Is it hubris for HN to routinely change published titles of articles?
As a regular reader and submitter, I'm always struck by how, on a regular basis, titles published by reputable newspapers, magazines, and — less frequently, but not never — scientific journals are edited and reworded and sometimes completely redone by HN moderators.
10 comments
There are a few guidelines about titles:
> Please don't do things to make titles stand out, like using uppercase or exclamation points, or saying how great an article is.
> If the title includes the name of the site, please take it out, because the site name will be displayed after the link.
> If you submit a video or pdf, please warn us by appending [video] or [pdf] to the title.
> If the title begins with a number or number + gratuitous adjective, we'd appreciate it if you'd crop it. E.g. translate "10 Ways To Do X" to "How To Do X," and "14 Amazing Ys" to "Ys." Exception: when the number is meaningful, e.g. "The 5 Platonic Solids."
> Otherwise please use the original title, unless it is misleading or linkbait; don't editorialize.
See https://news.ycombinator.com/newsguidelines.html
> Please don't do things to make titles stand out, like using uppercase or exclamation points, or saying how great an article is.
> If the title includes the name of the site, please take it out, because the site name will be displayed after the link.
> If you submit a video or pdf, please warn us by appending [video] or [pdf] to the title.
> If the title begins with a number or number + gratuitous adjective, we'd appreciate it if you'd crop it. E.g. translate "10 Ways To Do X" to "How To Do X," and "14 Amazing Ys" to "Ys." Exception: when the number is meaningful, e.g. "The 5 Platonic Solids."
> Otherwise please use the original title, unless it is misleading or linkbait; don't editorialize.
See https://news.ycombinator.com/newsguidelines.html
>unless it is misleading or linkbait
Ah, therein lies the rub.
Ah, therein lies the rub.
Setting aside the guidelines against clickbait titles or editorializing, one has to consider the tendency on HN for readers to only engage with the title, and not read the article at all. Assuming for the sake of argument that most titles aren't egregiously misleading or clickbait to begin with, many might still be triggering and need to be toned down.
Also, HN limits all titles to 80 characters, because I guess everyone is reading this site on a terminal, so any title longer than that has to be edited.
What I don't like (besides the character limit) is when they change the URL and throw the context of the thread off. There ought to be some way to link the history of comments to the history of editorial changes but I can't think of a solution that HN would consider elegant.
Also, HN limits all titles to 80 characters, because I guess everyone is reading this site on a terminal, so any title longer than that has to be edited.
What I don't like (besides the character limit) is when they change the URL and throw the context of the thread off. There ought to be some way to link the history of comments to the history of editorial changes but I can't think of a solution that HN would consider elegant.
I like the character limit because it forces me to really think hard about the piece in order to cram, say, a highly technical, oftimes abstruse 200-character scientific journal article title into the 80-character HN space.
HN should be able to handle a highly technical article title, though. And if people can't, then someone can explain it in the comments.
It becomes a problem when you lose necessary context in fitting the title in the limit, causing confusion or misinterpretation of the title. I've seen arguments started over it. There's one current example here[0] where "India’s Yes Bank" was shortened to Indian Bank" in the title to make up for the original being two characters over the limit. Turns out there's an actual "Indian Bank[1]" and now the title is misleading. All because we can't have a title with 82 characters.
Granted, it doesn't happen often, but every time it does happen, it would have been trivially easy to avoid by either not having a character limit for titles, or having a more reasonable one like 120 characters.
Edit: also apparently just happened here[2].
[0]https://news.ycombinator.com/item?id=22502649
[1]https://news.ycombinator.com/item?id=22504863
[2]https://news.ycombinator.com/item?id=22504837
It becomes a problem when you lose necessary context in fitting the title in the limit, causing confusion or misinterpretation of the title. I've seen arguments started over it. There's one current example here[0] where "India’s Yes Bank" was shortened to Indian Bank" in the title to make up for the original being two characters over the limit. Turns out there's an actual "Indian Bank[1]" and now the title is misleading. All because we can't have a title with 82 characters.
Granted, it doesn't happen often, but every time it does happen, it would have been trivially easy to avoid by either not having a character limit for titles, or having a more reasonable one like 120 characters.
Edit: also apparently just happened here[2].
[0]https://news.ycombinator.com/item?id=22502649
[1]https://news.ycombinator.com/item?id=22504863
[2]https://news.ycombinator.com/item?id=22504837
Twitter wouldn't exist if it weren't for the character limit.
Twitter's character limit was originally intended for posting on SMS, and remained long after it was even relevant. And they extended the limit recently anyway. And people work around it by linking multiple posts, so it doesn't even really matter.
And... I'm specifically talking about title length, not article length. Tweets don't even have titles.
And... I'm specifically talking about title length, not article length. Tweets don't even have titles.
Personal view: While it may not feel ideal, and initially the thought of it happening did feel off, I feel that the title changes are actually very helpful as it often results in clearer titles, which sometimes aren't very to-the-point or are unnecessarily sensationalized.
But, but — the original titles were created by either authors and/or editors who surely must be respected for their professionalism.
My reply may seem a bit off-topic: Whenever I think of titles of news articles (not research papers though), I often think of how some news organizations have pretty ad-hoc processes of coming up with headlines (titles), for instance this process outlined by Ars Technica [0]:
> Staffers have various methods for coming up with headlines. I prefer to think of my headlines first after researching but before writing the piece—not only does this mean I’m attacking the headline with my creative batteries fully charged, but the act of figuring out the headline helps also give the piece some shape in my head.
> Other folks have other methods, and we keep a #headlines Slack channel running for staffers to help workshop ideas. If you’re having trouble with a headline—or you can’t quite think of the right approach—everyone available lends a hand.
> The overall objective is to whittle the choices down to a pair of punchy but informative headlines that give readers a reason to click without being clickbait. The two headlines then both run on the front page in an A/B test, which shows different headlines to different visitors. The software uses statistical math to pick a headline "winner" based on which headline is selected more often by the visitors who see it in the first few minutes. A/B testing headlines gives us some idea of what types of headlines resonate best with different kinds of pieces; we use what the A/B test tells us along with a number of other data points to help us write better headlines.
This is not to imply that the headlines which are picked are non-optimal, but that there is a drive to not only experiment with headlines but also to get help from their peers on a dedicated Slack channel for it, as picking a good headline is not easy. While you're correct in stating that there needs to be respect for the authors and ideally thus not to change their headlines on HN, it very much seems that picking a headline in general is inherently tough as it involves condensing an article's overall point into less than 80-100 characters, which could result in non-optimal titles at times, and in some cases possibly lead to poor titles.
[0] https://arstechnica.com/information-technology/2020/01/no-of...
> Staffers have various methods for coming up with headlines. I prefer to think of my headlines first after researching but before writing the piece—not only does this mean I’m attacking the headline with my creative batteries fully charged, but the act of figuring out the headline helps also give the piece some shape in my head.
> Other folks have other methods, and we keep a #headlines Slack channel running for staffers to help workshop ideas. If you’re having trouble with a headline—or you can’t quite think of the right approach—everyone available lends a hand.
> The overall objective is to whittle the choices down to a pair of punchy but informative headlines that give readers a reason to click without being clickbait. The two headlines then both run on the front page in an A/B test, which shows different headlines to different visitors. The software uses statistical math to pick a headline "winner" based on which headline is selected more often by the visitors who see it in the first few minutes. A/B testing headlines gives us some idea of what types of headlines resonate best with different kinds of pieces; we use what the A/B test tells us along with a number of other data points to help us write better headlines.
This is not to imply that the headlines which are picked are non-optimal, but that there is a drive to not only experiment with headlines but also to get help from their peers on a dedicated Slack channel for it, as picking a good headline is not easy. While you're correct in stating that there needs to be respect for the authors and ideally thus not to change their headlines on HN, it very much seems that picking a headline in general is inherently tough as it involves condensing an article's overall point into less than 80-100 characters, which could result in non-optimal titles at times, and in some cases possibly lead to poor titles.
[0] https://arstechnica.com/information-technology/2020/01/no-of...