5 ms·
A Peek Into Netflix Queues
- pingswept 17y agoIf you look at the Boston map, there is one zone just outside route 128 near Lexington that consistently contrasts with its neighbors. That's Hanscom Air Force Base. It's particularly stark with the movie "Milk." The other areas of sharp contrast I noticed are Mattapan and Lynn. Interesting stuff.
- boundlessdreamz 17y agoGreat observation. Didn't expect twilight is the top movie in an airforce base!
- blasdel 17y agoThey are plenty of women in the Airforce, members often have families with children, and the Airforce skews strongly towards Evangelical Christianity.
- breck 17y agoBoth of the "Tyler Perry" movies (3/4 toward the Least Rented side), also show a sharp contrast along racial demographic lines. All in all, a very interesting visualization. This stuff redeems my love for the NYTimes.
- psadauskas 17y agoSame thing if you pick Denver, and pan NW slightly to see Boulder. Denver and Boulder are complete inverses.
- elblanco 17y agoI noticed a similar, but exactly opposite trend with Paul Blart: Mall Cop.
- fuzzmeister 17y agoIt's amazing to see the socioeconomic divide at play in this area (I'm from Concord). If you look just northeast of Hanscom on the map for Milk, you'll see a sharp dividing line between the richer towns of Carlisle, Bedford and Lexington and the more working-class towns of Billerica, Tewksbury and Dracut.
- amadiver 17y agoIs this data available to everyone? I'd love to see it for my area (Providence, RI). It might be interesting to compare this data with housing prices.
- thumper 17y agoI was looking around for the answer to this question, too. I was a little surprised at the ordering of the top couple of movies, and that got me wondering if "most popular" was normalized somehow. Are these the most popular from 2009? Were the numbers normalized for how long the movies were available via DVD? If there's no normalization, then this gives a snapshot about movie popularity... something like a 1-D vertical slice from a visualization like one of these two: http://www.nytimes.com/interactive/2008/02/23/movies/20080223_REVENUE_GRAPHIC.html http://www.nytimes.com/interactive/2008/02/23/movies/2008022... http://xach.com/moviecharts/2008.html http://xach.com/moviecharts/2008.html I don't want to knock this new visualization, but I'm wondering what inferences would be safe to draw from the data as presented. It seems like it is most useful for comparing regions to each other; for instance, I was surprised at how varied the movie ordering was for adjacent neighborhoods in Manhattan.
- elq 17y agoif you're a netflix member you can get a (not so close) approximation at http://www.netflix.com/LocalFavorites http://www.netflix.com/LocalFavorites
- manvsmachine 17y agoLooking at the map for my city of residence (Atlanta) confirms a pattern I've observed in real life: the black community generally will support its own, seemingly regardless of the details of the performance. Look at the patterns for The Soloist (Jamie Foxx), Lakeview Terrace (Samuel L. Jackson), Traitor (Don Cheadle), Obsessed (Beyonce), and the Tyler Perry movies. These movies have little to nothing in common, and their Metacritic scores vary widely, yet their distribution maps are essentially identical. A further point of interest: judging by these patterns, Atlanta really is as segregated as it is often made out to be. The only other city that I noticed with such stark contrast was D.C., which I've heard is another heavily segregated city (at least culturally).
- pvg 17y agoIt's probably my inner black-hearted cynic coming through but I thought this would be a visualization of how Netflix manipulates queues to favour new and recently signed-up users - something that's been the subject of anecdotal write-ups and small scale tests over the years. These are interesting and pretty in their own right although I already knew most of my neighbours are very fond of 'Mad Men'.
- unohoo 17y agothis is one of the best visualization I've seen in recent times. Wonder why Netflix doesnt do this on their website itself - an year end wrap-up on movies
- sachinag 17y agoThe social proof of having this in the NYTimes is worth a lot more than the SEO value of hosting it yourself.
- Wump 17y agoI wonder if one could use the rank of "Milk" as a predictor of the conservativeness/liberalness of a particular area.
- maukdaddy 17y agoThis might be the best viz I've seen on the NYT. Visualizing movies like Tyler Perry, Milk, etc. in Chicago correlates almost perfectly with socio-economic boundaries.
- RK 17y agoI wonder if anyone will use this data to help decide what part of a city they want to reside in...
- sid- 17y agoThough its interesting information, I think the data could help their competition like REDBOX to stock particular dvds at their locations