The structure of Hacker News just makes it very favorable to muddy the waters until the story disappears. And I think they correctly call that out in the article.

If the fundamental nature of short-duration comment consumption is antithetical to fact checking, that's certainly a problem.

But it's still superior to uncited points, as it is ultimately verifiable.

In a choice between the two, I'll take the ill (claims appearing more supported than they are) for the good (inculcating a culture of transparent citation).

For example have you ever been in an argument where you recall reading a paper or statistic, but don't have it on hand? What if that statistic was a false correlation drummed up by a highly politicized thinktank?

This is why it's so important to not leave potentially misleading or outright wrong citations unchallenged. This is sort of how the 'vaccines cause autism' claims can quickly spiral out of control. It requires people to behave earnestly and not mislead with their citations.

By who? Do you seriously read an HN comment that states something you didn't already know with a citation you don't bother to click and then just go forward assuming it's scientific fact?

I don't even take publications themselves as scientific facts until they have been reliably reproduced or provide ample evidence.

I can make a claim without a citation, or attribute it to "some article I read awhile ago". No one can verify my original claim. Someone may attempt to dig up another citation refuting it, or find a suspect source making my claim, but these are inefficient ways of fact checking. And if I wanted to be a dick, I could claim that wasn't what I was talking about / my original source.

I can also cite my claim. In this case, 99% may accept it at face value, but 1% may fact check my citation and loudly pronounce it doesn't conclude what I said it did. The 99% then gain the benefit of an erroneous report.

> The most admired arguments are made with data, > but the origins, veracity, and malleability of those data > tend to be ancillary concerns.

That's a concern I often share: "the data" can so often be a "winner's history".

People can have a productive discussion on the veracity (origins/malleability are not good reasons to ignore data by themselves in my opinion) of the data, other data can be presented to support or contradict the original point, and in a perfect world perhaps both sides would come away a little more aware of an issue.

    > The nice thing about having discussions based on fact...

Things are different, however, in discussions about human affairs, political, inter-personal topics, social movements, historical interpretations, art, design, music, aesthetics, business and other topics in the human experience. These discussions DO EXIST on HN. And NO, sorry, pure facts may not be enough or might be incomplete or out-of-reach for that subject-matter.

I'll bite.

Data can do a lot to improve discussions about history, politics and business.

I'd also argue that data can improve discussions about seemingly subjective things such as design.

> And NO, sorry, pure facts many not be enough or might be incomplete or out-of-reach for that subject-matter.

In fact I'd say that I'm close to saying those are the only discussions worth having about certain topics.

If I like a design and you don't that brings us nowhere. If either of us can bring some data and say 72% of the testers prefered it strongly, - but colorblind people struggled with it - that is something that might give both of us some value and might lead to better results.

When someone says they are afraid of flying, you show them facts of how safe commercial flight is compared to driving. You don't just quietly sit by while they try to pass legislation making flying illegal.

    > prefer debates be governed by [cherry-picked data] instead of whose argument is the most "humane"

There's a time and place for humane arguments and, in the absence of complete "data" (whatever THAT means), that may be all you have. At the end of the day, we're all emotional creatures and this often justifiably dominates considerations of human affairs. Not everything can be boiled down to problems in first order logic.

I'm trying to be charitable here, but it's difficult not to conclude that you're setting up the data as never being sufficient e.g. "whatever THAT means" in order to support humane arguments over data based arguments.

Holy shit. Observable evidence.

You own argument seem much more a moral one though. For one it doesn't even say what you said it says. For example:

> accusing HN commenters [...] abusing unreliable tools like "data" and "logic" to counter "humane arguments"

Actually says:

> humane arguments are dismissed as emotional or irrational

It doesn't say that human arguments are countered with data or logic. That statement is never made. It doesn't imply it either. If it was to be read as implied it would still be logic and data used in incorrect ways. So it seems hard to favor data if you don't even look at the text at hand.

Instead you seem to pretty much make a "slippery slope" argument, which if not outright is at least close to a moral argument. That questioning data dismisses the role of data, and that this leads to anti-intellectualism. Therefor we shouldn't question data in this way. That is, they are morally wrong to do it. Instead of the less moral argument that misuse of data is what leads to people questioning it.

You also back this up with all kinds of appeal to emotion from it being "a smug, flowery dismissal" to "basically saying I wish these annoying nerds would stop thinking". I don't see that reading as factual. If anything the article seems to argue that people should start thinking. One could argue that their reasoning for why people isn't thinking is incorrect, but that doesn't change the actual meaning in relation to thinking (that they think people should be thinking more). So again you are not arguing the text, but your moral conclusion from the text.

The more critical reading of your comments would simply be that you don't see the problem with data and therefor you don't understand the article's concerns nor the problems with your own comments. Which is pretty much what the article argues people don't do after observing this very forum.

I have a hard time even reconciling your one sentence in this comment. Because you are essentially saying that you rather have flawed data than a humane or morally superior argument. But that again seems very much in itself like a morally superior argument. It rests on that data is always better than a humane argument. Even when incorrect, and potentially more incorrect, than a humane argument.

There are two fundamentally different sources of truth: ethical consensus and physical data.

They are ultimately incompatible, in that only a single option can be your ultimate source of truth. Albeit other(s) can be valued to some degree.

The issue they took, and I would take too, with that portion of the New Yorker article is that it attacks the idea of physical data as a valid, supreme source of truth.

Consequently, the rhetorical paraphrasing of the passage into the comment dismissing nerds seems on point.

Objectively, the humanities has a poor track record of getting pissy and taking cheap shots at science as a viable supreme source of knowledge.

But I doesn't say that. Throughout the paragraph the emphasis is on the recklessness.

"hyperrational, dispassionate, contrarian, authoritative—often masks a deeper _recklessness_"

"_Ill-advised_ citations"

"_thought_ experiments"

"logic, applied _narrowly_, is used to _justify_"

"data, but the _origins_, _veracity_, and _malleability_ of those data [...]"

And that is also in line with the conclusion that "message-board intellectualism that might once have impressed V.C. observers like Graham has developed into an intellectual style all its own" and "can find themselves immersed in conversations that resemble the output of duelling Markov bots". The critique here is essentially that people don't have scientific literacy.

That people on hacker news think they can be rational, dispassionate and authoritative but that their recklessness with citations, logic and data (and their dismissal of other arguments) suggests they can't.

That is the argument.

All of this is questioning the substance of the arguments made, not the substance of such arguments made correctly. The entire point is that they aren't made correctly. You might even read the paragraph as acknowledging that it once worked, when it impressed observers.

> Consequently, the rhetorical paraphrasing of the passage into the comment dismissing nerds seems on point.

You can argue anything you want, but the argument here was made from a basis of facts. If the commentator wants to concede that their argument isn't based on the text itself, then they and you might have a point. But instead the rest of the argument isn't very believable. Which is the logic behind my argument in the first place.

> Objectively, the humanities has a poor track record of getting pissy and taking cheap shots at science as a viable supreme source of knowledge.

Again not relevant to the facts at hand. Especially since again they aren't making those statement. Otherwise we can bring anything into the discussion, from nationality to sat scores.

    > ... it [the cited paragraph] attacks the idea of physical data as a valid, supreme source of truth.

It attacks the idea of using glib, incomplete, or poorly examined "facts" as the basis for a valid argument.

If you want to say that poorly vetted "facts" are a basis for "supreme truth" and, further, that you can only choose that or else some kind of mushy ethical considerations... that's your prerogative, but you're wrong.

Those two points can't be construed any other way than as anti-intellectual. They were cheap shots, and the author should have known they were a cheap shots.

That those points are somewhat incongruous with the subsequent assertion and surrounding piece doesn't mean they're any less anti-intellectual.

] "The most admired arguments are made with data, but the origins, veracity, and malleability of those data tend to be ancillary concerns."

And indeed, I'd say that prefacing the immediately above (defensible) with the prior claims (indefensible) is the reason this entire comment thread exists.

The author could have made a much stronger argument about poor citation and fact checking, but they instead chose to include what feels like knee-jerk humanities logic horror, substantially muddying their point.

    >  ...dismisses the role of data in debate by suggesting that it is malleable or selective...

The points she makes are rock-solid. That said, I still do enjoy HN immensely and can tolerate the warts and libertarians.

Reproducibility crises or statistical hacking news articles are evidence of success. In a less-introspective system, those self-reevaluations didn't even happen!

Viz, the 300+ years it took for a shift from a geocentric to a heliocentric consensus.