I believe globalization has brought up some very unique problems with regards to how information is disseminated and how we process it. In the past, when you lived in a much smaller community, you...
I believe globalization has brought up some very unique problems with regards to how information is disseminated and how we process it. In the past, when you lived in a much smaller community, you could learn to come to trust particular sources of information when it came to particular matters. As communities grew from towns to villages and cities, new methods of processing information came about. Logical rules to help make sense of whether information could be trustworthy were developed and organizations came in place of individuals - rather than getting your news from the most social and connected people, you would get it from the news organization and you'd get information from books instead of teachers.
However, I do not believe we have come to a point at which we understand, on a society level, how we need to treat globalization as it exists today. What's happening on twitter is really just an amplification of what's happening in societies across the world and has to an extent happened throughout history. People have the inherent biases of thinking they are more knowledgeable than others and that their opinions are more important and that their values are the right things to value. When societies were smaller it was much less likely to run into bad actors in any particular community and so we looked to trusted sources of information for guidance - if you had a question about science, you might go to the local science teacher and ask for their advice. Today, you can find thousands of science teachers online, and invariably some of them will share your same opinion - you will be drawn to people who confirm what you already believe and it will solidify your beliefs.
Because of this we see conversations like the one driven on twitter by this study. Many on twitter are concerned (perhaps rightly) on the negative effects that 'techbros' can have on society and have begun to become aware of how bad actors are exploiting the current globalization to spread misinformation and further realize their own ideals. However, these same concerned individuals are falling into a trap of their own creation - they believe in their own ideals so strongly and have such a desire to prevent society from falling back into demise that they are drawn to the perpetual echo chamber that they exist in, only seeking out others who share the same ideas and vehemently convinced that theirs are the most important.
What we're missing today is a smart, methodical way to approach the nearly incomprehensible amount of information presented to us on a daily basis. We need a way to be able to respond both nimbly and adequately to new information as well as a way to understand whether someone is approaching new information in a similar manner. In the recent past we often were able to come to trust sources of news for being more unbiased than they are today. I think we are struggling to find that model today. We have prolific accounts spread about a variety of online sources - prominent tweeters, youtubers, streamers, and so on. Perhaps this will be a part of the new model for information dissemination.
Regardless of what the correct model is, I think the adage of 'never read the youtube comments' is increasingly applicable on larger scale conversations and particularly fast (chronologically) mediums of news dispersal and conversation. In a rush to provide commentary on matters important to the people who tweeted, they managed to spread misinformation. This spread of misinformation to further an agenda is also why I very rarely read articles about scientific papers, especially in the fields of nutrition and health science. I just hope we can get to a point where the system is able to provide adequate sources of information for the interpretation of said science to less scientifically-capable individuals lest we continue to swirl and suffer the downstream consequences of inadequate, malicious, and distorted interpretation to further spread the ideals of the interpreter.
Statistically confirming something like that without Machine Learning would have been insanely difficult. You'd need to manually tag attributed to every single word in multiple languages. That...
And yes, it's fine to confirm with more statistics things we already know, but most discussion focused on the "machine learning" aspect, which is what was most heavily mocked, at least within my filter bubble.
Statistically confirming something like that without Machine Learning would have been insanely difficult. You'd need to manually tag attributed to every single word in multiple languages. That alone, from a methods standpoint, seems significant. Confirming things we already know with interesting or novel methods is useful not just because it adds another bit of evidence to the pile that confirms whether we "know" something, but it also means you're developing a research method that you can apply to other things in the future.
But in that case, the issue wouldn't necessarily be one of not reading the article. It would be lay people who just don't understand how research works well enough to understand what's cool/interesting about something or why.
Overgeneralizing from patterns is a big no-no when doing real scientific inquiry. You may think this is obvious, as a lay person, but to linguists who study language phenomena as a formal science,...
Obviously "Languages that are more geographically proximate, more historically related and/or spoken by more-similar cultures [have] more aligned word meanings."
Overgeneralizing from patterns is a big no-no when doing real scientific inquiry. You may think this is obvious, as a lay person, but to linguists who study language phenomena as a formal science, it is absolutely a mistake to say this obvious. Historical Linguistics undertakes the study of language contact, borrowing, etymology, and semantic drift (among other things) and can show examples where related languages have diverged in terms of distributional, lexical semantics. There are evolutionary processes, esp. phenomena such as narrowing or metaphor, that predict the opposite of aligned word meanings in some situations!
A comparison from evolutionary biology that may help to think about is Darwin’s finches which exhibit divergent mutation and adaptive radiation across a small geographic proximity as influenced by their respective environments. We have linguistic evidence of non-mutually intelligible dialects, such as pidgin languages and other geographically collocated languages, or the poles on dialect chains (see also Pacific Languages: An Introduction (Lynch 1998) ch. 2). We also have instances of language isolates, such as Ainu or Basque. These instances provide interesting counter-examples for predicting lexical similarity based on typological, phylogenetic, or geographical distributions.
Pretty good article, although since the author has written his piece the tweeter mob seemed to have done a 180 in it's opinion on the article, over the span of about 48 hours. I went ahead and...
Pretty good article, although since the author has written his piece the tweeter mob seemed to have done a 180 in it's opinion on the article, over the span of about 48 hours. I went ahead and subscribed to the author's newsletter, since I dig the writing style. I do think that a plea to read before you form an opinion will largely go unheeded though, it's a ship that's just way too far gone, the battle was lost somewhere during the heyday of digg.
I think this is an atrocious example to pick, for a very real problem. Do I really care that numbers are more semantically aligned across languages that prepositions? There are so many more...
I think this is an atrocious example to pick, for a very real problem.
Do I really care that numbers are more semantically aligned across languages that prepositions?
There are so many more important questions that this sort of rigorous research could have answered.
Such as semantic links between authoritarianism across countries and cultures.
Sure, you might spend up with a whole lot of effort with little to show. But this is what AI is good for, spotting important correlations that humans can't see. Lets use it to solve the big problems.
I think the original tweets (that I did not read) got it right on two accounts. First, this research is wasted applying AI to a trivial problem where we already knew the answer. Second for not reading the paper in the first place.
If it's a novel AI, then applying it to known problems seems like the perfect way to start. It teaches familiarity with the tooling, the data, and any gotchas that may exist in the process. I...
First, this research is wasted applying AI to a trivial problem where we already knew the answer.
If it's a novel AI, then applying it to known problems seems like the perfect way to start. It teaches familiarity with the tooling, the data, and any gotchas that may exist in the process. I would guess that these researchers are now much better equipped at applying similar methodology to other problems.
Some of this stems from the fact that nobody has time to go to the original sources for all the things they want to learn about. Sure, that's a better way to do it. But we read articles that...
Some of this stems from the fact that nobody has time to go to the original sources for all the things they want to learn about. Sure, that's a better way to do it. But we read articles that summarize detailed scientific papers because we can't read all the papers ourselves, but we still want to have some information about them. Then others comment on the originals, the articles, the headlines, and the reactions of others, etc. This results in the game of telephone that this blog post describes. I don't think it's realistic to tell everyone to go read the original paper. The dynamics of social media amplify the worst responses, unfortunately, but that's human nature. I don't think it's something that can ever be "fixed". But chastising people for not taking time they don't have before daring to speak about something seems like a waste of breath. People like to blow off steam online, and they're going to continue to do it.
There is a more subtle issue, which is that we often use an article as a prompt to talk about something else that is sort of related. For me this is most noticeable on Hacker News. Until recently,...
There is a more subtle issue, which is that we often use an article as a prompt to talk about something else that is sort of related.
For me this is most noticeable on Hacker News. Until recently, any article about Go became an opportunity to talk about why Go doesn’t have generics even if the article isn’t about that at all. Any article about Amazon becomes a place to complain about counterfeit products on Amazon. And so on. There are certain keywords that attract certain discussions.
I don’t think this can be blamed on headlines being bad since Hacker News does a reasonable job.
Although these comments do get upvotes, it doesn’t seem to be like on Twitter since having lots of karma on Hacker News isn’t something people care about much.
I think these are just things people want to talk about more than the actual article. I don’t think there is any way to stop it other than more heavy-handed moderation to stay on topic, and I’m not sure that’s desirable.
In any medium, humans are going to human. In person conversation usually happens this way as well - people talk about a subject and often times others are reminded of something tangential,...
In any medium, humans are going to human. In person conversation usually happens this way as well - people talk about a subject and often times others are reminded of something tangential, related, a personal anecdote, or a feeling and drive conversation in a new direction.
The idea that people need to discuss only the article always seemed somewhat silly to me. Certainly we want to keep comments free of people stating things that are blatantly false, but we shouldn't be discouraging completely regular human conversation just because it's not what's discussed in the article but instead something in a similar vein.
I think the problem with online conversation is that it's so vast. There are potentially thousands to millions of people looking at the same thing, so it's not surprising that an absurd amount of conversation can happen. The problem is that everyone wants to see the conversation they came to the article for, or they came away from the article with and that can be hard to find in the noise of everyone else discussing. The way you moderate discussion is intrinsically tied to how large the audience is. When it's large enough, it makes more sense to split conversations into their own posts and threads for logical sake. When it's a smaller community like this, however, I think its fine to leave multiple conversations going at the same time.
Yeah... I think there's another fundamental problem that just boils down to it being a numbers game based off of "absorption" time. Content that is easier to make snap judgements about will almost...
Yeah... I think there's another fundamental problem that just boils down to it being a numbers game based off of "absorption" time. Content that is easier to make snap judgements about will almost always get more "points" and shares than content that requires more time and thoughtfulness to genuinely react to. If it takes ten seconds to read a headline, have a feeling about it, and share it to one other person versus ten minutes to absorb an article and form and share an actual opinion, in the time it takes someone to read the entire article and post their thoughts there are dozens of other folks who have already "finished" with it to their desired depth of engagement.
Then reading these hot takes in lieu of the actual primary source is kind of like having your food pre-chewed for you. I didn't read the article, but I presume that this person did, and that's almost as good, right? So I can just share their take, bam, and then I can move on to consuming the next thing!
And there's always a next thing to move on to. The quicker you can move on from your previous take, the sooner you get that next dopamine hit for getting involved in the next trend/drama/post/etc. The internet is a never-ending treadmill that disseminates content faster than is maybe reasonable, and your attention is a valuable commodity that people compete (viciously!!) for. A lot of smart people spend a lot of time thinking about how to keep you on their app longer, how to get you to share more things with your circles, how to get just a second of your attention or to provoke the smallest engagement. The same kinds of folks who took the clocks out of casinos are now all about endless scrollbars and other tricks to keep you somewhere longer than you originally intended.
I don't think there's any stopping the tide, frankly. Too much money in it. On an individual level I at least try to be aware and I try to resist my own urges to over-consume, and I try not to post or share glib/useless/mean stuff. I think the environment has harmed me, in a way. But I don't think I can stop anyone else, so that might be as much as I can do.
Very well said, and very much agreed. I don't know how we fight the tide, either. Rapid content will always outpace slower content; inflammatory content will always generate more of a response...
Very well said, and very much agreed. I don't know how we fight the tide, either.
Rapid content will always outpace slower content; inflammatory content will always generate more of a response than more measured content. It's not a fair fight even in the abstract, and, as you identified, there are plenty of people designing entire systems around exploiting this imbalance and habituating users to behaviors that are beneficial to the platforms but detrimental to the well-being of their users. Probably the worst part is that this is deliberately opaque, so the platforms that are actively exploiting us can successfully position themselves as allies rather than adversaries. The arms around us have hands picking our pockets, but to us they simply feel like a hug.
I think the internet operates fine in this regard. The problem is a human problem. Without a robust measure of intrinsic content quality, any aggregator like Reddit, Twitter, Tildes, Stack...
Given the nature of how sites like Reddit or Twitter work, where algorithms promote content with more interactions regardless of whether they are based in fact or not, I don't know how we can fix this without fundamentally changing how the internet operates.
I think the internet operates fine in this regard. The problem is a human problem. Without a robust measure of intrinsic content quality, any aggregator like Reddit, Twitter, Tildes, Stack Exchange, etc. can only surface quality content and filter poor quality content as well as the community/moderators can. That is, it’s a very hard (impossible?) problem to objectively measure quality, so it’s much more efficient to use aggregate, subjective measures (such as counting upvotes/downvotes/likes/claps etc.).
There are certainly other models for curating content, such as the ways that peer-reviewed academic journals work, but they are so much slower and work-intensive, and domain-specific that they really aren’t feasible at larger scales. The only large-scale, generalized efforts I’m aware of that approximate that sort of model are Wikipedias, but even they have issues of vandalism and "notability" restrictions. I suppose there are other domain-specific wikis and KBs that may also use a similar model of crowd-sourced peer-review, but they also tend to be domain-specific.
Ban clickbait articles. Outlets and articles that attract the kind of answers you refer to... Maybe. Or if you want to be more lenient here, at least edit the title to include the important...
I don't know how we can fix this without fundamentally changing how the internet operates.
Ban clickbait articles. Outlets and articles that attract the kind of answers you refer to... Maybe.
Or if you want to be more lenient here, at least edit the title to include the important information of [clickbait, important context needed] by mod action. That should tip off the careless reader at least a little bit.
But how do you define clickbait? Headlines exist to seek attention on purpose. "These Voters Back Trump Unapologetically. Here’s Why." Is the top NYT article right now, and it's almost begging you...
But how do you define clickbait? Headlines exist to seek attention on purpose.
"These Voters Back Trump Unapologetically. Here’s Why." Is the top NYT article right now, and it's almost begging you to click on it.
Caveat: I haven’t thought this through very carefully. I feel like the reason I wouldn’t categorize that headline as “clickbait” is that it is, in fact, a reasonable summary of or at least...
Caveat: I haven’t thought this through very carefully.
I feel like the reason I wouldn’t categorize that headline as “clickbait” is that it is, in fact, a reasonable summary of or at least indication of the contents of the article. It’s not misleading, nor is it simply a question without an answer. It’s true that it is editorialized to garner attention, but at least reading it tells me (accurately) that it describes an article going through the reasons that some Trump voters proudly support him.
That said, I fully agree that distinguishing “clickbait” from “not clickbait” is a difficult problem on an individual basis, and nearly impossible to accomplish at scale.
It also doesn't seem to be misleading the way clickbait often is. The title outright lying or at least heavily implying counterfactual circumstances is what makes good clickbait. It also makes for...
It also doesn't seem to be misleading the way clickbait often is. The title outright lying or at least heavily implying counterfactual circumstances is what makes good clickbait. It also makes for comments deserving of RTFA. Because as much as journalism has sold out, they often still have enough integrity to be more forthcoming with the truth in the actual article.
I believe globalization has brought up some very unique problems with regards to how information is disseminated and how we process it. In the past, when you lived in a much smaller community, you could learn to come to trust particular sources of information when it came to particular matters. As communities grew from towns to villages and cities, new methods of processing information came about. Logical rules to help make sense of whether information could be trustworthy were developed and organizations came in place of individuals - rather than getting your news from the most social and connected people, you would get it from the news organization and you'd get information from books instead of teachers.
However, I do not believe we have come to a point at which we understand, on a society level, how we need to treat globalization as it exists today. What's happening on twitter is really just an amplification of what's happening in societies across the world and has to an extent happened throughout history. People have the inherent biases of thinking they are more knowledgeable than others and that their opinions are more important and that their values are the right things to value. When societies were smaller it was much less likely to run into bad actors in any particular community and so we looked to trusted sources of information for guidance - if you had a question about science, you might go to the local science teacher and ask for their advice. Today, you can find thousands of science teachers online, and invariably some of them will share your same opinion - you will be drawn to people who confirm what you already believe and it will solidify your beliefs.
Because of this we see conversations like the one driven on twitter by this study. Many on twitter are concerned (perhaps rightly) on the negative effects that 'techbros' can have on society and have begun to become aware of how bad actors are exploiting the current globalization to spread misinformation and further realize their own ideals. However, these same concerned individuals are falling into a trap of their own creation - they believe in their own ideals so strongly and have such a desire to prevent society from falling back into demise that they are drawn to the perpetual echo chamber that they exist in, only seeking out others who share the same ideas and vehemently convinced that theirs are the most important.
What we're missing today is a smart, methodical way to approach the nearly incomprehensible amount of information presented to us on a daily basis. We need a way to be able to respond both nimbly and adequately to new information as well as a way to understand whether someone is approaching new information in a similar manner. In the recent past we often were able to come to trust sources of news for being more unbiased than they are today. I think we are struggling to find that model today. We have prolific accounts spread about a variety of online sources - prominent tweeters, youtubers, streamers, and so on. Perhaps this will be a part of the new model for information dissemination.
Regardless of what the correct model is, I think the adage of 'never read the youtube comments' is increasingly applicable on larger scale conversations and particularly fast (chronologically) mediums of news dispersal and conversation. In a rush to provide commentary on matters important to the people who tweeted, they managed to spread misinformation. This spread of misinformation to further an agenda is also why I very rarely read articles about scientific papers, especially in the fields of nutrition and health science. I just hope we can get to a point where the system is able to provide adequate sources of information for the interpretation of said science to less scientifically-capable individuals lest we continue to swirl and suffer the downstream consequences of inadequate, malicious, and distorted interpretation to further spread the ideals of the interpreter.
Statistically confirming something like that without Machine Learning would have been insanely difficult. You'd need to manually tag attributed to every single word in multiple languages. That alone, from a methods standpoint, seems significant. Confirming things we already know with interesting or novel methods is useful not just because it adds another bit of evidence to the pile that confirms whether we "know" something, but it also means you're developing a research method that you can apply to other things in the future.
But in that case, the issue wouldn't necessarily be one of not reading the article. It would be lay people who just don't understand how research works well enough to understand what's cool/interesting about something or why.
Overgeneralizing from patterns is a big no-no when doing real scientific inquiry. You may think this is obvious, as a lay person, but to linguists who study language phenomena as a formal science, it is absolutely a mistake to say this obvious. Historical Linguistics undertakes the study of language contact, borrowing, etymology, and semantic drift (among other things) and can show examples where related languages have diverged in terms of distributional, lexical semantics. There are evolutionary processes, esp. phenomena such as narrowing or metaphor, that predict the opposite of aligned word meanings in some situations!
A comparison from evolutionary biology that may help to think about is Darwin’s finches which exhibit divergent mutation and adaptive radiation across a small geographic proximity as influenced by their respective environments. We have linguistic evidence of non-mutually intelligible dialects, such as pidgin languages and other geographically collocated languages, or the poles on dialect chains (see also Pacific Languages: An Introduction (Lynch 1998) ch. 2). We also have instances of language isolates, such as Ainu or Basque. These instances provide interesting counter-examples for predicting lexical similarity based on typological, phylogenetic, or geographical distributions.
But was machine learning used appropriately or inappropriately? Does anyone know? Or care?
This seems like good reason to ignore the article.
Pretty good article, although since the author has written his piece the tweeter mob seemed to have done a 180 in it's opinion on the article, over the span of about 48 hours. I went ahead and subscribed to the author's newsletter, since I dig the writing style. I do think that a plea to read before you form an opinion will largely go unheeded though, it's a ship that's just way too far gone, the battle was lost somewhere during the heyday of digg.
I think this is an atrocious example to pick, for a very real problem.
Do I really care that numbers are more semantically aligned across languages that prepositions?
There are so many more important questions that this sort of rigorous research could have answered.
Such as semantic links between authoritarianism across countries and cultures.
Sure, you might spend up with a whole lot of effort with little to show. But this is what AI is good for, spotting important correlations that humans can't see. Lets use it to solve the big problems.
I think the original tweets (that I did not read) got it right on two accounts. First, this research is wasted applying AI to a trivial problem where we already knew the answer. Second for not reading the paper in the first place.
If it's a novel AI, then applying it to known problems seems like the perfect way to start. It teaches familiarity with the tooling, the data, and any gotchas that may exist in the process. I would guess that these researchers are now much better equipped at applying similar methodology to other problems.
Some of this stems from the fact that nobody has time to go to the original sources for all the things they want to learn about. Sure, that's a better way to do it. But we read articles that summarize detailed scientific papers because we can't read all the papers ourselves, but we still want to have some information about them. Then others comment on the originals, the articles, the headlines, and the reactions of others, etc. This results in the game of telephone that this blog post describes. I don't think it's realistic to tell everyone to go read the original paper. The dynamics of social media amplify the worst responses, unfortunately, but that's human nature. I don't think it's something that can ever be "fixed". But chastising people for not taking time they don't have before daring to speak about something seems like a waste of breath. People like to blow off steam online, and they're going to continue to do it.
There is a more subtle issue, which is that we often use an article as a prompt to talk about something else that is sort of related.
For me this is most noticeable on Hacker News. Until recently, any article about Go became an opportunity to talk about why Go doesn’t have generics even if the article isn’t about that at all. Any article about Amazon becomes a place to complain about counterfeit products on Amazon. And so on. There are certain keywords that attract certain discussions.
I don’t think this can be blamed on headlines being bad since Hacker News does a reasonable job.
Although these comments do get upvotes, it doesn’t seem to be like on Twitter since having lots of karma on Hacker News isn’t something people care about much.
I think these are just things people want to talk about more than the actual article. I don’t think there is any way to stop it other than more heavy-handed moderation to stay on topic, and I’m not sure that’s desirable.
In any medium, humans are going to human. In person conversation usually happens this way as well - people talk about a subject and often times others are reminded of something tangential, related, a personal anecdote, or a feeling and drive conversation in a new direction.
The idea that people need to discuss only the article always seemed somewhat silly to me. Certainly we want to keep comments free of people stating things that are blatantly false, but we shouldn't be discouraging completely regular human conversation just because it's not what's discussed in the article but instead something in a similar vein.
I think the problem with online conversation is that it's so vast. There are potentially thousands to millions of people looking at the same thing, so it's not surprising that an absurd amount of conversation can happen. The problem is that everyone wants to see the conversation they came to the article for, or they came away from the article with and that can be hard to find in the noise of everyone else discussing. The way you moderate discussion is intrinsically tied to how large the audience is. When it's large enough, it makes more sense to split conversations into their own posts and threads for logical sake. When it's a smaller community like this, however, I think its fine to leave multiple conversations going at the same time.
Yeah... I think there's another fundamental problem that just boils down to it being a numbers game based off of "absorption" time. Content that is easier to make snap judgements about will almost always get more "points" and shares than content that requires more time and thoughtfulness to genuinely react to. If it takes ten seconds to read a headline, have a feeling about it, and share it to one other person versus ten minutes to absorb an article and form and share an actual opinion, in the time it takes someone to read the entire article and post their thoughts there are dozens of other folks who have already "finished" with it to their desired depth of engagement.
Then reading these hot takes in lieu of the actual primary source is kind of like having your food pre-chewed for you. I didn't read the article, but I presume that this person did, and that's almost as good, right? So I can just share their take, bam, and then I can move on to consuming the next thing!
And there's always a next thing to move on to. The quicker you can move on from your previous take, the sooner you get that next dopamine hit for getting involved in the next trend/drama/post/etc. The internet is a never-ending treadmill that disseminates content faster than is maybe reasonable, and your attention is a valuable commodity that people compete (viciously!!) for. A lot of smart people spend a lot of time thinking about how to keep you on their app longer, how to get you to share more things with your circles, how to get just a second of your attention or to provoke the smallest engagement. The same kinds of folks who took the clocks out of casinos are now all about endless scrollbars and other tricks to keep you somewhere longer than you originally intended.
I don't think there's any stopping the tide, frankly. Too much money in it. On an individual level I at least try to be aware and I try to resist my own urges to over-consume, and I try not to post or share glib/useless/mean stuff. I think the environment has harmed me, in a way. But I don't think I can stop anyone else, so that might be as much as I can do.
Very well said, and very much agreed. I don't know how we fight the tide, either.
Rapid content will always outpace slower content; inflammatory content will always generate more of a response than more measured content. It's not a fair fight even in the abstract, and, as you identified, there are plenty of people designing entire systems around exploiting this imbalance and habituating users to behaviors that are beneficial to the platforms but detrimental to the well-being of their users. Probably the worst part is that this is deliberately opaque, so the platforms that are actively exploiting us can successfully position themselves as allies rather than adversaries. The arms around us have hands picking our pockets, but to us they simply feel like a hug.
I think the internet operates fine in this regard. The problem is a human problem. Without a robust measure of intrinsic content quality, any aggregator like Reddit, Twitter, Tildes, Stack Exchange, etc. can only surface quality content and filter poor quality content as well as the community/moderators can. That is, it’s a very hard (impossible?) problem to objectively measure quality, so it’s much more efficient to use aggregate, subjective measures (such as counting upvotes/downvotes/likes/claps etc.).
There are certainly other models for curating content, such as the ways that peer-reviewed academic journals work, but they are so much slower and work-intensive, and domain-specific that they really aren’t feasible at larger scales. The only large-scale, generalized efforts I’m aware of that approximate that sort of model are Wikipedias, but even they have issues of vandalism and "notability" restrictions. I suppose there are other domain-specific wikis and KBs that may also use a similar model of crowd-sourced peer-review, but they also tend to be domain-specific.
Ban clickbait articles. Outlets and articles that attract the kind of answers you refer to... Maybe.
Or if you want to be more lenient here, at least edit the title to include the important information of [clickbait, important context needed] by mod action. That should tip off the careless reader at least a little bit.
But how do you define clickbait? Headlines exist to seek attention on purpose.
"These Voters Back Trump Unapologetically. Here’s Why." Is the top NYT article right now, and it's almost begging you to click on it.
Caveat: I haven’t thought this through very carefully.
I feel like the reason I wouldn’t categorize that headline as “clickbait” is that it is, in fact, a reasonable summary of or at least indication of the contents of the article. It’s not misleading, nor is it simply a question without an answer. It’s true that it is editorialized to garner attention, but at least reading it tells me (accurately) that it describes an article going through the reasons that some Trump voters proudly support him.
That said, I fully agree that distinguishing “clickbait” from “not clickbait” is a difficult problem on an individual basis, and nearly impossible to accomplish at scale.
It also doesn't seem to be misleading the way clickbait often is. The title outright lying or at least heavily implying counterfactual circumstances is what makes good clickbait. It also makes for comments deserving of RTFA. Because as much as journalism has sold out, they often still have enough integrity to be more forthcoming with the truth in the actual article.