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

Poll: The new HN comment system – agree, disagree, or wait and see?

news.ycombinator.com

Post reply on HN
Print view View original

51–60 of 157 posts

Re: Poll: The new HN comment system – agree, disagree, or wait and see?

#51

Post by heurist »

I don't like it. I don't mind waiting to see what happens - I'm actually kind of curious to see how it changes the community - but my guess is it will decimate the number of comments per thread because it filters the signal as well as the noise. 1000+ karma users aren't going to endorse even 10 comments per thread on average (unless they are exceptionally enthusiastic), so only the most popular threads will have any possible discussion. In my experience the most popular threads are often the least interesting so I probably won't be able to read or participate in worthwhile discussions while this system is in place, though this depends on the balance of comments vs. upvotes which pg et. al have at their disposal to tinker with.

I think this solution approaches the problem backwards. Instead of blocking all comments until they pass some endorsement threshold, why not make 1000+ karma users' downvotes count as 'denouncements' and hide the post if they get three or four denouncements? This should have essentially the same effect on noise levels without filtering out massive amounts of quality posts.

Re: Poll: The new HN comment system – agree, disagree, or wait and see?

#54

Post by derefr »

hayksaakian wrote: post #18
mkr-hn wrote: post #15

Remember when a chorus said hiding comment scores would be the end of HN?

Comment scores are actually irrelevant. The relative score is all that matters, and a comment's position on the page tells you all you need to know.

Rank is only apparent when a comment has siblings. When two people are having a reply-to-reply back-and-forth debate, I'm usually left very curious to which side HN agrees with.

Re: Poll: The new HN comment system – agree, disagree, or wait and see?

#56

Post by jcampbell1 »

sillysaurus3 wrote: post #16

It's worth comparing this poll to the one about whether HN should display comment scores: https://news.ycombinator.com/item?id=2595605 Only 38.5% of users were in favor of comment scores being hidden, but it was decidedly a positive change for the site.

> it was decidedly a positive change for the site.

Was it?

When things were transparent, there were no reason to upvote interesting trollish comments.

Now they get blindly up-voted for being interesting. For example, consider this article:

https://news.ycombinator.com/item?id=7436401

About 5 facebook engineers are answering questions. They are of course buried by a meta discussion about whether facebook should use PHP at all.

Since the change, I have written a few 50+ karma comments, and all were trolls. The change makes me like my own comments less. Getting a "high score" just requires posting something popular and mildly offensive, where as before, people could weigh, "what conversation is most important".

Re: Poll: The new HN comment system – agree, disagree, or wait and see?

#57

Post by rahimnathwani »

dperfect wrote: post #40
pg wrote: post #10

Earlier quoted context omitted.

Of course. We'd be pretty stupid not to revert something that wasn't working. Plus in any case if you tweak parameters like the number of endorsements a comment needs and how much karma you need to endorse, you can approach current HN as closely as you want.

I'd venture to guess that the difference between a requirement of just 1 endorsement from users with karma >= 1 and the current HN is fairly significant (a downvote differs conceptually from a nonexistent endorsement), so you may not be able to approach it too closely. I suppose only time will tell though.

This reminds me of the 'penny gap problem': that the steep drop-off in users from raising the price from zero to a penny is much larger than the (much smoother) drop from further price increases.

Re: Poll: The new HN comment system – agree, disagree, or wait and see?

#59

Post by jonesetc »

pg wrote: post #2

I suggest another option: wait to see what it's like.

Even if it's just a test, it seems a reasonable scenario that just the test will cause permanent damage. I understand trying to promote better discussion, but this system is downright unwelcoming to any newcomers. When I registered to comment I thought the downvote restrictions were odd and I was willing to try it out, but if this was in place I'd have closed the tab and never come back. Maybe subject-matter experts wouldn't be so put off, but novices and students like myself would feel shunned. That's not saying I'm the most valuable poster, but I am far from a troll and I contribute real opinions and questions. I think lesser experienced posters like myself are important.

I also fear that this may fail silently with an echo chamber as a result. I think this is a real concern here. Unless there is a great amount of monitoring on those posts that don't make it through, none of the automatic posters may even miss the ones who never get a voice. That doesn't make it a positive change though.

I hope you reconsider this change,

A concerned member of this community

Post reply on HN