Does anyone have anything resembling statistics on how often the HN auto-de-clickbaiter actually improves article titles? I only notice it when it wrecks them, which it seems like it does really often, but maybe there are dozens of cases where it's just quietly making things a little bit better. Or not.
[EDITED to add:] Of course I'm not suggesting an automated system that assesses which changes are improvements, or anything like that. But maybe there's something somewhere that logs every title change that gets made, and every now and then someone could look at the most recent 50 and see whether it's making things better or worse on net.
1. A blogger writes a short clickbait title with no information about what the article is about. (The real Miss Julie)
2. The hacker news de-clickbait filter takes it from a generic, no context title to something grammatically incorrect and nonsensical. (The Miss Julie)
3. Someone changes the title but still doesn't give any context.
"Fish swim in the ocean" becomes "Fish the ocean" becomes "Scientist observe fish".
I don't think "The real Miss Julie" is a clickbaity title.
It's a perplexing title if you haven't heard of Strindberg's play, for sure, but that's no different from a title like "Improving Kubernetes" or "Riemann Hypothesis proved", which will make sense only if you know about Kubernetes or the Riemann Hypothesis.
If you _have_ heard of Strindberg's play and make the connection with the title, then the obvious interpretation is something like "it turns out that there was an actual person Strindberg's character was based on, and this article is about her". Which, as it happens, is exactly correct.
I think it's worth distinguishing between "I can't tell what this is about from the title, because it's clickbait" and "I can't tell what this is about from the title, because I happen not to be familiar with the subject matter". This is the second of those, not the first.
(For the avoidance of doubt, I am not in the least suggesting that there's anything wrong with happening not to be familiar with the subject matter.)
hat's no different from a title like "Improving Kubernetes" or "Riemann Hypothesis proved"
It's completely different because there is no context, so it could be about anything, especially on a general technology 'news' site.
"Riemann Hypothesis" is going not ambiguous and neither is anything with "Kubernetes" in the name. A generic sentence with a generic first name is. I think this is obvious to most people.
HN is not a technology site. Not exclusively, anyway. Some days not even primarily! HN is an aggregator, so we pluck articles from varied contexts all the time. "Intellectual curiosity".
But "Miss Julie" has exactly as much inherent context as "Riemann Hypothesis", or "Kubernetes". In all cases, a web search will return, as the first result, the correct reference.
If we took a random sampling of people off the street, I'd be curious to see how many would recognize Miss Julie vs the other two. I lean toward more for Miss Julie!
Anyway, what's a better title? "The Real Miss Julie (being a character in an eponymous 1888 play by Swedish playwright August Strindberg)"?
Bad examples. Anyone named Riemann might hypothesize on any subject. And Kubernetes is a word in Ancient Greek.
I had a Natural History professor in college named Riemann. He was a Darwinist, the irony! And I've met Greek scholars in the same venue.
At some point, you just have to say "My liberal arts education was cut short before we got to 19th C Scandinavian playwrights". It's fine. You're in excellent company.
(At the risk of offending the canon, I have to say that I really did not enjoy Strindberg, and although I acknowledge the cultural significance, I would happily exchange those hours for one kanelbulle.)
This is exactly what I thought you would reply. You were saying the other two were ambiguous, now are you saying that a wikipedia page means they aren't ambiguous?
At some point, you just have to say "My education was cut short before a class on logic".
Also:
Hacker News (HN) is a social news website focusing on computer science and entrepreneurship.
No, I'm saying that there's nothing broadly ambiguous about any of these three references, but that of course you can find people with a localized confusion for each.
Your position seems to be that Miss Julie is somehow different or something obscure, and I disagree. It's just an instance of localized confusion.
Also, more authoritatively than the 'pedia:
> On-Topic: Anything that good hackers would find interesting. That includes more than hacking and startups. If you had to reduce it to a sentence, the answer might be: anything that gratifies one's intellectual curiosity.
Anyway, this is a silly conversation. Now you know about Miss Julie. I wish I could say you'd enjoy reading it, since we've spent so much time talking about it. Unfortunately, I didn't enjoy the play when I read it many years ago. It is well-respected though, and perhaps I should give it another chance in my adulthood.
> But "Miss Julie" has exactly as much inherent context as "Riemann Hypothesis", or "Kubernetes". In all cases, a web search will return, as the first result, the correct reference.
Hmm, yes misapprehension indeed. Ergo, visa vi, concordantly, find me anywhere on any site where someone sees "Riemann Hypothesis" and asks 'which Riemann' or "Kubernetes" and asks 'is that the greek name?'
Oh man, you're killing me here. Too deftly ironic for trolling, so I'll respond, for your entertainment!
I work in Fintech, and I use vim. You just blew out my context scope.
But I see this is more localized confusion. Your liberal arts curriculum also didn't extend to Latin or French apparently. Or you can blame autocorrect if you like!
vis-à-vis
Also, seriously ironically: I first learned that phrase from the same guy who made me read Strindberg in high school (he also taught French). There's something going on here.
Re: the rest: you're still missing the point, so I've nothing more to add.
Your interpretation of HN's title conventions is inaccurate, titles are not required to have 'context' and that alone does not make them clickbait. That's never been a thing - it might not be to your taste but your 'pipeline' is just plum wrong, e.g. https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...
> > "Miss Julie" has exactly as much inherent context as "Riemann Hypothesis", or "Kubernetes"
> No it doesn't. Let's think super hard. "Miss Julie" could be anyone named Julie.
That is, frankly, bullshit. "Let's think super hard"... "Romeo and Juliet" could be any two people named Romeo and Juliet. But that's not what you thought when you saw the names, is it?
Just because your Allgemeingbildung is sucky enough to only know (or at least have vaguely heard of) Shakespeare, but not Strindberg, doesn't make "Miss Julie" any more generic than "Romeo and Juliet". It just means you need to work on widening your literary horizons, not just your hacker ones.
Genuine interest: why is the title considered clickbaity?
I understand that the automated system transformed it and in this case would consider the automation to be wrong, however I don't really see the article itself as clickbaity.
[EDITED to add:] Of course I'm not suggesting an automated system that assesses which changes are improvements, or anything like that. But maybe there's something somewhere that logs every title change that gets made, and every now and then someone could look at the most recent 50 and see whether it's making things better or worse on net.