zlacker

[parent] [thread] 27 comments
1. Solace+(OP)[view] [source] 2019-08-08 11:58:11
Could you clarify why you thought this? What evidence do you have that supports this? The big thread shows that the top comment agrees that 8chan should be left alone. [0] and the comment chain shows that there seems to be something like a significant minority against 8chan, but it doesn’t appear to be a prevailing majority.

0. https://news.ycombinator.com/item?id=20610395

replies(6): >>jamesh+E5 >>tomp+Kh >>mises+bD >>rebuil+pi1 >>haberm+Lj1 >>azangr+ol1
2. jamesh+E5[view] [source] 2019-08-08 12:51:50
>>Solace+(OP)
It’s a persistent Misreading of Internet forums as a mode of discourse, both in how people consume them and how people participate in them, that we tend to regard their discussion threads as a mechanism for determining group consensus on a topic. Cloudflare is dropping 8chan? Let’s get together and decide whether we collectively think that is a good thing or a bad thing. Once we’ve established that fact, we can move on and refer back to that decision in future discussions, like a mathematical lemma.

If you instead think of a forum thread as an airing of opinions - a chance to find out what is the range of perspectives on the topic that exist in the community, and be exposed to nuances you wouldn’t have thought of on your own, the exercise takes on a different tone. People who came to that thread thinking that it’s obviously a good thing are exposed to arguments that disagree, and vice versa; maybe some people are persuaded to shift their viewpoint, or maybe not, but everybody learns that a topic that they might have assumed was uncontroversial is actually one on which reasonable people might disagree.

It can be jarring for the nerd-inclined to accept that just because they have arrived at their opinions through, obviously, clear rational analysis of facts, that does not mean that everybody else, when presented with the same facts, will necessarily reach the same opinion. The illusion that you can read an HN thread and say ‘well, the pro arguments seemed more coherent and got more upvotes than the anti ones, so presumably the community consensus is pro’ ignores the fact that the anti arguments were also made by members of the HN community, and we’re not bound by collective decision making. You are allowed to read the thread and adjust your own priors and come to your own conclusions, having hopefully been exposed to some perspectives you might otherwise have missed.

replies(1): >>Solace+O6
◧◩
3. Solace+O6[view] [source] [discussion] 2019-08-08 13:02:44
>>jamesh+E5
Yes, I agree with what you're saying. But I'm asking why the person posting believes that the HN community overwhelmingly believes X and their evidence for that. I presume they do have evidence and conclusions and I'd like to know about it.
replies(1): >>astine+l8
◧◩◪
4. astine+l8[view] [source] [discussion] 2019-08-08 13:15:01
>>Solace+O6
I think that people tend to perceive HN as overwhelmingly believing whatever the opposite of their opinion is any time there is a significant debate on something. Unless there is overwhelming support for our own position, we feel that we are in a hostile environment.
replies(5): >>Solace+kd >>waterh+qf >>hestip+Qe1 >>monocu+jJ1 >>jcranb+eN3
◧◩◪◨
5. Solace+kd[view] [source] [discussion] 2019-08-08 13:46:33
>>astine+l8
Part of helping to work against this is to challenge and ask for genuine evidence with an open heart. I don't want to assume that that is what the poster is believing, but it also clashes with my understanding of reality.
replies(1): >>tomhow+pZ1
◧◩◪◨
6. waterh+qf[view] [source] [discussion] 2019-08-08 13:59:32
>>astine+l8
Cf. https://en.wikipedia.org/wiki/Hostile_media_effect
7. tomp+Kh[view] [source] 2019-08-08 14:14:30
>>Solace+(OP)
The current top comment. IIRC it was fluctuating wildly while it was on the front page. After it's gone, comments can get reshuffled, because some people might keep replying/reading/up&downvoting (arguably those with more of a "vested interest" - likely those that disagree with the original article)
replies(2): >>Solace+Hi >>mehele+dc2
◧◩
8. Solace+Hi[view] [source] [discussion] 2019-08-08 14:20:26
>>tomp+Kh
Doesn't that provide more evidence that it is not the prevailing majority, then?
replies(1): >>thetru+Zt
◧◩◪
9. thetru+Zt[view] [source] [discussion] 2019-08-08 15:39:51
>>Solace+Hi
Also, does it suggest that the most engaged HN readers (who come early to topic discussions) have a starkly different opinion to late comers?
10. mises+bD[view] [source] 2019-08-08 16:43:00
>>Solace+(OP)
I'd approach it from a different point of view (Cloudflare can choose with whom it does business), but still got the same general idea. Interestingly enough, some of the highest-voted posts aren't always the "prevailing opinion" - some times. Lots of comments get ranked highly because others recognize they are cogent and support them. They may disagree, and so comment, but might still vote in favor if the argument is well-reasoned. I do this personally, when I can.
replies(1): >>segfau+Pb1
◧◩
11. segfau+Pb1[view] [source] [discussion] 2019-08-08 20:18:49
>>mises+bD
Hacker News doesn't simply arrange comments by the number of upvotes they receive, it also considers the karma of the commenters and the freshness of the comments. Also, when submissions aged and comments settled, the current top comment would always get a lot of upvotes due to its position, and lock them "in place" by the strong positive feedback.

So I doubt if reading the top comments is a very objective method for evaluating controversial discussion (it has a strong correlation, but maybe not the best). Often, I see very heated discussion and competing comments moving up and down until nobody is interested in spending more energy in the debate.

P.S: Invasive profiling and tracking can be a very effective (and possibly, the only) method to uncover insights on the dynamics of online forums like Hacker News. If we track users' every move, it could make great contribution to sociology and psychology researches, and may even help answering unsolved questions in order to building a better community for everyone. Unfortunately, it's too dangerous and unethical to use, I won't support it, but I'm always curious to know the results.

replies(1): >>dang+2c1
◧◩◪
12. dang+2c1[view] [source] [discussion] 2019-08-08 20:20:14
>>segfau+Pb1
> it also considers the karma of the commenters

Karma doesn't affect ranking on HN. This comes up often enough that I wonder where the idea came from. Do other forums work this way?

replies(2): >>segfau+od1 >>belisa+QK3
◧◩◪◨
13. segfau+od1[view] [source] [discussion] 2019-08-08 20:27:49
>>dang+2c1
Thanks for the official statement ;-)

Interestingly, I originally believed HN is pure-upvote based, then I learned it from other HN comments that says karma affects ranking and I believed it.

So I'd say it's just an unsubstantiated rumor/misinformation getting circulated in the comment section from time to time, combined with the impression of HN having a "magic algorithm", so many believed it without any fact-checking. Also, the quick-moving nature of HN comments somehow created a confirmation bias that makes the idea appeared to be true.

replies(1): >>Faark+mo1
◧◩◪◨
14. hestip+Qe1[view] [source] [discussion] 2019-08-08 20:36:13
>>astine+l8
It's interesting how one or two dissenting views amongst a majority neutral or even supportive results in "This place is <insert bias> now!" I wonder why absolute agreement is required for some people to not feel attacked or marginalized.
15. rebuil+pi1[view] [source] 2019-08-08 20:56:04
>>Solace+(OP)
Is looking at what the top comment says a good way of gauging consensus? I read that thread and walked away with the impression that the majority view was that "deplatforming" 8chan was mostly OK. This because most of the on-topic comments seemed to hold that viewpoint.
16. haberm+Lj1[view] [source] 2019-08-08 21:03:13
>>Solace+(OP)
Top comment in this big thread argues the opposite: https://news.ycombinator.com/item?id=20616055
17. azangr+ol1[view] [source] 2019-08-08 21:12:39
>>Solace+(OP)
> Could you clarify why you thought this?

I learnt of the news about 8chan from this thread: https://news.ycombinator.com/item?id=20616055 — which was on the HN’s first page before it was replaced by the slightly longer thread you linked to. The top comment in that thread is decidedly against the chans.

There’s also been a lot of mentioning of Popper and his paradox of intolerance in these threads. A post [1] in the thread you referred to (it also was among the top ones when that thread appeared on the front page), for example, began by saying that "Popper taught us that we can't be tolerant towards intolerants" ("taught us" implying that this statement has grown to become general wisdom).

If HN’s prevailing sentiment has since turned in favor of 8chan, I am very happy to hear that.

1 - https://news.ycombinator.com/item?id=20611816

◧◩◪◨⬒
18. Faark+mo1[view] [source] [discussion] 2019-08-08 21:31:57
>>segfau+od1
I would have believed you, since I feel often the same names are highly upvoted. Though this can have natural causes. Like them writing better comments. Or just popularity based on name recognition. Such feedback loops naturally existing without being explicitly implemented by HN sounds highly plausible.
◧◩◪◨
19. monocu+jJ1[view] [source] [discussion] 2019-08-09 00:32:21
>>astine+l8
Perhaps the most insightful comment I have read here in weeks.
◧◩◪◨⬒
20. tomhow+pZ1[view] [source] [discussion] 2019-08-09 04:17:14
>>Solace+kd
I'll have a go at addressing this...

My assessment of that thread is the same as it always is when a thread gets a huge number of comments: sentiment fits a roughly normal distribution, with the mean position being something approximating "this is a really difficult question and either course of action has significant risks and pitfalls", and every step away from the mean point of view placing increasing importance on one particular aspect and decreasing importance on the other aspects.

If that weren't the case, there wouldn't be a huge number of comments, as we would quickly find consensus and move on to the next topic.

If you look at the top three root comments on this thread:

- The first one [1] points out that different standards are applied between 8chan vs Facebook/Twitter/etc, and disagrees with Cloudlfare's decision on free speech grounds. But then many people disagree and debate this position.

- The second one [2] asks a neutral question about Cloudflare's exposure to legal liability for content on its platform if it is making decisions about what content is allowable or not. Then people discuss that question.

- The third one [3] acknowledges the complexity of the topic, devoting each of the first two paragraphs to what the writer considers to be almost-equally meritorious but opposing points of view, then concludes that on balance the Cloudflare decision is right. But then many people disagree and debate that position.

To properly answer your challenge, one would have to examine all 1400+ comments and classify them by their level of support for/against the Cloudflare decision, which is somewhere between impractical and impossible.

But from my scanning through the comments, I don't see any "prevailing" or "overwhelming" position emerge, and I see many of the commenters wrestling with the inherently vexed nature of the issue.

[1] https://news.ycombinator.com/item?id=20610548

[2] https://news.ycombinator.com/item?id=20610552

[3] https://news.ycombinator.com/item?id=20610453

replies(1): >>Solace+gx2
◧◩
21. mehele+dc2[view] [source] [discussion] 2019-08-09 07:44:26
>>tomp+Kh
The ordering is also not just by score. Newer comments get some time at the top as well and then decay to what I assume is their scored position.
◧◩◪◨⬒⬓
22. Solace+gx2[view] [source] [discussion] 2019-08-09 11:48:54
>>tomhow+pZ1
Sorry, I’m not challenging that requires a thorough breakdown. The challenge is the low bar of whether or not a topic has an overwhelming majority of opinion. Which you agree the evidence doesn’t support on a fairly casual glance and analysis of top voted comments and their responses.
replies(1): >>tomhow+1B2
◧◩◪◨⬒⬓⬔
23. tomhow+1B2[view] [source] [discussion] 2019-08-09 12:23:47
>>Solace+gx2
To be clear, my contention is that we should expect there to be no overwhelming or prevailing opinion, and that a quick look at the top-voted root comments and their subthreads seems to support this expectation.

For what it's worth, I think we're taking this discussion a bit too seriously, as the person you were initially replying to was being at least a little humorous and self-deprecating.

The parent comment they replied to made an assertion of the form HN's prevailing view on blah is X, and they replied to the effect of that's funny, my perception was that the prevailing view was opposite-of-X, which is a neat example of the hostile-media effect, and I think the commenter was aware of that.

It's interesting though, that it was the counter-point that you saw the need to challenge, not the original assertion :)

Do you assert that the prevailing or overwhelming opinion was in favour of one particular position? Can you provide evidence for that?

I'm very conscious that we could go around in circles on this :)

replies(1): >>sergio+GC2
◧◩◪◨⬒⬓⬔⧯
24. sergio+GC2[view] [source] [discussion] 2019-08-09 12:43:15
>>tomhow+1B2
> Do you assert that the prevailing or overwhelming opinion was in favour of one particular position? Can you provide evidence for that?

I think like he's arguing exactly the opposite, as he's implied multiple times:

> The challenge is the low bar of whether or not a topic has an overwhelming majority of opinion.

> But I'm asking why the person posting believes that the HN community overwhelmingly believes X and their evidence for that.

replies(1): >>tomhow+4E2
◧◩◪◨⬒⬓⬔⧯▣
25. tomhow+4E2[view] [source] [discussion] 2019-08-09 12:56:59
>>sergio+GC2
If they are of the view that there was no clearly prevailing position on that topic, then we're in consensus and we're all done with the discussion :)

>> But I'm asking why the person posting believes that the HN community overwhelmingly believes X and their evidence for that.

My read on it is that the person wasn't making an assertion of fact on this, they were making a wry observation that their perception of a prevailing view was the opposite of their parent commenter's perception of a prevailing view, thus demonstrating the hostile media effect in action.

◧◩◪◨
26. belisa+QK3[view] [source] [discussion] 2019-08-09 20:42:30
>>dang+2c1
/.
replies(1): >>segfau+T24
◧◩◪◨
27. jcranb+eN3[view] [source] [discussion] 2019-08-09 21:04:46
>>astine+l8
Is it reasonable to say that it is a hostile environment? I don't it takes too many vocally hostile people to create an environment that is hostile.
◧◩◪◨⬒
28. segfau+T24[view] [source] [discussion] 2019-08-09 23:58:12
>>belisa+QK3
Slashdot karma affects ranking? Good to know.
[go to top]