Hacker News new | past | comments | ask | show | jobs | submit login
Ask HN: Does the HN commenting UI work for you?
4 points by petenixey 59 days ago | hide | past | favorite | 4 comments
I've been coming to Hacker News since 2005. Over that time it's grown by many orders of magnitude.

The quality of the links has fluctuated but on the whole have stayed strong and, if anything, improved over time. I don't know what the filtering process is but it works well - the articles feel pure and deeply interesting.

But the comments are trickier. As the size of the audience has grown, so have the number of comments. Which has the potential to be really interesting, except for the propensity of the UI to drive both readers and commenters deep into rabbitholes.

I used to love seeing the breadth of reactions to an article (or a new company). I could scan the comments to see what Hacker News thought. I quickly could get an idea of different perspectives and opinions. I could absorb the reaction of the Silicon Valley Coding Diaspora.

These days though it feels, at best, a deep, cloistered exploration of one POV, and at worst a triggered cascade of pedantry.

There are still amazing people and amazing observations in there. But they're not easy to get to. There's no easy way to scan the comments to get a sense of the breadth of reaction - it's just a case of wading through all of them.

Back when I was in my 20s and soaking up everything I possibly could, I would happily invest that time. These days it doesn't feel like such a good trade-off.

And I fear that in turn that penalises the potential breadth of reaction that this remarkable audience posesses. And reduces the participation of others who have less time - many of whom are more senior (decide for yourself whether you value that perspective).

If I could wave a wand, I'd love to see the threads collapsed by default. I feel like there are lot of potential UI wins that could further broaden engagement.

Would love to hear everyone else's thoughts...

(Also... and I'm sure I'm thousandth person to ask this... how on earth is it that there is no dark mode to this dev-mecca?!)

I haven't been reading here as long as you have, so I don't have your perspective. But:

As a reader: I don't think there's only one POV, but maybe because if a topic interests me, I tend to read the whole thread, or at least the first page. And reading deeply enough inside threads, I almost always see multiple POVs, interesting responses and somewhat polite disagreements. Almost every week there's something someone says which is very interesting, resonates with me, or (more rarely) from which I learn something new and sometimes unexpected.

But you're right, wading through comments takes time. I don't do this if I'm very busy. So some weeks I have just a single comment page open on my browser (both mobile and desktop) and I read through it in between things, 5 minutes here, 5 minutes there, until it's finished. I don't read anything else except the front page's headlines.

As a writer of comments: Writing a good comment takes time. Several times I've written a long detailed comment and then didn't post it because I felt I didn't have time to polish it enough. But not always. And reading through other people's comments also takes time out of this, of course. This personally is what reduces my participation.

I don't think having the threads collapsed would help me much. For me it'd be much of the same, I'd still expand them all anyway and read through them, hoping to see those comments that make it worthwhile.

As for dark mode, I don't care about it much. Personally I almost never use it in applications. In my mobile web browser I use the built-in dark mode when I read something just before sleep, but that's it. I personally prefer "warm light" solutions like flux or in recent years built-in OS solutions.

As for your references to the "Silicon Valley Coding Diaspora", that might have been (and still is?) the core audience, but imo the readers / writers nowadays are much more varied and international than that. Which is very important for me. I think focusing solely on "Silicon Valley" would be a waste for this site.

at worst a triggered cascade of pedantry.

and at its best.

There's no easy way to scan the comments to get a sense of the breadth of reaction - it's just a case of wading through all of them.

Friction, I think of as a feature, not a bug. Slowing people down raises the threshold for engagement. Requiring effort reduces drive-by's in the comments.

It's not just that HN has changed, you have changed too. Probably more than HN. Yes, you don't have the time. But also less is new...I mean I heard of Python on HN.

But one of the ways HN has changed is the content and comments are more loosely coupled with the context of YC. Funding rounds are unlikely to make the front page anymore.

And that's because startups are mainstream news...perhaps due in no small measure to YC's activities.

Yet, most likely it is mostly you -- after all, it's your cranky old man post.

Good luck.

If I could wave a wand, there would be a way to have multiple dimensions to vote across, and a way to tailor how much those votes affect what I see, or the order thereof.

I'd make revisions possible past 2 hours, but make it quite visible that revisions had been done, and a way to visualize the differences.

I’ve been experimenting with a smart RSS reader that is, I think, successful with news articles. It trains a text classifier to filter content I like. I added the RSS feed for HN comments to it and thought it was too much, but I still have it ingest the ‘top comments’ and like some of what I find, I think the highest probability score it’s assigned to a comment is around 0.5 though, contrasted to 0.9 that I regularly see for some machine learning papers and articles about the Ukraine war.

This is not really the right approach to comments because they are embedded in a tree, the parent and children should matter for classification. I think though a comment viewer would be possible and I also think metafilter would be a good source of comments, possibly better than the metafilter articles.

(Oddly in the last month I have been thinking about inferred preference scores as more ‘real’ in some sense than revealed preferences. HN articles that have a high comment/score ratio are frequently knockdown-dragout matches and by training a classifier on that ratio I have a ‘clickbait’ score that expects a lot of comments on ‘Is Bitcoin a Shitcoin?’ but very few on anything that has to do with graph databases.)

Applications are open for YC Summer 2023

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact