For anyone interested, an active HNer Edward Weismann (edw519) made an eBook[1] out of his contributions on HN. Certain of his opinions can be disagreed with but there is a lot of wisdom in his comments.
But Hacker News isn't a platform for posting essays. You're supposed to have back and forth discussions. If all we get a treatises from on high discussion threads aren't going to be worth reading.
Inspired by the recent 'What are the most uplifting comments you've read on HN?' discussion [1], I've compiled some of my favourites into an eBook. Pull requests welcome!
From the name I thought it was going to be a book of text generated by a Markov chain trained on HN comment data. Took me a while to figure out it wasn't.
Anyway, good collection of comments. It'd be useful to have a HTML version linked in the README.
It'd be interesting to separate upvotes from downvotes somehow. I've observed that many of my most-upvoted comments receive a lot of downvotes as well - usually for questioning HN conventional wisdom. And I find most of the comments that I really like, that I think are really outstanding, are at odds with the groupthink as well.
It seems to me that most of the claims that posts are "att odds with the groupthink" and claims that posts are "downvoted for questioning HN conventional wisdom" refer to posts that are in line with the frequently-articulated positions of at least as large of a segment of the HN community as any that they are at odds with.
Its natural to want to think that those who disagree with or downvote you are engaging in "groupthink" or defending "conventional wisdom" and those who agree with or upvote you are independent thinkers or rewarding independent thought.
But I don't think upvoting is any less prone to groupthink or downvoting any less prone to having a substantive basis.
Well, sure. You're not going to get upvoted at all if everyone disagrees with you.
But I've noted a very distressing tendency toward comments accumulating significant downvotes not for being wrong, but for being different. Hell, I just got multiple downvotes for saying you should use containers for development even if you're developing on a Linux desktop, because containers are repeatable and your desktop is not. That's kind of a duh thing, but apparently people find it offensive. That's one example, but I've seen it happen a lot, and to a lot of posters making technically correct and logically valid points.
> But I've noted a very distressing tendency toward comments accumulating significant downvotes not for being wrong, but for being different.
I've noted a very distressing tendency for people on HN to assume that no one could possibly disagree with their subjective interpretation of the relevance and degree of substantive contribution of a post, and to therefore assume that downvotes of posts that the complaining poster agrees finds to be insightful must be based on differentness, not on differing subjective interpretations on relevance and degree of substantive contribution. And, further, to state the conclusion of this stack of assumptions as a simple fact.
So do you think disagreement alone is grounds for downvoting? Maybe I'm weird, but I'll often upvote comments I disagree with, if I think they're interesting.
Downvotes as cultural reinforcement is a real problem. And it's a bad idea, imho. I've seen it in the technical community since the idea of comment voting was introduced back in the glory days of Slashdot, 17 years or so ago. It distresses me to see HN doing the same thing.
> So do you think disagreement alone is grounds for downvoting? Maybe I'm weird, but I'll often upvote comments I disagree with, if I think they're interesting.
Disagreement is fairly explicitly -- i.e., per pg -- an acceptable reason for downvoting, though I don't prefer to use it that way and will often upvote posts I disagree with that I feel add value to the discussion.
But that's irrelevant to my point in GP -- my point is that people to tend to present self-serving assumptions about the motivation behind downvotes of particular posts as if they were established facts.
> Downvotes as cultural reinforcement is a real problem.
Downvotes as cultural reinforcement is a central element of community moderation -- "cultural reinforcement" is how the particular character of a community is maintained, and "community moderation" is expressly the decision to give that reinforcement to the community. And that's just as true for downvotes-for-lack-of-substance as for downvotes-for-disagreement.
Having seen both unmoderated and primarily centrally moderated communities, I think community moderation is a fairly good approach (not without its tradeoffs), and HN's form of community moderation is, IMO, one of the better ones.
Yeah. As much as I knock cultural reinforcement of wrongheaded assumptions (it's no different in principle than fundie churches), community policing is better than the alternatives.
Note that finding comment scores is no longer possible for recent comments, as the API disabled returning comment scores for dates after that spreadsheet was generated.
