Claim: "You'll find nausea and vomiting in just about every discussion about consuming psilocybin mushrooms.
Challenge: "I know this statement is false, because I've read thousands of discussions. I challenge you to attempt to prove your assertion with evidence. If you make an attempt, you will quickly realize that you are objectively incorrect."
Response (from the claimant, thus far): <null>
Reply (third party): "It would seem that at least some people think Nausea is a problem."
Key distinguishing phrases: "just about every", "some"
Result: Downvote the person who speaks precisely, and objectively correctly.
To those who disagree, first consider how one might physically implement a test of the initial assertion: "You will find <x> or <y> in ~every <discussion about consuming psilocybin mushrooms>".
Might this work: [CountOf forum posts for <DateRange>] - [CountOf forum posts containing keywords for <DateRange>]
Does it seem likely that CountOfKeywordHits / CountOfTotalPosts is going to be up in the range of ~"just about every"? I'm guessing no.
Consequence: Here is where my point lies. If people on HN can't be bothered to not only speak/read accurately and precisely, but punish (downvote, vilify, throttle, stonewall) those who do so (or advocate for such ideas), what might be the general quality of ideas that get persisted in people's minds as a consequence of reading the information that exists in such an environment?
And if this is happening on HN, imagine what the state of affairs is on other platforms (Reddit, Facebook, the media, etc). Now, compare this idea ("facts" that people consume are often literally incorrect) to the current level of ~"general understanding" the broader public seems to exhibit when it comes to the "accurate understanding of relevant facts related to what is going on", across a wide range of important topics (something that seems to come up in discussion, now and then). Might some correlations appear?
And if we looked a bit harder, might a plausibly causative relationship be observable, to some degree? Does the idea that people's internal mental model is affected by the quality of information they consume seem implausible, or controversial? Or, does this perhaps vary based on which (out)group of people we're talking about?
But wait...how true is "facts that people consume are often literally incorrect"? This seems like a rather extraordinary claim. And you know what they say about extraordinary claims!
Again, consider how one might physically implement a test of this extraordinary assertion? Not so simple as doing a keyword search this time. I know of no AI that could do it. Maybe it would require a manual analysis. So, here's an interesting idea: let's say one was to go through this entire thread, and tag each post with a boolean or score indicating the epistemic soundness of claims made within it.
Heck, while we're at it, what others dimensions might yield some interesting output? And if we're going to do one thread, why not do 5 or 10, picking topics from a variety of subject areas, including ones that are bound to be controversial, but are simultaneously extremely important public matters (environment, pandemics, politics, tax policy, etc), where the concerns of people who advocate for old-fashioned ideas like high quality discourse, precision in speech and interpretation, and intellectual honesty, can't be so easily dismissed/framed as stupid, a waste of time, invoking flame wars, "gish-galloping", etc.
So, considering this scenario, what would an analysis of actual written words, consistently interpreted in accordance with the HN guidelines, reveal to us of what is going on in our community? If we did the same thing in other communities, what might we find there? How confident is everyone here that our sense-making is so much superior to that of the <various outgroups> that are commonly mocked for their "lack of intelligence", and tendency to "believe whatever they uncritically read online"? I for one would like to find this out. I wonder if anyone else would. I wonder if anyone cares.
But enough gish-galloping for one day. I imagine I'm just concerned about nothing, things are really not all that bad out in the real world. Most likely, everything is not only mostly under control, but getting better on a daily basis. And besides, it's not like any of this is our responsibility. No, every problem that does exist in the world is someone else's fault. The world isn't what we make of it, it is what the others make of it. As it has always been, and as it will always be.
You're right, the "just about every" is unproveable. But you can't win the argument that nausea and vomiting isn't a concern by saying that the use of "just about every discussion" is wrong, and if you've decided to focus on that aspect you've now switched the topic. I doubt very much the original author intended that to be a scientific fact, and more of a expression of its prevalence. Where some see accuracy others see pedantry and an attempt to exclude an otherwise point simply because of inaccurate language. Also, I didn't attempt to defend the "every", but instead re-frame the claim in more balanced terms, which proves that, at least for some, it's a concern.
> You're right, the "just about every" is unproveable.
(This looks like another assertion! But I digress.)
I don't recall saying that. In fact, my actual words convey the exact opposite meaning (at least to my interpretation):
>> I challenge you to attempt to [prove] your [assertion] with evidence. If you make an attempt, you will quickly realize that you are [objectively incorrect].
Perhaps something was lost in translation along the way? Having a very open mind, I am a big believer that such ambitious things (and many more!) are possible, but such optimistic ideas seem to not be very popular nowadays. Under certain scenarios at least.
But anyways....are you sure?
I can agree, at least not with the explicit specifications we currently have. But then again, it wasn't my assertion, so why is it I who must bear the burden of these shortcomings? Does that seems fair? Or, am I looking at it wrong?
Here is how I would approach this sort of a problem - I like to first put on my thinking cap, and think, much as one might do when writing code at work.
The assertion: "You'll [find] [nausea and vomiting] in [[just about] [every] [discussion]] [about consuming psilocybin mushrooms]."
I've added some highlights around words and phrases, in an attempt to facilitate more precise communication over this limited medium.
Of course, one would have to manually determine the values for [subjectHitCount] and [totalDiscussionCount], but I hope we can leave it at the pseudocode in my prior comment?
Wait a minute. I've arbitrarily decided what the interpretation of "just about every" is. Who do I think I am!!!
We can't look it up, because there's no commonly accepted meaning. What to do!
Now, you and @colechristensen can decide among yourselves what an acceptable value is for [percentageCutoff] - I'm not concerned about that aspect of it.
> But you can't win the argument [that nausea and vomiting isn't a concern] by saying that the use of "just about every discussion" is wrong...
100% correct, and I am in complete agreement.
> ...and if you've decided to focus on that aspect [you've now switched the topic].
Oh? Is it I who've switched the topic? (Also...is this yet another assertion? Oh, never mind.)
By my reading of this thread, it seems to me that that assertion was higher up in the thread. Specifically, here:
(You may also note that I received no response to my rebuttal. But I did receive a downvote, so that's something!)
The assertion we are actually dealing with in this sub-thread, is: "You'll find nausea and vomiting in just about every discussion about consuming psilocybin mushrooms."
Perhaps I should have explicitly quoted the assertion so as to minimize that possibility for misinterpretation. I will try to be more precise in the future. But then on the other hand, being precise seems (as far as I can tell, which is not very well judging by voting and responses) to be not terribly popular around here, despite it being a forum consisting of a high concentration of developers.
> I doubt very much the original author intended that to be a scientific fact, and more of a expression of its prevalence.
Is this my fault as well? If no meaning can be accurately derived from a statement, and reasonable inferences are not allowed, then what is the point of engaging in conversations at all using this language? Just for fun? (And I mean that question literally.)
I challenge you to attempt to prove your assertion with evidence. If you make an attempt, you will quickly realize that you are objectively incorrect.
And then, in light of that realization, perhaps there may be value in reviewing and contemplating, at length, your earlier statement:
> Think about why these kinds of arguments are made on different topics which you disagree with and then rethink why you are making them here.