Unfortunately there are too many useful rhetorical devices in the list, devices that serve a purpose. Ironically, the list starts off with "correlation is not causation," an expression that in most cases is raised for very good reasons -- for example, any popular science story that includes the word "linked" but without adequate qualifiers.
If more people were science-literate, these platitudes wouldn't be necessary. But they aren't, so they are.
I agree. The author says "If there's some specific reason you think a a study is wrong, describe it". Unfortunately, this is very often precisely the problem: many times not in the study itself but in the way it's reported.
"Correlation is not causation" mistakes are not always obvious. Especially when there's ideological bias. E.g: the claim that owning a gun increases the chances that you will be violently killed by X%.
> The author says "If there's some specific reason you think a a study is wrong, describe it". Unfortunately, this is very often precisely the problem: many times not in the study itself but in the way it's reported.
Too true. My recent favorite was a popular account of a marijuana study. The popular article was titled "Marijuana causes psychosis" or words to that effect. The popular account went on about how teenagers went crazy after smoking killer weed. The study itself said, "We don't know whether marijuana use sometimes causes psychosis, or psychosis sometimes causes marijuana use. More study is needed."
"Correlation is not causation" started useful but has become less useful the more cliched it became. And OP is right that it's particularly useless among people who've already heard it.
The whole reason people felt the need to mention "correlation is not causation" is that correlation is evidence of causation. Some people seem to think this catchphrase means the two are unrelated which is also false.
> The whole reason people felt the need to mention "correlation is not causation" is that correlation is evidence of causation.
No, without evidence that assumption is false. Correlation can only be evidence for an unexplained link, and even that is often undermined by desperate researchers' predisposition to offer any detected correlation as though it couldn't result from chance.
Given A and B, absent a plausible causative mechanism, and a correlation between them, possible explanations include:
* Chance -- quick, publish!
* B caused A.
* A caused B.
* An unevaluated cause C connects A and B.
If this seems to go to extremes in skepticism, well, remember that skepticism of new results is -- or should be -- the scientist's job.
Title: "Drug development: Raise standards for preclinical cancer research"
Quote: "Fifty-three papers were deemed 'landmark' studies (see 'Reproducibility of research findings'). It was acknowledged from the outset that some of the data might not hold up, because papers were deliberately selected that described something completely new, such as fresh approaches to targeting cancers or alternative clinical uses for existing therapeutics. Nevertheless, scientific findings were confirmed in only 6 (11%) cases. Even knowing the limitations of preclinical research, this was a shocking result."
> Some people seem to think this catchphrase means the two are unrelated which is also false.
But without evidence, without a rigorous scientific evaluation, that's a scientist's default assumption, an assumption that relies on the null hypothesis. Using the null hypothesis, one assumes there's nothing there, that the association between A and B results from chance, then looks for reliable evidence that might lead us to a different conclusion.
> No, without evidence that assumption is false. Correlation can only be evidence for an unexplained link
I did not say it was conclusive evidence; I said it was evidence. I'm well aware that "A is correlated to B" does not prove "A causes B" or even "A causes B or B causes A", but it is a data point in favor.
Saying "We should evaluate other evidence before we decide if A causes B" is reasonable skepticism. Acting as though "A is correlated to B" has no bearing whatsoever on the question of whether A causes B is another matter.
(Not that I actually disagree with most of your post, mind you! The real message of "correlation is not causation" is "don't overrate this specific data point; it's a common mistake". But the realist shouldn't underrate it either.)
> I did not say it was conclusive evidence; I said it was evidence.
But it isn't. The null hypothesis requires us to assume that there's nothing but chance at work, and let evidence force a different conclusion. The fact that A and B appear correlated is not by itself evidence of anything other than chance.
> I'm well aware that "A is correlated to B" does not prove "A causes B" or even "A causes B or B causes A", but it is a data point in favor.
No, this is false. Without testing a hypothesis, and without a careful examination of a mechanism, the correlation has precisely no meaning apart from chance.
Here's an example selected at random from a vast literature that tries to make this point:
Title: "Creating a phony health scare with the power of statistical correlation"
Quote: "In the United Kingdom, the more mobile phone towers a county has, the more babies are born there every year. In fact, for every extra cell phone tower beyond the average number, a county will see 17.6 more babies. Is this evidence that cell phone signals have some nefarious baby-making effect on the human body? Nope. Instead, it's a simple example of why correlation and causation should never be mistaken for the same thing."
I could link to a thousand similar stories, many being mistaken for actual scientific results.
> But the realist shouldn't underrate it either.
A realist -- a scientist -- always begins by assuming the association is the result of chance (the null hypothesis), and then examines evidence that might argue for another explanation. This is why all self-respecting scientific papers include a p-value. The p-value describes the probability that the result arose from chance, not the hypothesis under test.
Quote: "In statistical significance testing the p-value is the probability of obtaining a test statistic at least as extreme as the one that was actually observed, assuming that the null hypothesis is true."
Translated into layman's language, the p-factor is the probability that the observation -- the "correlation" -- arose by chance.
A properly educated scientist always assumes the null hypothesis is true, i.e. that the observation arose from chance factors. She then tests this assumption with evidence.
Random PG quote I've kept on my clipboard for days looking for the right day to paste it. GOD I HOPE HE SEES.
Copying != stealing, because apparently time stopped about 8 years ago. And this is digg.
IE is hard to develop for even though all the problems it had have been solved for approximately 11 thousand years before anyone alive today was even born.
First they came for [whatever we're talking about] and I said nothing ...
Cliches like this are OK. What really annoys me is that the top rated comment is invariably a straight-forward contradiction of the original article's thesis. Like this comment here.
I personally call it Pwnage Culture. Karma-based sites love it when a user debunks an article or another user's comment.
You get extra points if you start your reply with a cold remark such as "No." "Wrong." "False." And then you put some random citations to Pubmed or Wikipedia at the bottom of your comment.
Law of Reddit XCVII:
You can make a redditor believe anything, no matter how outlandish, so long as you preface it with "Contrary to what you hear in the MSM...."
But I guess that's a tad more specific than generic second opinion bias, which is what we see here
There is nothing wrong with a top comment being a contradiction. I believe that in this way, it might bring even more useful comments and thoughts since it takes a different angle.
The real problem is that those contradicting comments usually are plain simple negative (not to be confused with criticism, which is alright, I am referring to the grumpy negative ones), demotivating.
There's nothing wrong with useful critique or argument. But as others have said it's very easy to dash off a very quick rebuttal of a minor point in an article, ignoring the main thrust of whatever the article is about.
When I've read six pages of interesting and useful article it's depressing to see the top voted comment on HN is about a footnote on page 4 with little relevance to anything interesting.
Sometimes refutation comments are good and valuable though.
In general the decisive factor in the value of a comment is what I will call the "lebowski rule". If a comment contains a statement of someone's opinion and nothing more it is rarely worthwhile.
By way of illustration, this comment would fit that criterion up to this point.
However, I will explain my reasoning which will hopefully rescue it. In the internet age most people are aware of the variety of opinions on a topic, especially the most obvious pro/anti points on a given topic. What can often happen on sites like reddit or HN is that the first comment that espouses a well-known opinion then becomes effectively an impromptu poll. This breaks the link + comments model though because the opinion of the link is inherent in the link itself so a post voicing unadorned agreement would be easily recognized as valueless and thus generally not created or voted up. Whereas the contrarian opinion has no representation and thus becomes the proxy poll for disagreement. The problem is that people voting for/against place their votes differently, either for the link or for the most popular "middlebrow dismissal" comment. But because these are two separate beasts they can't compete against each other directly. Which is why you get the behavior of highly upvoted rejection comments, since that is effectively a reflection of the controversy around the subject. Even though that rarely improves the level of discussion.
In contrast, a more thoroughly thought out refutation which follows through lines of reasoning and evidence will be more likely to contribute to both the disemination of new knowledge to readers as well as higher quality discussions.
Additionally, the mere attempt to try to improve the quality of a comment through fleshing it out will lead to a higher rate of abandonment of less worthwhile posts.
P.S. Here is one area where I think the HN software doesn't help, because it encourages shorter, more quickly written posts which tends to favor the exploration of shallower depths of thought.
Cliches get to be cliches because they're a good idea that everyone wants to reuse. And recycle. Until they're familiar and then over-familiar.
The top-rating issue is simply symptomatic of the lack of insight of many of the folks clicking the little arrows over their first morning cup of coffee. News delivers a brief stimulus; commentary even briefer: the spasmodic twitch of the mousing fingertip merely signals the assent (or dissent) of the mayfly mind.
> The top-rating issue is simply symptomatic of the lack of insight of many of the folks clicking the little arrows over their first morning cup of coffee.
Totally agree and it's hard to blame them. The real problem might be that people who are over-familiar with the cliche don't bother to down vote.
Reminder to all: take the time to down vote over used cliches that bother you.
Unfortunately, not everyone can down vote, and the fastest route to getting enough points to do so is to make the comments we are complaining about to gather the upvotes.
I think the worst is is when the top rated comment is a correction or contradiction of a peripheral assertion in the original article that fails to address the broader implications of the original article.
Though I acknowledge this is a really tough pattern to battle since it's much easier to be certain a quick factual correction provides concrete value than it is to assess a reply with a complicated or nuanced argument. And there is nothing wrong with comments that make a factual correction, it just leads to poor results at the top of the discussion page.
For each of these irritants, perhaps someone has just recently discovered them and is keen to appear knowledgable the next time the debate comes up.
If I had my way, a comment that matched something like /^This\./ would result in an immediate hellban. I don't know why this particular habit annoys me so much.
While this is true for long time readers, there are new readers on this website every single day.
Perhaps someone hasn't learned about the Dunning-Kruger effect and learns something new.
Perhaps the point that he wants to make (I think) doesn't come through, he wants comments to be more then short stubs, comments that really make a point.
Also smugbait 'Why you will (do something that's likely to become popular)' and the imperious 'Begin/stop doing something right now'. Yes, those headlines do get my attention. In a way that makes me wish to slap the writer for their lack of manners.
The biggest problem with Hacker News comments -- actually, make that all comments -- is fundamental: the sort of people that leave comments are (more often than not) people that want something, ANYTHING, to say, not people that have something to say.
Somehow I think your comment applies to TV, radio and writers. The signal-to-noise ratio is getting worse and worse I think (as a side note, I thing signal-to-noise is also one of the frequent HN comments :D)
I don't think that's a problem. I think that's by design. Internet forums are similar to bars where people gather to hang out and talk about common interests. There are other outlets where the people that have something to say can say it.
> it's OK if people just leave anything as a comment here because it will be modded down if it's irrelevant to the discussion.
It would be great if down voting was more prevalent. Many people don't down vote for anything but the absolute worst comment. Some people downvote for disagreement and not poor comment.
If that were the case, the solution would be simple: just skip reading comments.
On the contrary I often learn useful things from HN comments. Sometimes I even submit stuff to HN because I am curious about the opinion of HN commentators.
I think the karma system can also be a contributor to this. Any comment you make has the chance to get a few upvotes, regardless of how little it contributes to the conversation. So people just spam comments to get a few points here and there.
This phenomena can be seen in forums without karma also but i think in places with karma it is usually worse, worst offenders being digg and reddit where everybody tries to make a silly joke as those are easy points.
A better algotithm calculating the karma as an average over all your comments maybe could help..
It would also favour the short and succinct over the long and rambling. Which sounds like a good thing, if you believe that useful information can be distilled losslessly.
It might also be interesting to rate-limit commenters (perhaps based on their previous upvotes per comment).
If you are new to HN, you may only post one comment per week.
The "better" (as measured by karma) your comments are, the more comments per week you may post.
This might make commenters consider whether their comment is really so important that they sacrifice another opportunity to comment in the future.
I have mixed feelings about karma. It's nice to have filters that bring (mostly) better comments to the top, but I think it results in more silenced voices than it does posts-for-karma's-sake.
It's validating to get a few upvotes, but more than that it's disheartening to spend time on a comment and recieve no feedback whatsoever.
I'm an outgoing person, but I find myself shying away from commenting on HN for that reason. Admittedly, I don't usually have some brilliant insight, but it does make it harder to feel like part of the community.
The "average" (on your user page, seemingly over your last few comments) seems to matter more for how high your posts appear. I've found myself deliberately not replying when someone makes an interesting reply to a comment of mine a few days afterwards, because posting that reply brings my average down. Which is definitely a bug.
Telling us how ridiculously little time it took you to build something cool. Most of the time I enjoy what you built, but telling me that you did it "over the weekend" cheapens it for me. Are you bragging, trying to excuse it's shortcomings, or lying? All three most likely. Unless you're a team of people in a weekend competition [1], I'd skip the part about only caring to put a couple days into it before showing it off.
This is not criticism, but just an idea I had while ago while writing similar articles: Maybe only people read this who only consider the gain you get from reading their comment. Probably people who don't care about the other users at all will not read it.
[citation needed] - This isn't Wikipedia, so skip the passive-aggressive comments. If you think something's wrong, explain why.
The reason to post this is not so much thinking the comment is necessarily wrong as that it makes an extraordinary or absolutist claim absent any evidence, which if accepted as fact is going to change the shape of the whole discussion...perhaps appropriately, but the onus is on the person making the claim to support it.
A link to a logical fallacy, such as ad hominem or more pretentiously tu quoque - this isn't a debate team and you don't score points for this.
Considering the requests that if one thinks something is wrong, one should explain why, I'd say that identifying errors in reasoning is quite appropriate; though some of them are so frequent that I just downvote rather than pointing them out yet again. Fallacies of composition are astonishingly frequent, for example.
I consider linking helpful; that's just being a good citizen. Ad hominem is recognizable, but e.g. "Begging the question" is more often misunderstood and linking helps clarify.
I sometimes find myself debating with people who never do any research, who expect to get by with flowery rhetoric. For such people, links can serve as a reality check.
Yes, but on the other hand, fallacious reasoning is rarely the source of disagreement or controversy in a debate. At least in my experience. All you're doing by saying "ad hominem" is shifting the focus to technical delivery of the argument rather than focusing on the real matter of the discussion.
And most people forget that calling something an ad hominem attack does not mean that it's wrong. Most things which get called ad hominem are usually not, in fact -- saying "you are dumb and your arguments is wrong because of X, Y, Z" is different than saying "your argument is wrong because you are dumb" (the latter would be ad hominem, the former is merely an irrelevant statement)
> All you're doing by saying "ad hominem" is shifting the focus to technical delivery of the argument rather than focusing on the real matter of the discussion.
I agree, that looking for logical fallacies in the argument often is a straw man in itself, which is ironic and which is the reason many are skeptical about pointing them out altogether. But in the end you can't abandon logic and talk relying only on intuition and emotions and discarding all reasoning, which would happen if people "rather focus on the matter of discussion than shift focus to technical delivery".
I think you may be confusing HN with some other place. I have literally never seen someone being called dumb or anyone being "proven" wrong for being stupid. Honestly, I think most fallacies pointed out here are fair and especially "ad hominem" which is probably the easiest to grasp. Can you find any counter-examples?
All you're doing by saying "ad hominem" is shifting the focus to technical delivery of the argument rather than focusing on the real matter of the discussion.
I'm not sure I agree. The use of ad hominem is the shift away from the discussion proper. Calling ad hominem (or any logical fallacy) out is merely a recognition of that shift, and could be considered a (weak?) attempt at getting back on topic. When used correctly, that is.
The "citation needed" thing is pulled out any time people disagree, however, even when the comment is something where a source could be found rapidly using Google. It is not the onus of someone in a friendly discussion to have to anally cite every comment they make that involves some random fact: instead, it should be the onus of every person involved in the discussion to keep an open mind and do some due diligence. If you disagree with the other person's fact, then you bring up your own fact and cite the hell out of it to make it clear how wrong the other person is, you don't just demand citations. The real reason people do this is just because they want to undermine trust in the other person's position without having to do any actual work, and it is a great strategy as no matter how well cited something is, there will always be something, potentially in the citation itself (going one or even two levels out), you can take issue with and scream "citation needed" about.
Shrug. Most of the "citation needed" posts I've seen are in response to comments like "almost all recipients of welfare spend it on smokes and alcohol." There probably doesn't exist any studies that disproves that vague claim. Yet whoever posts it shouldn't be allowed to get away with it, hence "citation needed."
"Shrug. Most of the "citation needed" posts I've seen are in response to comments like "almost all recipients of welfare spend it on smokes and alcohol."
[citation needed]
Personally, I don't feel that you need to back your assertion up with facts. If I wish to take issue with it, the onus is on me to go and scrape HN to produce the definitive validation or refutation of your claim. Which was Saurik's point, I think.
Sure, when you leave a comment asserting something, you don't need to provide a full bibliography of references, but if someone asks for it (ie, someone follows up with [citation needed]) then the onus is on you to show that your facts have a basis in reality. I don't see anything wrong with that.
You can't really disprove something which doesn't have any factual basis.
Surely if someone asks for a citation then they need to provide at a minimum some countering anecdotes. Otherwise their "[citation needed]" is even weaker than my anecdote-supported assertion... And if they're going to present countering evidence, why not just do so without the superfluous "[citation needed]"?
> Surely if someone asks for a citation then they need to provide at a minimum some countering anecdotes.
The burden of proof lies with the initial claim AKA Russell's teapot. In a strict academic forum, the person making the claim should provide evidence at the time of making the claim or at least be prepared to defend it with facts.
[citation needed] is shorthand for "Bold claim. I disagree & very much doubt you can back your claim up.". Which I think is ok. Problems arise if it is used for "I disagree & I like to look smart". That is why I would prefer people use the long form, not cryptic shorthand.
anigbrowl is talking about "[making] an extraordinary or absolutist claim absent any evidence", and you're talking about "even when the comment is something where a source could be found rapidly using Google". These are not comparable cases, and so, I guess I don't see how your comment really addresses anigbrowl's. This is what I see. anigbrowl: there are cases where it's appropriate. You: it shouldn't be used everywhere.
There is this (subjective) notion of relevance. When it's relevant, a simple [citation needed] communicates a lot more than merely these words.
To be fair, the article is ALSO not talking about the case anigbrowl is talking about.
I've seen "[citation needed]" abused more often as I've seen it used when somebody is "[making] an extraordinary or absolutist claim absent any evidence"
And the main claim still holds: provide a link that disproves, descredits or calls into question the extraordinary claim. Or at least explain why you think it's extraordinary.
I guess you and the author are sensitive to different kinds of abuse than I am. What annoys me more are wild claims, and I often go away feeling it's not easy to counter this much reliance on misinformation. I am not very active on HN, and read only about 1% of the threads that make it to the front page. Whenever I saw [citation needed], I felt it was apt. Also, I did not get the impression that it was passive-aggressive behaviour.
Identifying errors in reasoning isn't the same as explaining why someone is wrong, though. Given the basic definition of an argument as a structure linking premises to a conclusion, a fallacy is simply a flaw in the structure. It doesn't invalidate the premises or the conclusion, only how they relate to one another. So while you're certainly free to attack the structure behind a conclusion, really you're better off attacking the conclusion itself with an argument of your own.
> Given the basic definition of an argument as a structure linking premises to a conclusion, a fallacy is simply a flaw in the structure. It doesn't invalidate the premises or the conclusion, only how they relate to one another.
True, and there's a name for that logical error -- it's the "fallacy fallacy" or "argument from fallacy", the error of assuming that, because there's a fallacy in one's argument, therefore the conclusion must be wrong.
> So while you're certainly free to attack the structure behind a conclusion, really you're better off attacking the conclusion itself with an argument of your own.
Yes. On the other hand, such a reply may serve to strengthen or clarify an argument, and is therefore sometimes appropriate.
I'm actually quite happy that the list contains these items. Yes, when overused they are annoying (like everything else) but by itself they are all (except the "general internet comment") valid.
I used to dislike the "this is why I love Hacker News" comment. It's not so prevalent anymore, but it was very common 3 or 4 years ago.
And while we're on the subject, I dislike the use of the word sure on nerd sites. I say nerd sites because I haven't seen it (ab)used so much on other sites or movies.
Basically, people say "I'm sure" to mean "I think" and "I'm pretty sure" to mean "I kind of think."
I just went to reddit and the fist thread on my front page had 3 instances of "pretty sure." I'm pretty sure it's the most common phrase and it gets posted thousands of times per day in contexts where the user is not sure at all.
You don't see either as often here as certain other sites, but two more that are invariably abused: "Most people..." and "You've obviously never..." The former is almost always code for "Most right-thinking people who fit into my personal worldview, which I am terrified of breaking...", while the latter is "I've got some domain knowledge here which on face value contradicts the precise words you've just written. It therefore follows that it's impossible for me to be wrong about..." and is more often than not followed by a demonstration of exactly how wrong it is possible to be.
Any time I see either of these, it's end-of-thread-time.
I tend to use such phrases as a shorthand for "I can say that X is true based on my own experience, but have not explicitly researched this topic and so am open to disagreement from someone more knowledgeable".
On the other hand, expecting such nuance to come through in the phrase "I'm almost positive" or "nearly certain" (my preferred variants) is likely a bit unreasonable.
The comment that gets my anal beads in a knot the most: "Citation please" or "Can you link to a study that.." etc.
Sometimes people (myself included) just like to comment. Fart out a completely subjective opinion without worrying about the little army of pedantic party poopers coming along to spoil the fun.
It's just a discussion board after all. Must everything be quite so serious. all. the. time. ?
It's fine to post subjective opinions ("I think yellow flying monkeys are awesome, and should be venerated as gods"), but those kinds of comments are usually targeted at statements of facts that get thrown around ("Yellow flying monkey's poop cures cancer and diarrhea").
Is it necessary to make groundless claims all the time? The reaction you are seeing has a context of serious subject matter. If you want silly disposable comments, try Reddit
I understand this is a place for serious debate and constructive criticism, but sometimes it's more uptight than getting interviewed at a border control.
I visit and get value from it, but this article is spot on. Well, I still think [citation needed] is relevant sometimes, in the case of a legitimate/interesting (or maybe questionable) citation that would produce more value with the citation. (But sometimes it goes like this "The sky is blue"[citation needed])
Yes, sometimes you want to see a source, but why would you use that phrasing? The passive form doesn't feel like a conversation. Like the original poster wrote, it comes across as passive-aggressive.
Imagine someone walking into a coffee shop and ordering coffee by saying: "Coffee required" -- while drawing imaginary square brackets around his head with his fingers.
It isn't a request for a source. It is like saying 'if you want others to believe this thing you are claiming publicly, you need to back it up better than asking us to take you on your word.'
Probably intent. The problem with uncited bold claims is that they can derail entire threads if not nipped in the bud immediately.
I understand that HN is not a debate team, but I don't think it's asking all that much for people to cite sources on something that's likely to be controversial.
It's a politeness thing, really. If you're making claims about something or other, link to WP, or Google, or somewhere out of consideration for your fellow reader. It doesn't take that much longer and significantly contributes to a discussion.
Meanwhile, not doing it in this troll-filled place called the internet tends to make people wonder A, why you didn't bother, and B, what kind of fast one you're trying to pull. Debate is adversarial, after all.
The [citation needed] is a quick and simple way to shift the onus back onto the OP where it belongs, and also serves to warn people who maybe didn't read the comment all too closely that shenanigans may be afoot. And often times that's all you get, because a distressingly high portion of people won't reply when challenged in this manner.
It's a convenient and recognized request for more information.
It's pretty much the equivalent of "RSVP" on an invitation -- a shorthand request that the person it's directed to take an action out of consideration.
Reading it as passive-aggressive strikes me as the attitude of someone who's looking for a social slight. In my experience, that search is rarely unsuccessful.
One way to interpret it as passive aggressive is if its merely an attempt to steer the argument in the direction of the fallacy of appeal toward authority. (abortion is wrong) (citation needed?) (my bible) (I'm not Christian) (holy way begins)
Another way is just a psuedo-politeness. (... and applying ohms law aka power = voltage / current ...) (citation needed?) (quick google search results in ...whoops)
It does make perfect sense for situations where the literature is lacking. Yesterdays discussion of ultra minimal RISC architectures is a good example. So, seriously, the linked to paper only has three references? And I've read two and don't have access to the third? In the grand history of bored programmers daydreaming about turing tarpits I'm somehow familiar with 2/3 of all written articles yet never heard of the remaining 1/3 until yesterday? Citation needed. There's got to be more academic articles than listed. Just looking at the turing tarpit section along of various esolang sites...
(abortion is wrong) (citation needed?) (my bible) (I'm not Christian) (holy way begins)
Perhaps for some. For me it's clear that I've encountered someone whose worldview is fundamentally transrational and further discussion isn't necessary.
My purpose isn't to change the person's mind but to understand it. And if that understanding leads me to the conclusion that they're not worth wasting time in discussion, so be it. What's frustrating is when someone drops some vaguely provocative hint of something without giving a sense of what underlies it. Knowing the source of the bullshit is helpful.
It does make perfect sense for situations where the literature is lacking.
Also for where the person posting has a specific instance or reference in mind but cannot be assed to provide it. Again, my experience is that discussions with such people tend to be pretty unproductive. I only wish H/N had the ability present on, say, Reddit, to tag people as idiots. I've similarly created "idiot", "troll", "denialist", and "libertardian" circles on G+ simply to keep tabs of who's not worth engaging with (or noting to others that their conversation will likely be unproductive).
> I understand this is a place for serious debate and constructive criticism, but sometimes it's more uptight than getting interviewed at a border control.
and the most uptight are the ones who feel the need to pretend.
Personally, I'm tired of comments complaining about HN commenters (not saying author did so, though, those are helpful and non aggressive recommendations).
HN is by far the place on the internet where I find the most insightful comments.
The first 8 points are all totally valid comments and I don't think they can be rejected out-of-hand. The author seems annoyed by a perception that they're thrown into the mix in snarky, passive aggressive or ad hominem ways but that doesn't invalidate their use when they are accurately applied.
Facile, poorly reasoned content is abundant on the internet. Hacker News is a firehose of content – both good and bad – and commenters aim to filter that by tagging the content through comments.
It is not always a good use of your time to make a reasoned argument against blatant logical fallacies, scientific/engineering ignorance and failure to understand modern tech business practices. If you're spending more time discounting vapid content than its idiot author spent creating it, then the world is suffering a net loss.
The inevitable comments about Bettridge's law of headlines every time a title ends in a question mark are not helpful. They don't make an argument, as the article is hardly ever arguing for the question. It is just indicating "look at me, I know of this exotic jargon term, and it happens to apply to this article title". Like, this article could have been "would hacker news improve if we stopped posting certain kinds of comments?"; the answer might be "no", but saying "no, due to Bettridge's law of headlines" provides no value or understanding to the reader.
This, also, is not useful to point out every time an article title ends in a question mark. I'm not advocating for articles with titles that end in question marks, only that there is no value in reflexively pointing out "this article's title ends in a question mark" every time it happens: Bettridge's law of headlines exists, the point has been made, and even to the people who haven't heard about it, it doesn't give the reader any insight. I maintain that the only use of that particular comment is to make the person writing it feel smart for knowing a jargon term.
It does not penalize the article, as even before this concept was given a name it was already clear. The only people it "penalizes" are the people who have to see the comment every single time an article, no matter what the quality level of the content, happens to have such a question-marked title. Think about it this way: the article got voted to some place, and maybe is even on the front page; do you think yelling "Bettridge's law of headlines says 'no'" is changing that vote?
Do you think people reading the comments are going "ah, that is something I had yet to consider about this article; I had previously been curious to know the answer to this deep and burning question, but now that I read this comment, it is clear time that not only am I a fool, but I should down vote this article, all articles like it, and start my own crusade to scream the name of this wonderful law of headlines every time I see such an article".
The best you are getting is "oh, I didn't realize someone had assigned this a funny 'law'... that will be a great trivia item I can being up to demonstrate my epic knowledge of jargon". The result is then just another person in the article comments adding noise. It is nothing more than a way to add mild justification to the seemingly-dying practice of yelling "first": you just need to know, for each article, what bingo terms happen to apply, and then be he first to lay claim to them.
Writing the question you mean to answer or that you mean to explain the arguments for and against into your headline is neither sensationalized or misleading.
Chanting "Bettridge's Law" simply because the headline is a question certainly isn't helpful but that's because it's inaccurate.
As Bettridge himself said:
> they know that story is probably bullshit, and don’t actually have the sources and facts to back it up, but still want to run it
Identifying a headline & story as an example of Bettridge's Law of Headlines should highlight that the headline is nasty link-bait phrased as a question purely to manipulate the reader and that the article provides no insight – in fact it merely serves to further highlight the insidious nature of the headline and the nasty journalistic choices of the author.
Mindless evangelism of proprietary technologies, disguised as an attempt at an insightful comment (microsoft's mvps and evangelists are particularly guilty of this).
Emphasis on the "mindless". I'm horribly spoilt in that I'm always hideously disappointed when cool tech comes along that's not open source, but there's some very interesting work going on behind closed doors that is worth evangelising.
It would be nice to have a sort of an "argument" database, so instead of putting the argument in the comment again, you could just link to the database. Then computers could filter these arguments automatically for the people who already know it. Maybe the arguments could be parametrized, to make the connection to the problem at hand more obvious.
A lot of arguments are repeated endlessly on the Internet, and elsewhere. I've often thought that someone should build a graph database for this. Then maybe we could move past the basic argument and go a little deeper.
I agree. There was Debatepedia (http://dbp.idebate.org), but it seems dead. Unless the arguments are in sort of machine readable format, any filtering will be difficult. The problem is the format - I believe AI researchers still struggle with that one.
228 comments
[ 3.0 ms ] story [ 268 ms ] threadIf more people were science-literate, these platitudes wouldn't be necessary. But they aren't, so they are.
"Correlation is not causation" mistakes are not always obvious. Especially when there's ideological bias. E.g: the claim that owning a gun increases the chances that you will be violently killed by X%.
Too true. My recent favorite was a popular account of a marijuana study. The popular article was titled "Marijuana causes psychosis" or words to that effect. The popular account went on about how teenagers went crazy after smoking killer weed. The study itself said, "We don't know whether marijuana use sometimes causes psychosis, or psychosis sometimes causes marijuana use. More study is needed."
The whole reason people felt the need to mention "correlation is not causation" is that correlation is evidence of causation. Some people seem to think this catchphrase means the two are unrelated which is also false.
No, without evidence that assumption is false. Correlation can only be evidence for an unexplained link, and even that is often undermined by desperate researchers' predisposition to offer any detected correlation as though it couldn't result from chance.
Given A and B, absent a plausible causative mechanism, and a correlation between them, possible explanations include:
* Chance -- quick, publish!
* B caused A.
* A caused B.
* An unevaluated cause C connects A and B.
If this seems to go to extremes in skepticism, well, remember that skepticism of new results is -- or should be -- the scientist's job.
Example: http://www.nature.com/nature/journal/v483/n7391/full/483531a...
Title: "Drug development: Raise standards for preclinical cancer research"
Quote: "Fifty-three papers were deemed 'landmark' studies (see 'Reproducibility of research findings'). It was acknowledged from the outset that some of the data might not hold up, because papers were deliberately selected that described something completely new, such as fresh approaches to targeting cancers or alternative clinical uses for existing therapeutics. Nevertheless, scientific findings were confirmed in only 6 (11%) cases. Even knowing the limitations of preclinical research, this was a shocking result."
> Some people seem to think this catchphrase means the two are unrelated which is also false.
But without evidence, without a rigorous scientific evaluation, that's a scientist's default assumption, an assumption that relies on the null hypothesis. Using the null hypothesis, one assumes there's nothing there, that the association between A and B results from chance, then looks for reliable evidence that might lead us to a different conclusion.
I did not say it was conclusive evidence; I said it was evidence. I'm well aware that "A is correlated to B" does not prove "A causes B" or even "A causes B or B causes A", but it is a data point in favor.
Saying "We should evaluate other evidence before we decide if A causes B" is reasonable skepticism. Acting as though "A is correlated to B" has no bearing whatsoever on the question of whether A causes B is another matter.
(Not that I actually disagree with most of your post, mind you! The real message of "correlation is not causation" is "don't overrate this specific data point; it's a common mistake". But the realist shouldn't underrate it either.)
But it isn't. The null hypothesis requires us to assume that there's nothing but chance at work, and let evidence force a different conclusion. The fact that A and B appear correlated is not by itself evidence of anything other than chance.
> I'm well aware that "A is correlated to B" does not prove "A causes B" or even "A causes B or B causes A", but it is a data point in favor.
No, this is false. Without testing a hypothesis, and without a careful examination of a mechanism, the correlation has precisely no meaning apart from chance.
Here's an example selected at random from a vast literature that tries to make this point:
http://boingboing.net/2010/12/20/creating-a-phony-hea.html
Title: "Creating a phony health scare with the power of statistical correlation"
Quote: "In the United Kingdom, the more mobile phone towers a county has, the more babies are born there every year. In fact, for every extra cell phone tower beyond the average number, a county will see 17.6 more babies. Is this evidence that cell phone signals have some nefarious baby-making effect on the human body? Nope. Instead, it's a simple example of why correlation and causation should never be mistaken for the same thing."
I could link to a thousand similar stories, many being mistaken for actual scientific results.
> But the realist shouldn't underrate it either.
A realist -- a scientist -- always begins by assuming the association is the result of chance (the null hypothesis), and then examines evidence that might argue for another explanation. This is why all self-respecting scientific papers include a p-value. The p-value describes the probability that the result arose from chance, not the hypothesis under test.
http://en.wikipedia.org/wiki/P-value
Quote: "In statistical significance testing the p-value is the probability of obtaining a test statistic at least as extreme as the one that was actually observed, assuming that the null hypothesis is true."
Translated into layman's language, the p-factor is the probability that the observation -- the "correlation" -- arose by chance.
A properly educated scientist always assumes the null hypothesis is true, i.e. that the observation arose from chance factors. She then tests this assumption with evidence.
Copying != stealing, because apparently time stopped about 8 years ago. And this is digg.
IE is hard to develop for even though all the problems it had have been solved for approximately 11 thousand years before anyone alive today was even born.
First they came for [whatever we're talking about] and I said nothing ...
I would love this to be called "Moomin's Law"
You get extra points if you start your reply with a cold remark such as "No." "Wrong." "False." And then you put some random citations to Pubmed or Wikipedia at the bottom of your comment.
Exactly, I automatically downvote comments like that.
It's ok, you can downvote this, I know it'll make you feel better about yourself.
http://en.wikipedia.org/wiki/Conversion_(gridiron_football)
The real problem is that those contradicting comments usually are plain simple negative (not to be confused with criticism, which is alright, I am referring to the grumpy negative ones), demotivating.
When I've read six pages of interesting and useful article it's depressing to see the top voted comment on HN is about a footnote on page 4 with little relevance to anything interesting.
Sometimes refutation comments are good and valuable though.
In general the decisive factor in the value of a comment is what I will call the "lebowski rule". If a comment contains a statement of someone's opinion and nothing more it is rarely worthwhile.
By way of illustration, this comment would fit that criterion up to this point.
However, I will explain my reasoning which will hopefully rescue it. In the internet age most people are aware of the variety of opinions on a topic, especially the most obvious pro/anti points on a given topic. What can often happen on sites like reddit or HN is that the first comment that espouses a well-known opinion then becomes effectively an impromptu poll. This breaks the link + comments model though because the opinion of the link is inherent in the link itself so a post voicing unadorned agreement would be easily recognized as valueless and thus generally not created or voted up. Whereas the contrarian opinion has no representation and thus becomes the proxy poll for disagreement. The problem is that people voting for/against place their votes differently, either for the link or for the most popular "middlebrow dismissal" comment. But because these are two separate beasts they can't compete against each other directly. Which is why you get the behavior of highly upvoted rejection comments, since that is effectively a reflection of the controversy around the subject. Even though that rarely improves the level of discussion.
In contrast, a more thoroughly thought out refutation which follows through lines of reasoning and evidence will be more likely to contribute to both the disemination of new knowledge to readers as well as higher quality discussions.
Additionally, the mere attempt to try to improve the quality of a comment through fleshing it out will lead to a higher rate of abandonment of less worthwhile posts.
P.S. Here is one area where I think the HN software doesn't help, because it encourages shorter, more quickly written posts which tends to favor the exploration of shallower depths of thought.
The top-rating issue is simply symptomatic of the lack of insight of many of the folks clicking the little arrows over their first morning cup of coffee. News delivers a brief stimulus; commentary even briefer: the spasmodic twitch of the mousing fingertip merely signals the assent (or dissent) of the mayfly mind.
Mea culpa: I'm guilty of this, too.
Totally agree and it's hard to blame them. The real problem might be that people who are over-familiar with the cliche don't bother to down vote.
Reminder to all: take the time to down vote over used cliches that bother you.
Well, I suppose they could set up a voting ring, but that seems a lot of trouble to go to.
Though I acknowledge this is a really tough pattern to battle since it's much easier to be certain a quick factual correction provides concrete value than it is to assess a reply with a complicated or nuanced argument. And there is nothing wrong with comments that make a factual correction, it just leads to poor results at the top of the discussion page.
But if I disagree, I can say how I disagree. And this can move the conversation forward. Like this comment here.
Good one. The author could have complained about self referential comments too, and how writing one doesn't make you Douglas Hofstadter.
I liked it, though.
For each of these irritants, perhaps someone has just recently discovered them and is keen to appear knowledgable the next time the debate comes up.
If I had my way, a comment that matched something like /^This\./ would result in an immediate hellban. I don't know why this particular habit annoys me so much.
How about you do one for story submissions? #1 on the list can be ones whose titles are link-baiting "N Xs that Y" types.
b) there are downvote arrows
To me it's OK if people just leave anything as a comment here because it will be modded down if it's irrelevant to the discussion.
It would be great if down voting was more prevalent. Many people don't down vote for anything but the absolute worst comment. Some people downvote for disagreement and not poor comment.
My reading buffer most certainly can.
If the s/n dips sufficiently, I'm out.
It's the Yogi Berra Restaurant Syndrome, or the Evaporative Cooling Effect:
http://blog.bumblebeelabs.com/social-software-sundays-2-the-...
On the contrary I often learn useful things from HN comments. Sometimes I even submit stuff to HN because I am curious about the opinion of HN commentators.
This phenomena can be seen in forums without karma also but i think in places with karma it is usually worse, worst offenders being digg and reddit where everybody tries to make a silly joke as those are easy points.
A better algotithm calculating the karma as an average over all your comments maybe could help..
The metric I would most like to know is how many votes my comments got divided by the total number of words in my comments.
It's validating to get a few upvotes, but more than that it's disheartening to spend time on a comment and recieve no feedback whatsoever.
I'm an outgoing person, but I find myself shying away from commenting on HN for that reason. Admittedly, I don't usually have some brilliant insight, but it does make it harder to feel like part of the community.
[1] http://garage48.org/blogger/list-of-projects-built-over-week...
"just use bootstrap/heroku/nodejs/techoftheday"
"hn is turning into reddit"
"http://ycombinator.com/newsguidelines.html Anything that good hackers would find interesting."
complaints about: negativity, wrong tone and cursing.
looking for reasons to get offended.
The reason to post this is not so much thinking the comment is necessarily wrong as that it makes an extraordinary or absolutist claim absent any evidence, which if accepted as fact is going to change the shape of the whole discussion...perhaps appropriately, but the onus is on the person making the claim to support it.
A link to a logical fallacy, such as ad hominem or more pretentiously tu quoque - this isn't a debate team and you don't score points for this.
Considering the requests that if one thinks something is wrong, one should explain why, I'd say that identifying errors in reasoning is quite appropriate; though some of them are so frequent that I just downvote rather than pointing them out yet again. Fallacies of composition are astonishingly frequent, for example.
I sometimes find myself debating with people who never do any research, who expect to get by with flowery rhetoric. For such people, links can serve as a reality check.
And most people forget that calling something an ad hominem attack does not mean that it's wrong. Most things which get called ad hominem are usually not, in fact -- saying "you are dumb and your arguments is wrong because of X, Y, Z" is different than saying "your argument is wrong because you are dumb" (the latter would be ad hominem, the former is merely an irrelevant statement)
I agree, that looking for logical fallacies in the argument often is a straw man in itself, which is ironic and which is the reason many are skeptical about pointing them out altogether. But in the end you can't abandon logic and talk relying only on intuition and emotions and discarding all reasoning, which would happen if people "rather focus on the matter of discussion than shift focus to technical delivery".
I'm not sure I agree. The use of ad hominem is the shift away from the discussion proper. Calling ad hominem (or any logical fallacy) out is merely a recognition of that shift, and could be considered a (weak?) attempt at getting back on topic. When used correctly, that is.
[citation needed]
Personally, I don't feel that you need to back your assertion up with facts. If I wish to take issue with it, the onus is on me to go and scrape HN to produce the definitive validation or refutation of your claim. Which was Saurik's point, I think.
Sure, when you leave a comment asserting something, you don't need to provide a full bibliography of references, but if someone asks for it (ie, someone follows up with [citation needed]) then the onus is on you to show that your facts have a basis in reality. I don't see anything wrong with that.
You can't really disprove something which doesn't have any factual basis.
The burden of proof lies with the initial claim AKA Russell's teapot. In a strict academic forum, the person making the claim should provide evidence at the time of making the claim or at least be prepared to defend it with facts.
No facts required! One man's half-baked opinion versus another's! See the surprise ending tonight at 9!
There is this (subjective) notion of relevance. When it's relevant, a simple [citation needed] communicates a lot more than merely these words.
I've seen "[citation needed]" abused more often as I've seen it used when somebody is "[making] an extraordinary or absolutist claim absent any evidence"
And the main claim still holds: provide a link that disproves, descredits or calls into question the extraordinary claim. Or at least explain why you think it's extraordinary.
True, and there's a name for that logical error -- it's the "fallacy fallacy" or "argument from fallacy", the error of assuming that, because there's a fallacy in one's argument, therefore the conclusion must be wrong.
> So while you're certainly free to attack the structure behind a conclusion, really you're better off attacking the conclusion itself with an argument of your own.
Yes. On the other hand, such a reply may serve to strengthen or clarify an argument, and is therefore sometimes appropriate.
And while we're on the subject, I dislike the use of the word sure on nerd sites. I say nerd sites because I haven't seen it (ab)used so much on other sites or movies.
Basically, people say "I'm sure" to mean "I think" and "I'm pretty sure" to mean "I kind of think."
I just went to reddit and the fist thread on my front page had 3 instances of "pretty sure." I'm pretty sure it's the most common phrase and it gets posted thousands of times per day in contexts where the user is not sure at all.
Any time I see either of these, it's end-of-thread-time.
On the other hand, expecting such nuance to come through in the phrase "I'm almost positive" or "nearly certain" (my preferred variants) is likely a bit unreasonable.
Sometimes people (myself included) just like to comment. Fart out a completely subjective opinion without worrying about the little army of pedantic party poopers coming along to spoil the fun.
It's just a discussion board after all. Must everything be quite so serious. all. the. time. ?
Yes. Yes it must. It's what makes HN unique and wonderful. There are thousands of other discussion boards that aren't as serious.
I understand this is a place for serious debate and constructive criticism, but sometimes it's more uptight than getting interviewed at a border control.
I visit and get value from it, but this article is spot on. Well, I still think [citation needed] is relevant sometimes, in the case of a legitimate/interesting (or maybe questionable) citation that would produce more value with the citation. (But sometimes it goes like this "The sky is blue"[citation needed])
Is it? [citation please]
Imagine someone walking into a coffee shop and ordering coffee by saying: "Coffee required" -- while drawing imaginary square brackets around his head with his fingers.
I understand that HN is not a debate team, but I don't think it's asking all that much for people to cite sources on something that's likely to be controversial.
It's a politeness thing, really. If you're making claims about something or other, link to WP, or Google, or somewhere out of consideration for your fellow reader. It doesn't take that much longer and significantly contributes to a discussion.
Meanwhile, not doing it in this troll-filled place called the internet tends to make people wonder A, why you didn't bother, and B, what kind of fast one you're trying to pull. Debate is adversarial, after all.
The [citation needed] is a quick and simple way to shift the onus back onto the OP where it belongs, and also serves to warn people who maybe didn't read the comment all too closely that shenanigans may be afoot. And often times that's all you get, because a distressingly high portion of people won't reply when challenged in this manner.
It's pretty much the equivalent of "RSVP" on an invitation -- a shorthand request that the person it's directed to take an action out of consideration.
Reading it as passive-aggressive strikes me as the attitude of someone who's looking for a social slight. In my experience, that search is rarely unsuccessful.
Another way is just a psuedo-politeness. (... and applying ohms law aka power = voltage / current ...) (citation needed?) (quick google search results in ...whoops)
It does make perfect sense for situations where the literature is lacking. Yesterdays discussion of ultra minimal RISC architectures is a good example. So, seriously, the linked to paper only has three references? And I've read two and don't have access to the third? In the grand history of bored programmers daydreaming about turing tarpits I'm somehow familiar with 2/3 of all written articles yet never heard of the remaining 1/3 until yesterday? Citation needed. There's got to be more academic articles than listed. Just looking at the turing tarpit section along of various esolang sites...
Perhaps for some. For me it's clear that I've encountered someone whose worldview is fundamentally transrational and further discussion isn't necessary.
My purpose isn't to change the person's mind but to understand it. And if that understanding leads me to the conclusion that they're not worth wasting time in discussion, so be it. What's frustrating is when someone drops some vaguely provocative hint of something without giving a sense of what underlies it. Knowing the source of the bullshit is helpful.
It does make perfect sense for situations where the literature is lacking.
Also for where the person posting has a specific instance or reference in mind but cannot be assed to provide it. Again, my experience is that discussions with such people tend to be pretty unproductive. I only wish H/N had the ability present on, say, Reddit, to tag people as idiots. I've similarly created "idiot", "troll", "denialist", and "libertardian" circles on G+ simply to keep tabs of who's not worth engaging with (or noting to others that their conversation will likely be unproductive).
and the most uptight are the ones who feel the need to pretend.
HN is by far the place on the internet where I find the most insightful comments.
Facile, poorly reasoned content is abundant on the internet. Hacker News is a firehose of content – both good and bad – and commenters aim to filter that by tagging the content through comments.
It is not always a good use of your time to make a reasoned argument against blatant logical fallacies, scientific/engineering ignorance and failure to understand modern tech business practices. If you're spending more time discounting vapid content than its idiot author spent creating it, then the world is suffering a net loss.
This is valid in the same way that it is valid to point out when a headline contains a false and self-serving claim.
I don't think anybody is under the slightest illusion that having heard of Betteridge's law demonstrates any particular intelligence.
Do you think people reading the comments are going "ah, that is something I had yet to consider about this article; I had previously been curious to know the answer to this deep and burning question, but now that I read this comment, it is clear time that not only am I a fool, but I should down vote this article, all articles like it, and start my own crusade to scream the name of this wonderful law of headlines every time I see such an article".
The best you are getting is "oh, I didn't realize someone had assigned this a funny 'law'... that will be a great trivia item I can being up to demonstrate my epic knowledge of jargon". The result is then just another person in the article comments adding noise. It is nothing more than a way to add mild justification to the seemingly-dying practice of yelling "first": you just need to know, for each article, what bingo terms happen to apply, and then be he first to lay claim to them.
As Bettridge himself said:
> they know that story is probably bullshit, and don’t actually have the sources and facts to back it up, but still want to run it
Identifying a headline & story as an example of Bettridge's Law of Headlines should highlight that the headline is nasty link-bait phrased as a question purely to manipulate the reader and that the article provides no insight – in fact it merely serves to further highlight the insidious nature of the headline and the nasty journalistic choices of the author.
Mindless evangelism of proprietary technologies, disguised as an attempt at an insightful comment (microsoft's mvps and evangelists are particularly guilty of this).
Sometimes discussion boards need to have opinions on them and not just facts, IMO!
Ah well.. a man can only dream.