Ohai Movie Critics,
Turns out you are no longer needed!? Isn’t that great? No longer do we need to tediously watch movies and then provide context, consideration or critiques. That’s right – just like last week when we got rid of friends and family to help us with our judgement and life choices, this week it turns out we can totally automate movie critics.
All we have to do, is wait 20 years, and then go back and figure out how often a particular movie was referenced by other movies! Sweet!
Of course…using this metric, I feel like there is a damn good chance that either Bill and Ted’s Excellent Adventure or Wayne’s World is going to make the list of All Time Most Influential Movies Ever.
The results of the study itself are pretty straightforward,
[blockquote cite=”AAAS” type=”left, center, right”]The researchers found their automated method of movie citations is better at predicting greatness, especially in movies 25 years old or older, than these runners-up: the expertise of movie critics (a group of critics or a single critic), the wisdom of the crowd, the numbers of awards won and the amount of box office sales, among others.[/blockquote]
Their category for greatness? A movie’s inclusion in the National Film Registry. Okay sure. The problem is, while the method may be a better predictor, people tend to want information about a movie while it is still, you know…in theaters. Or hell, even while it’s still in their iTunes “New and Notable” category. Not to mention, we’re pretty sure most critics (or the general public for that matter) would be able to accurately predict that the Wizard of Oz was a damn good movie 25+ years after the fact.
The study’s authors make a point of saying that – hey, this isn’t really for movies, we’re trying to use this method to find out what scientific papers are most important, based on a complex citation algorithm – except after 13 years of college, I’m pretty sure this isn’t new. I really hope that somehow, someone has buried the lead here and just done a poor job of writing up this study, because otherwise, the Army (who funded this study) may be wishing they’d just asked around a bit.
Here’s your algorithm: What are the most rigorously peer reviewed journals in the field? What are the “leading” journals in the field? How many times was a paper cited by other papers in those journals? Over what length of time? Bob’s your uncle. It’s called a systematic literature review, relies on the use of a citation index, and hot damn are there a lot of them out there.
It’s called quantitative research. There’s some classes you can take. Call us.
The article referenced in this post is: M. Wasserman, Xiao Han T. Zeng “Cross-evaluation of metrics to estimate the significance of creative works.” Proceedings of the National Academy of Sciences, 112(3)