My principle concern here is with transparency and fairness. I'll argue that HN is reasonably good about the latter. For the former, though, there's some pretty serious spelunking that's required to determine practices and policies. The task isn't impossible, you've written numerous times on why shorter and more flexible guidelines are preferable to long and strict ones. And I often (though sometimes grudgingly) come to understand, if not necessarily accept, your position.
I'm well aware of the recent flood of Reddit submissions, and I've made the point myself a couple of times in the past weeks: <https://www.hackerneue.com/item?id=36435319> and <https://www.hackerneue.com/item?id=36321773>, and specifically noted that "Reddit" in titles is trending well above the historical high-water mark set in 2012 (46 mentions for the year), a point I'd made to suggest that HN was not over-penalising the topic.
I'd like to distinguish submissions and even threads (which Algolia search turns up readily) with front-page stories which it doesn't, and which really can't easily be determined without explicitly crawling the front page archive. That's something I'd begun undertaking in late May of this year, (initially to answer a different question: what state(s) get the most love on HN), and even now is answering a range of questions and curiosities about HN.[1]
The front page has both a significance and scarcity which submissions and threads do not: 10,950 slots in a regular year, an additional 30 on leap years. The archive and patterns it reveals over front-page HN activity, itself a mix of influences from submitters to community activity (votes, comments, flags), to HN moderation (automated and explicit), and of course, the larger world, SV and beyond, is catnip for me, and I suspect interesting to others. It's the confluence of influencing factors that I find most interesting --- submitters alone have suprisingly little influence in general on final disposition, something that's hard to recognise from the outside.
I've least concern for my own recent submission's fate (it rapidly ranked far higher than I'd expected, then fell quickly in StoryPos (story position) --- you've explained in email that was due to the flamewar detector). I would suggest that that introduced a new point that a Reddit admin pledge had previously been made to NOT reassign the subreddit. That said, yes, judging significance for a slew of similar stories, particularly as Reddit lays claim to four-, three-, and two-digit membership subreddits, isn't viable for HN.
The fact that there are topics HN has difficulty in discussing reasonably is somewhat painfully evident to many. I would suggest that this could be seen as a challenge to better enable those discussions, many of which are highly significant. And yes, many have tried and failed here.
There was a suggestion a few months ago for HN to feature a dedicated "customer-service-of-last-resort" section. I'd like to see that developed further. Even if the threads are repetitive, they afford a capability the present landscape is sorely lacking. And address a problem which YC in its role as a tech incubator has helped create, which is to say, internalises a cost to YC that's presently been socialised.
<https://www.hackerneue.com/item?id=34941474>
________________________________
Notes:
1. Another way to look at this is what the minimum mean votes of a front-page HN article is (looking at StoryPos 30), which for 2023 has been 161 points. The threshold was last below 50 in 2009. There we see ... 76 posts between 2023-5-30 and 2023-6-22: <https://hn.algolia.com/?dateEnd=1687485600&dateRange=custom&...> Still high, but not quite so overwhelming as your points>50 threshold. It's also well above the number of stories which have made the front-page archive for 2023 to date.
It would be interesting to compare specific keywords which either don't appear on the front page, with high votes, or which do appear on the front page but with low votes. That's not trivial to accomplish presently.
I'm afraid I must disappoint you. We're never going to do that, because (1) it would go against what we're trying to optimize for (https://hn.algolia.com/?dateRange=all&page=0&prefix=true&sor...), and (2) we don't want to run a customer support forum for other people's products. That would be a circle of hell, and not an outer one.
Re "highly significant" discussions - this is an internet message board. The genre is transient and trivial, and as we know from McLuhan, that factor dominates any of the content channeling through it. That's why when people try to do "highly significant" things in internet threads, it always comes across as overblown and overwrought.
A few days ago.
> Why change the weighting given stories on a given topic without notice?
We don't give notice about that kind of thing.
> And how do you reconcile this change with HN's long-standing policy
It follows it strictly. HN has had over 500 threads about Reddit, containing over 25,000 comments, in the 3 weeks since this kerfuffle went kablooey. (And that's just the threads with "reddit" in the title - there have been plenty more.) For any other topic that repetitive and drama-filled, we would have penalized it much more and much earlier.
The rule is that we moderate less, not more, when YC or a YC startup is the story. Note the word "less". We still moderate, we just do it less—and we've done it way less on this Reddit tsunami than we otherwise would have. In fact we probably went too far in the other direction.
I mean just look at those search results: https://hn.algolia.com/?dateEnd=1687485600&dateRange=custom&.... That is nuts, way over the top, and not at all standard practice for HN. The idea (if anyone is entertaining it) that we've somehow suppressed this story is pretty silly when it's one of the most-discussed (and most repetitively, and indignantly) topics of all time.
If there's significant new information, you'd be welcome to let us know at hn@ycombinator.com so we can take a look. I didn't see significant new information at https://www.hackerneue.com/item?id=36434885, and I certainly didn't see a substantively different discussion in the HN comments - merely the same rehash.