As with many domains, media behaviours follow a power law distribution: "a small group" is responsible for the majority of posts, the majority of comments, the majority of moderation actions, etc. And a small group of subreddits are where a great deal of membership, submissions, comments, and traffic occur. What spez's comment confounds is quantity with significance, most especially power and value generation.
I don't presently have deep insights into the workings of Reddit and numbers associated with its activity.
I do however have a sixteen-year trove of Hacker News front page activity that I've been studying for the past few weeks and drawing various insights from, discussed in several recent comments on both HN and the Fediverse (#HackerNewsAnalytics: <https://toot.cat/@dredmorbius/tagged/HackerNewsAnalytics>).
The trove (2006-2-20 through 2023-6-13) 178,642 front-page stories as archived under the "past" link (<https://news.ycombinator.com/front>), which I've crawled recently. This isn't all submissions to HN (the front page sees only about 3% of all submissions), though it's most of activity (just under half of all comments appear on those stories, integrating with a 2022 analysis from Whaly.io[1]).
Whaly also gives a denominator of active HN accounts, 767,496 (2005--2021), which I'll presume is still roughly accurate. Keep in mind that this is certainly smaller than the total registered accounts, of which I've no data, though I suspect is in the millions.
The front page dataset includes submissions from 43,598 distinct accounts. So straight off the top, only five percent of active profiles have contributed to the front page.
And majority of that 5% appears only once: 24,433, or 56%.
The top ten profiles account for 4% of FP posts.
The top 50: 11%.
253 profiles submitted 25% of front-page articles.
2,092 profiles submitted fully half of all front-page articles. That's just over a quarter percent of all, again, active HN profiles, let alone total registered accounts. A "small group" by any measure, but one that's had a profound influence on Hacker News.
Keep in mind that this is front page submissions and not comments or other activity measures. I'm reporting on what I have insights on, and that's the front page, topic titles, dates, vote & comment counts, and submitters. The Whaly analysis does look at comments, along with other activity, and offers a breakdown of types of significant HN members:
- "Seeders (16%): People that post but that never comment on other user posts"
- "Casual Users (79%): People that reasonably comment & post (between 1 and 100 comments per year) on the social network."
- "Hardcore Users (5%): People that interact a lot with the social network (more than 100 comments on the year)"
My point is that spez's characterisation is dishonest, misleading, dismissive, and irrelevant. I'm disappointed NPR, also in the media business and presumably highly cognizant of such dynamics, didn't push back harder on those talking points, or balance the article in other ways. Small groups can and do matter, most especially in media. And spez is at war with his power users who've contributed tremendously to the enterprise he's now attempting to cash out on, which seems ... unwise.
________________________________
Notes:
1. 2021 retrospective: <https://whaly.io/posts/hacker-news-2021-retrospective> and top 10k commenters in 2021: <https://whaly.io/posts/top-10k-commenters-of-hacker-news-in-...>, latter discussed here: <https://news.ycombinator.com/item?id=29778994>