Uncalculated Variables

or, how algorithms just don’t get me, and in fact, may be insulting my intellegence and taste in movies.

I was recently catching up on one of my favourite podcasts, The Film Vault. While this podcast may be an acquired taste for some due to the caricatures the hosts put on of themselves, it is nonetheless a quality look at a diverse range of movies. Of course, this podcast is mostly beside the point, as the point is mostly about something they mentioned that I have now spent most of my free commuting time on: Taste.io.

Since last year’s change to the Netflix rating and recommendation system, I have been without a way to decide what to watch next, since you know, we have been allowing computers to make that decision for us. This is where taste.io comes in. Essentially, it is a website/browser app that makes recommendations based on a four point rating system, along with a slightly more detailed “attributes” system. Of course, you have to rate hundreds of movies first, or it’ll just recommend popular movies.

This I did with abandon. According to the handy dandy statistics page I have rated 973 movies. Their computer has helpfully informed me that I have spent a minimum of 2 months and 14 days of my life watching movies. Of course, this doesn’t include repeat viewings or any number of movies it just hasn’t shown me yet to rate, or which of the approximately 7 trillion Clint Eastwood westerns I cannot remember. Seriously, I know I have seen some, I just am not sure which.

Anyway, wasted youth is not fully what this is about.

As you fill in the gaps in your viewing, it starts to create a personal profile. This personal profile rates your movie personality on 4 scales of juxtaposed traits.

1. Happy vs. Grim
2. Simple vs. complex
3. Calm vs. Intense
4. Modest vs. Cynical

So, after 973 movies, the computer has told me that I am a happy go lucky zen simpleton who doesn’t trust anyone.

WTF!?!!!!!111!

I have posted a great deal about how I wish to see myself, and very few of those words seem to match the man in the mirror. Here are my words: Scowling intense complicated sarcastic jerk.

How am I and the computer so far off? Uncalculated variables my friends.

First, temporal positioning. I saw Backdraft when I was 14. I saw Commando when I was 10. They both had beautiful women and explosions in them. They both featured themes of the past haunting someone. I hated Backdraft and I based my world philosophy on Commando. Thus I rated Backdraft as a meh and Commando as amazing.

These ratings might not be how I would rate them today if I saw them for the first time. We all know some movies just don’t hold up to re-viewings later in life, I am shedding a tear for you Highlander. I am reasonably confident, as an adult I would enjoy the slightly more complex narrative that was probably Backdraft, and might find the boom boom one liner script that was Commando a bit less fulfilling today.

Clearly, since the algorithm did not bother to ask my age at any time, it cares little for at what age you probably saw a movie. I was a kid when I saw Dune, and an adult when I saw Fieval Goes West. I was 8 when I saw Dune, of course I hated that shit. I can’t even remember the damn movie. I can remember where my coked out uncle parked the car though. We were in P2 D2 that day. Of course I remember it, Star Wars was amazing to me at 8. At 41 I couldn’t stand the new Star Wars movie. That’s ok, it wasn’t made for people like me.

The algorithm does not care, or compute. I am simple.

Next we have social emotional status. I was 20 when I saw Lion King. I saw Lion King approximately 346729 times. Why you might ask? Because I was living with a couple with a toddler, and that shit rolled all damn day long, and I was too high to get off the couch. I am not proud of this fact, but sometimes we have to face where we are and what we are doing, even if the computers don’t care. Socially and emotionally I was in a dark place, and my friends had the patience to let me work it out while watching an uplifting can do attitude movie that resonated with me. I had to put my behind in the past, and I had to be told to do so many times before the message sunk in.

The point is, sometimes movies appeal to us not because they are of quality or lasting value, but rather because of who we were when we saw them. I am in fact a grim person based on life experience, but sometimes I like to escape that, or sometimes I need to be shown the light. When being me, I will choose a dark dystopian future over anthropomorphic talking cliches every time.

The algorithm doesn’t care or compute. I am Happy.

Next we have biological purposes. This is going to be a combination of TMI and killing trees (Kleenex comes from trees right?), so please feel free to skip to the end of this section, esspecially if you know me in real life. I am a human being, and as such am flawed and subject to some more base desires. To that end, allow me to mention two movies on the calm versus intense spectrum Career Opportunities and Basic Instinct.

For a teenage boy, these movies offer an attractive woman in various outfits doing various things. One movie instersperses these scenes with comedy and warmth, the other features murders and premature ejaculation. Which one do you think I remember fondly, and which do you think I hated for teaching me about edging and blue balls? To be fair, one was a PG-13 tight sweater fest and the other was more moist and sweaty, but we had fewer choices back then. Hot Spot hadn’t been released yet. The point is, the calm light movie won out in my estimation, not just because Jennifer Connelly is better looking, but because murder is icky and a turn off.

The algorithm doesn’t care or compute. I dislike intense films. Also, the algorithm didn’t judge me, so neither should you.

Lastly, we have the only area in which the algorithm and I see eye to eye, though I am still offended, because I don’t think it is for the same reason. The algorithm was comparing modest with cynical, which on the surface seems to negate the juxtaposition rating. These are not opposite terms on a spectrum usually.