Live data from Hacker News

Please read the paper before you comment

buttondown.email

71–80 of 341 posts

Re: Please read the paper before you comment

#71
post #55

From the title this seems to be a violation of HN guidelines. "Please don't comment on whether someone read an article."

The title is the verbatim title of the linked article and the article is about the topic of people not reading papers when discussing them. I don't see how you'd come to the conclusion it's an inappropriate title.

Re: Please read the paper before you comment

#72
post #48

Some years ago, Ars Technica ran a story on guns. https://arstechnica.com/science/2011/04/guns-in-the-home-lot... About halfway through a one-page article was this quote. "If you have read this far, please mention Bananas in your comment below. We're pretty sure 90% of the respondants to this story won't even read it first." Sure enough, there were about 3 pages of comments before someone mentioned bananas. Read the…

What it demonstrates is that people have already made up their mind on the topic and aren't genuinely interested in new perspectives. This helps explains the inflammatory nature of Internet debates in general.

Re: Please read the paper before you comment

#73

The key to cleaning up my Twitter feed was to unfollow people who were too good at Twitter. That is: People who are working to maximize engagement, follower count, retweets, and likes. In my feed, the highest signal to noise ratio comes from people who aren't commenting on any and every trending topic. Instead, they're busy doing real work, away from Twitter, only occasionally checking in to share something. Likewise…

It doesn't even matter if they're good at it. Anyone who is even trying to "win" at twitter is just going to actively make your feed worse. It's not just that they spam you with low-value tweets, but also that following them helps ensure that you land in that awful filter bubble of people who see twitter as a competition.

Re: Please read the paper before you comment

#74
post #2

When I first learnt to program the thing that I really enjoyed was learning to make a terminal app that took in the users input and used it as a conditional. I’d make all these little toy apps where my family had to type in the right word to continue. Eg “Who is the leader of the autobots?” => “Well done. Transform and roll out!!” It strikes me we could encourage better discourse with one word locked comments that th…

> Micro quiz questions, one up on ‘are you human’ to ‘prove you read one key thing’

I have gotten good results from including these in my expository writing. Having the reader stop and go through actively using the material at even a basic level seems to really help them integrate and retain it. They're multiple choice questions embedded throughout the text.

There's no reason you can't force everyone to correctly answer them before being able to comment. That being said, writing good incremental questions like this is actually quite a bit of work.

Re: Please read the paper before you comment

#75
The problem is people want their internet points. If you take the time to read, and then comment, your comment won't get enough visibility on all social media platforms, especially if dozens or hundreds of other comments quickly begin appearing. You have to 'show up early' to get visibility. Optimizing for that dopamine hit, users choose to maximize points over thoughtful discussion.

Re: Please read the paper before you comment

#76
post #40

Earlier quoted context omitted.

So you're saying we need a platform with both characters/post and post/day limits?! I mean... that's beautiful, because I created a second twitter account just so I can follow a select few people for the pretty much the same reason you gave without having to worry they're getting lost to the noise.

The social network I honestly want is something akin to the tradition of Christmas cards: a rare update (at most once a quarter) about how life is going with the option of attaching a few photos. No comments section, but instead a "I'd like to get in touch and chat sometime" button which helps schedule a video chat. No retweets, shares, engagement metrics.

The ideal social network in my opinion is instant messaging. Its the only major platform which hasn't been infested with algorithms, marketing, "influencers" and all the other crap. Its just you and close contacts. The downside is you won't see random updates from people you don't interact with regularly but honestly its not the biggest loss.

Re: Please read the paper before you comment

#77
post #40

Earlier quoted context omitted.

So you're saying we need a platform with both characters/post and post/day limits?! I mean... that's beautiful, because I created a second twitter account just so I can follow a select few people for the pretty much the same reason you gave without having to worry they're getting lost to the noise.

The social network I honestly want is something akin to the tradition of Christmas cards: a rare update (at most once a quarter) about how life is going with the option of attaching a few photos. No comments section, but instead a "I'd like to get in touch and chat sometime" button which helps schedule a video chat. No retweets, shares, engagement metrics.

I like the idea, but how is that different from a personal website/blog with private access?

Re: Please read the paper before you comment

#78
This also goes the other way. Someone who people think of as a thought leader or a science communicator says something that seems reasonable, but is not actually based on any science gets amplified and soon if you even dare to ask for evidence you are painted as "the enemy" and "russian bot" for daring to think on your own and actually request papers on the matter.

Then people will start linking you papers that are only tangentially related to the question at hand and get offended when you actually go and read the paper and point out that the paper has (almost) nothing to do with the actual question at hand.

Re: Please read the paper before you comment

#79

Earlier quoted context omitted.

Some users (including myself) disable the :visited command.

Thinking of reasons one would want to disable that caused me to do a Google search and find this: https://developer.mozilla.org/en-US/docs/Web/CSS/Privacy_and... I guess it can't be observed from a script? That would be a neat way to get potentially sensitive info.

As mentioned in the article, firefox also limits what you can actually do with the visited selector so there is no way you can use css to change something which is detectable with js so the comments hidden thing won't work.

Re: Please read the paper before you comment

#80
post #64
post #48

Some years ago, Ars Technica ran a story on guns. https://arstechnica.com/science/2011/04/guns-in-the-home-lot... About halfway through a one-page article was this quote. "If you have read this far, please mention Bananas in your comment below. We're pretty sure 90% of the respondants to this story won't even read it first." Sure enough, there were about 3 pages of comments before someone mentioned bananas. Read the…

> Read the paper? You're lucky if they read the article This is natural if you follow the amount of effort required. Effort to comment ~ close to zero. Effort to go through an article worth several pages (or worse, a paper!) without dropping off is much, much higher.

Even lower effort/risk - make a comment in response to a wrong-headed comment from another commenter that hasn't read the citation. All you have to know is that their comment is wrong, whether or not it has any relation to the new information.
Post reply on HN