of this and it seems like it's empirically testable without fiddling with the way the site much.

How would you reduce it?

Re: A List of Hacker News's Undocumented Features and Behaviors

#189

Post by tptacek »

dang wrote: post #188
pvg wrote: post #186

Earlier quoted context omitted.

The emotional dynamic in downvoting is very strong. c Have you lot thought much about ways of reducing that, without changing the effect votes have on ordering? I subscribe to the 'emotional dynamic' theory of this and it seems like it's empirically testable without fiddling with the way the site much.

How would you reduce it?

You've already done some things to reduce it; for instance, by not greying out your own comments when you've been downvoted. Since comment scores have lots of jitter, you could also smooth out the displayed score, or delay updates. There's almost no time-value to comment scores, but I'd bet most of the pain in a downvote is confined to ~15 minutes from the posting of the comment.

Re: A List of Hacker News's Undocumented Features and Behaviors

#190

Post by pvg »

dang wrote: post #188
pvg wrote: post #186

Earlier quoted context omitted.

The emotional dynamic in downvoting is very strong. c Have you lot thought much about ways of reducing that, without changing the effect votes have on ordering? I subscribe to the 'emotional dynamic' theory of this and it seems like it's empirically testable without fiddling with the way the site much.

How would you reduce it?

Just the obvious stuff - you've already done things like that for upvotes. If we stipulate that having a grey, minus-y comment is some heavy public shaming shit to lay on users, it seems reasonable to try to apply it more judiciously. Are there really five distinct levels comment badness, each with a specific level of illegibility that the poster, let alone everyone else needs to know about? Is a +0-2=-2 comment, or a +20-22=-2 comment really terrible enough for sitewide opprobrium? What would happen if comments just sat at 1 until they hit -4 or below? Or if, for the purposes of display, everyone started with an invisible bonus of N? Theese and similar seem simple (like all work to be done by others) to try without as much as telling anyone.
Post reply on HN