Exactly. If they had limited it to cases where "it" is referring to Pennywise, that would be one thing, but not when anyone uses a very common pronoun!

I think it's quite interesting to include. Apparently Barbie says "Barbie" more than It says "it", which is fascinating!

I’ve never seen It, but having seen Barbie, it’s not all that surprising lol.

whole conversation makes me just think: But how can we not say it, if we don't know what it is!

Ni!

I guess this is the downside of making a data analysis thing as a side project to hopefully get something going, but not having the time to take care of all potential edge cases.

I guess "Them!" is also affected by this, and maybe The Thing or The Birds...

This is IMO one of the coolest use cases of AI. With a half-decent prompt, an LLM is pretty good at tasks like those.

Both you and GP seem to have stopped reading the article early...

He specifically calls out `"real"` title drops just a few sections later.

To be fair, the article starts out seeming real for about the first third. It's only after the first list — Barbie, Damini, Sita,... Azhar, It — that it descends into obvious parody. Quote:

"What's interesting about the (Fiction) list here is that it's pretty international: only two of the top ten movies come from Hollywood, 6 are from India, one from Indonesia and one from Turkey. So it's definitely an international phenomenon."

Here the writer slides seamlessly from talking about movies with title drops to talking about movies with single-word titles which are also the name of the main character, but is still saying things like "What's interesting about this list..." and "...an international phenomenon," as if those are remotely the defining characteristics of the list he just gave. (The defining characteristic, again, is "movies named after the protagonist." That's all.)

Then there's a section break. Since the article clearly outed itself as parody right before the break, I think it's totally reasonable for anyone to stop reading it at that point. (Although maybe not 100% reasonable to come back and comment on HN about it, except maybe to express disappointment and save other people the bother of reading that far themselves.)

Anyway, after the break the author says, "You might have noticed [an icon on each movie that is] named after one of its characters." But scroll back up and you'll see that icon is missing from 4 of the movies in that list of 10: "Saina", "Nussa", "Arif v. 216", and "It". Of those 4, 3 are clearly named after a main character. The fourth (like "Ecks vs. Sever") is named after two characters (Arif and 216) but the graph shows that the author is counting instances of the name "Arif" alone, not instances of the phrase "Arif v 216".

So not only is the article trying to be funny, it's not even playing by consistent rules — it's a parody of an academic paper but also just flat-out lying about the data! That's not only annoying but uncool.

I would actually be interested in reading a real article on the phenomenon of title drops in movies, e.g. by someone who'd gone through a bunch of movies and tallied which of them contain title drops. But the linked article is just garbage.

Yes. I would expect the writer to at least check the top 10 movies.

While scrolling, I was surprised by Arif V 216. As I've seen the movie, there is no chance of such a high count of title drops. Skimming the instances, it was just showing the name "Arif." I returned to HN comments to check whether it was mentioned, and only you did.

Cool idea, lazy execution.

I think it's just a guy that dug around in some data for fun and wanted to show it to people.

I'll bet he ingested a file full of known real names in order to know whether the title was a character's name, thus "Nussa", "Ecks vs Sever", etc not showing as "titles matching names".

The sibling comment that he should have at least manually checked his top 10 list is VERY fair, however.

after doing a naive approach he then drills down into more proper title drops.