I was under the impression that comments on news sites were mostly to divert the readership from directly contacting the author or the company through channels that are listened to; merely a solution to keep the wacko segment of the readership occupied and satiated. At least in practice, this is how it appears to function; they're almost exclusively 3rd party plugins and nobody that works for the news organizaion seems to ever engage with them.
I'm admin for several news sites. We chose a third-party plugin for comments because it offers easier account management for the user and better spam filtering than anything home-grown, and like any third-party plugin, we don't have to update that piece of software ourselves. We do read the comments and engage when appropriate.
It's a news aggregator, not a source. In fact, it's only really got two primary functions: allow users to post arbitrary links, and allow users to comment on them.
Most news sources however have three primary functions: Identify, Write, and Publish news.
Commenting in HN is key to its existence; Commenting in news sources just happens to be a thing available.
Saying that comments works in HN is like saying comments work on reddit; it's trivially true, assuming the site is in any way functional/popular. But that doesn't mean comments are useful on NYT; take away comments and there really wouldn't be much lost from its primary goals.
Ah, but HN often doesn't link to general news sites - most of them are dedicated to certain disciplines.
If you're in the US, as an exercise, look up the news site of your local newspaper, and read the comments there. They're often of higher quality than the larger state/national news sites, and even then they're mostly crap.
But on good sites, absolutely! I'd consider comment sections more valuable than the news content, which is often an AP repost with a couple more sentences.
It split the tent. A ten thousand year old idea. Now poor screwed over urban dwellers and poor screwed over rural dwellers are fiercely blaming each other for messing everything up. Setting up a cycle of blame among the powerless disenfranchised groups is so old and still works so well.
Revolutions to me indicate regime change rather than the population fighting itself. France went through a revolution, Yugoslavia through a civil war which led to its breaking up.
What is happening right now pits one fragment of the population against another aka polarization, that's a harbinger of a possible civil war, not of a revolution.
> Google announced last week, for instance, that it had built an artificially intelligent tool that can read through comments and identify whether or not they are "toxic".
I'm Interested in knowing if you see submission guidelines/terms of conditions to be censorious as well then. The sensible approach when dealing with speech protections is that censorship can only come from the State since its muffling of speech is universal throughout a society and is backed by an implicit threat of violence. There is no inherent right to a soapbox upon which to stand on. A private party being in a position where they must broadcast any speech from any source is actually in this arrangement an abridgement of their own speech.
Now that being said, there is a reasonable point of view that says while it should not be legally mandated, private party filtering and editing of content acquired from third parties through some sort of public accommodation _seems_ dishonest. But the truth is, publications writ large exist to transmit a particular experience, whether that be cultural, ideological, or technical(in the sense of the mechanics of conversation a la HN). While this would not be ideal if taken to a no-exceptions type extreme, as long as State censorship is fought off and defended successfully, there types speech that gets muffled through one publication can and will find a light through another.
We seem to have a choice between that and S#1T Posting and Spamming. Now, I suppose Poe's law could limit some sarcasm, but I'm not sure that would be a real loss either.
This reminds me of a literary test to vote. In this case it's: "Here's our news story. Answer some questions to be sure you understand OUR point of view, the words and ideas WE want you to acquiesce, in order to express your so-called opinion." Ungood.
US state voting literacy tests were problematic because (i) the criteria for being able to skip the test tended to be racially discriminatory and (ii) it wasn't practical for most of the affected citizens to move to a state which treated them better.
(i) does not apply; regardless of whether you intend to argue for or against a point made in the article, you're responsible for knowing the same things about what the publishers think they said. Furthermore, even oarsman:regatta-style test bias is of negligible relevance because (ii) also does not apply; as we can see from the 2016 election results, enough people "disenfranchised" by NYT/WP/etc. policies were able to communicate effectively in other Internet venues to produce a catastrophic outcome for the old gatekeepers.
Most of the Internet comments that produced 2016 were either written by folks who'd easily pass this sort of test, or on venues that won't add this sort of test in the future; I wouldn't worry about the impact of this on dissent. If you want to fight against anything a specific media organization does, fight against selective reporting/censorship [1]; but it's not a big deal, the Internet has demonstrated itself up to the challenge of routing around it.
1: If you have a signal with 0s and 1s, and a filter that lets only the 1s through, what is the information content of the original signal? The filtered signal? Incidentally, this is also a major problem in science today.
Even with your wording, it doesn't seem that bad. Shouldn't the reader understand the POV of the author? Even if - or especially if - one considers the author's position as wrong or evil?
Overtonwindow is referring to historical policies in parts of the US that were used to suppress the black vote. These had nothing to do with actually ensuring a literate electorate. White voters were often either exempt from the tests altogether or held to a much lower standard. The concern is something like wording the questions to be ideological shibboleths to let the "right sort of people" comment rather than checking for awareness of specific facts or claims.
This response is why such measures were so effective.
Many of the tests were so subjective they could simply fail anyone they wanted to. Especially certain groups who had just recently gained the power to vote.
if all the literate people were male and all the illiterate people were female, requiring literacy would not necessarily be sexism. you're conflating 2 different things. as did they. perhaps it did exclude more recently-emancipated black people. oh well. requiring literacy for something as important as an election is not obstructionist, IMHO
The fee you mentioned, however, is, IMHO.
if I had to hire a scientist for a job and the only applicants that were qualified enough were white males, hiring a white male would not make me a racist sexist.
i look forward to the day when people just stop trying to correlate everything both positive and negative to race, gender, age, and any other attribute other than "are you a good person" and "can you do this job well"
>i look forward to the day when people just stop trying to correlate everything both positive and negative to race, gender, age, and any other attribute other than "are you a good person" and "can you do this job well"
completely abstracted of the rest of history maybe that'd be possible. the problem is that black people in antebellum south were prevented from learning to read. so really it went something like: let's not teach black to read and then require literacy in order to vote.
>oh well. requiring literacy for something as important as an election is not obstructionist, IMHO
i would gladly wager that an overwhelming majority of voters don't read more than 10,000 words about any given election and get most of their data from television/radio/speech. so tell me why literacy is of paramount importance?
If it were universally applied, evaluated identically each time, and then used a clear pass/fail manner, you would be right. For example, the mathematical answer to "1 + 1 = ?" can be evaluated objectively.
But those "literacy" weren't anything even close to similar.
I've often thought that those who make decisions for society should be required to pass a test to show they understand all sides of an issue. It is absurd to me that politicians are currently allowed to make the wide variety of decisions that they make. I don't think anybody really wants to vote for them to do that. (There was an old episode of Star Trek that had officer candidates use the holodeck to be exposed to a wide variety of situations and outcomes. I would think that whatever we could do to emulate that should be a requirement for making associated decisions.)
The danger there is that the people who write the test end up having some power. They can choose a biased set of questions or even demand wrong/biased answers. Being political issues, often nobody knows the right answers anyway.
So we'd need elections to choose who writes the tests and then who gets to vote for them?
The great thing about democracy is that no matter what goes wrong to the country internally, the blame lies with the people who voted for those politicians, or didn't bother to vote, which is most people. It's a way to thrash out difficult decisions where the people deciding are the same people who end up affected by them. There are gaps in this idea of course, like persecuting minorities and voters being more concerned about their tribal membership
In the US, we've had variations of that in the US under the guise of a literacy test.[1] Promoting "understanding" sounds like a great idea until the test itself is made a weapon against groups that are considered "undesirable" by the majority of the populace.
> great thing about democracy is that no matter what goes wrong to the country internally, the blame lies with the people who voted for those politicians, or didn't bother to vote, which is most people
In the abstract, this is true.
In the perception, it's always the leaders that screwed up - or whomever they can blame.
In democracy, as with most politics, perception is reality
I've often thought that apart from the very fewest national security issues all government should be conducted in the open: all cabinet meetings should be public, all political areas should be monitored and the results available to all, and all private political discussion between politicians should be banned on pain of imprisonment. If they've nothing to hide they've nothing to fear, right?
The problem I see with that proposal is that it would entirely eliminate the ability of Congress to compromise on anything, for fear of a video going viral showing them ceding ground on contentious issues and being used against them next election.
I expect there will be some mockery of the questions if they are not careful. Imagine questions written by MSNBC & FoxNews for the same story. I can see the questions setting up an echo chamber effect.
The cool thing about this kind of filter is that it specifically has nothing to do with what the potential poster is about to post. As long as the questions remain about the facts in the article, intuitively it seems terrible for virality (as it extends the time frame between casual emotion and action), so I doubt any large news outlet would actually implement this.
It can at most be about the facts of the article as understood by the writer (Murray Gell-Man applies here), which the commenter may very well know better than the writer.
As long as they are facts stated in the article (and the article has been read), I don't see an issue. However, a good NLP might still make automated spam posting possible.
The example they give is correctly stating what an acronym in the story stands for. The trouble starts if you have to answer questions like "Which Senators are pro-life" or "Who threatened to exit the EU" which can be leading or misleading or have multiple correct answers depending on your world view.
Then turn the questions into factual "Which Senators are named as pro-life in this article?", "Which country's representative threatened to exit the EU?". Even when you disagree, you should be able to give an accurate account of the opinion you disagree with.
The problem is if you disagree you get another confirmation of the leanings of the organization. I agree with everdev, you really need to stick with indisputable facts. Your first question is problematic because you might dispute the characterization regardless of what the article says.
What is the problem with confirming the leanings of the organization just before you are going to comment about completely disagreeing with the article's characterization?
If the article mentions that "Senator X of the pro-life camp said ..." you answer the quiz question with "X" (because it's an indisputable fact that the article calls X pro-life) and then you are free to comment how "pro-life" is a biased term, or that X is a shill for Y who supported murdering babies in the past, or whatever.
You don't have to agree with the article's stance to know how to answer the questions, you just have to know what the article's stance is.
According to the AP stylebook, "pro-life" would need to be replaced with "anti-abortion" because that is more neutral and carries fewer weighted connotations with it. I would suspect a large population would disagree with the AP conclusion, but it goes to show how difficult it is to avoid bias even in a simple fact-based question.
If the article uses "pro-life" in the text, then the question asking you to confirm that you read that part doesn't add any new bias.
If you didn't read the part but skimmed over it, you might be motivated to leave a comment to mention the AP stylebook's recommendation, to the benefit of other commenters.
Simple fact trivia won't do anything to demonstrate a person has understood the article, merely that he read it.
It's the fundamental difference between functional illiteracy and "plain" illiteracy.
This difference shows up all the time when analyzing data because some jurisdiction will only report the illiteracy number while some other will include the functional illiteracy as well.
I think you're ignoring that many people might not even read the article at all. Proving comprehension sounds incredibly hard; proving someone has at least tried would probably be a huge step in the right direction compared to the mayhem which is most comments online.
Proving real comprehension would be (is?) hard/impossible in a constrained format such as this would have to be (pointy-clicky multiple choice).
Proving someone "tried* is also [nearly] impossible here: just take standard test-taking strategies for standardized tests of reading comprehension, and apply them to this context (because that's all this is, but not even done by people whose entire job is to make "good" tests)
> I think you're ignoring that many people might not even read the article at al
I rarely read the articles posted on HN. The discussion is frequently much more interesting and often more enlightening. I generally only read the article if the filter of other people's comments suggests that it would be worth my time.
It's like learning how to answer standardized tests: the "reading comprehension" questions are usually among the worst offenders. The questions are asked in the order they appear in the passage (eg, the third question will not cover something prior to the first question, chronologically in the passage, unless it is an over-arching question). You can dramatically increase your "reading comprehension" scores by knowing things like this: and the same would be true of "questions about about the article" before commenting - just know to skim for the handful of pertinent keywords, answer, and move on.
I have fantastic reading comprehension from a standardized test point of view - I was always able to finish the section well before the time limit. But I also knew I didn't have to actually follow the directions - I didn't need to read the passage, I needed to find the answers: a very different, and far simpler, task.
My dream of having people answer 5 simple questions to be allowed to get a voting card would come true.
Example of questions: name three neighbour countries, who is the current president, is France in Europe or Asia, what was the last world war,...
I understand that this will unfairly remove people who are illiterate, or had a difficult childhood, or are simply dumb but they are expected to make a choice which will impact the whole country. If they do not know anything about this country and the world it is in then sorry, they are not in a poisition to choose.
Stupidity has very little political bias, fortunately.
Such questions are rarely useful. They tend to be either trivial to the extreme (such as the continent of France or the first letter in the alphabet), politically opinionated, or more complex than you think. (What was the last world war? You and I might think it was the Cold War but with in depth knowledge would have more nuances to their answer.)
This is not a matter of political biais. If someone wants to vote extreme left or right and knows who is the current president - fine.
I do not want biais, I want people who get to vote know the very basics of the world they live in.
Someone who voted X because yeah, the current president is wrong, but this "current president" happens to be a TV celebrity and not the president - such a person does not deserve to have voting rights.
You would be surprised where people put France...
I doubt that a significant number of people wouldn't be able to name the current president. The problem is that anything more complex than that is prone to bias.
Unfortunately, in the US, there's a pretty ugly historical precedent for this practice, and it doesn't leave much faith in people's ability to actually write the questions fairly.
Everywhere anything like this has been done, it has been a travesty. Elitists should have the courage of their convictions, like e.g. ancient Athens had. At least there's less bullshit that way.
Is it being elitist to require someone to know wher your country is located? Or who your president is?
It is not a shame to be illiterare or dumb, or mentally unfit. This does not mean that the future of a country should be put in these people hands.
I will never be a pilot because of health issues. I do not see it as discrimination but just the pragmatic reality of this world.
Everyone knows where my country is located and who "my" president is. (I didn't vote for him!) Which country's location and/or president concerns you? If you say "Slovenia" or "Bhutan", then yes it is elitist to think any voter should have to know that location or person.
I wonder whether another way of approaching the problem would be to have to two commenting sections for a site. One would be moderated algorithmically and the other wouldn't. People could choose according to their tolerance.
I think that would only hinder conversation. The original idea of questions also seems dumb to me, since it feels like even many good posters who actually read the article would be too lazy to answer the questions, and if there was another comment section which didn't require answering questions, they'd just post in there. It also seems like having two comment sections would divide the conversation.
Ignoring the ethics: what reason would you have to answer truthfully? Someone could easily circulate the "correct" responses, and then you pass the barrier with basically no difference. What are you going to do, eliminate voting anonymity and chase people up afterwards to see if they can actually answer the way they said they did?
One form of question could be this. Then even if people parrot the right answer, they are still being improved and have to acknowledge their profound wrongness every time they visit the ballot box.
Your child is very sick. Do you
(A) Do what is recommended by peer-reviewed scientific literature, of which one of the central tenets is that you only believe things which are backed up by data and solid arguments.
(B) Do what is recommended by your religion, of which one of the central tenets is that faith is sufficient reason to believe in something, and neither data not solid arguments are needed.
The system that determines "rightness" is the system that makes your medicines work, your planes fly reliably, your bridges safe to drive over, and the computer and communications network work that allowed you to type that silly question into.
That's a good (and presumably unintentional) demonstration of the problem - the European Union is a hugely complex, very political beast and the right answer to that question depends heavily on your political frame of reference and priorities. Even just the free trade-related aspects can be accurately described either as giving up control of our own laws and regulations to a centralized bureaucracy or enabling trade by removing non-tariff barriers depending on how you want to frame them - since almost all non-tariff barriers other than differences in regulations are banned by WTO rules they're the same thing, one description just emphasises the benefits and the other the downsides. (I've even seen a certain FT writer who insists he's not pro-EU use both framings - the positive one when discussing the EU and the negative one when discussing post-Brexit trade deals outside the EU.)
True, but maybe there could be some sort of democratic consensus system to establish the questions (one which by design avoids the superficial circularity).
That just tests the ability to stick to consensus - what happens when you get the equivalent of "does the Sun orbit the Earth" in a society before Copernicus and Galileo? You also move away entirely from the original point of the questions - someone with scientific knowledge doesn't necessarily know anything about society and politics. They're separate worlds.
Even that is apparently fraught because a few people (though no sects I'm aware of) profess belief in a flat Earth for religious reasons. For example, Biblical literalists interpreting Job 38 as implying a flat Earth ("Where wast thou when I laid the foundations of the earth? [...] Whereupon are the foundations thereof fastened? or who laid the corner stone thereof").
You have been downvoted, but I too would think that was a good idea - unfortunately there are many potential traps here, not the least of which is the ability to control the questions.
In addition to that, you end up breaking the democratic system, which is so very good at keeping our disputes and disagreements from boiling over into a civil war.
