CM Final Project Documentation

Lessons Learned:

  1. If completion is your goal, change directions early on in the project. Even if you’re making baby steps with an error-ridden code/stubborn website, move on. Replicate and modify a tried-and-true code.
  2. Don’t try to follow the code of the smartest kid on your class. Just don’t.
  3. Don’t commit to something just for the sake of the project. Whatever you choose should be at least mildly interesting.
  4. Avoid Instagram like the plague when doing data scraping (it gets updated to frequently to avoid being scraped, so example codes on the web often won’t work.

The Original Plan

I’ll probably be burned on a stake for saying this, but I hate museums. There is nothing more unappealing to me than spending an afternoon looking at old paintings by dead white guys (however, if the paintings are by dead women, I might perk up juuuust a bit). Perhaps I’m too low-class for culture. Because I am an intelligent human being, I decided to base my project on the object of my hatred (because there really isn’t much else to do in Paris besides go to museums anyway). My plan was to capture other peoples’ museum experiences by scraping the web for images and text (i.e. tweets and/or Instagram photos posted by museum goers) and somehow present it. Early iterations of the plan included projection, and the idea of VR was tossed into the mix

Problem number 1

(The first problem was actually not remembering anything from the museums I’d been to, but i remedied that by finding the app from a Centre Pompidou and picking a few pieces from there. Thanks Matt and Marianne for the suggestion that saved me from having to stand in the monstrous Pompidou line.)

Museum goers are just as boring as the museums themselves. Because my roommate and I have a habit of posing with paintings/sculptures and/or giving them new captions, I assumed there would be other people on the internet doing the same. I was horribly mistaken. The only captions I found were along the lines of “[x] painting by [y] artists at [z] museum. *yawn* I decided to reboot: I’d choose a bunch of paintings by a specific artist and surround his works with a bunch of related text from social media in a VR environment (it is worth noting here that at this point I’d never worked with VR before. I’ve never even experienced VR. Occulus Rift might as well be the tooth fairy).

Problem number 2

I tried to use Richard and Nicole’s code as examples. The server that hosted Nicole’s code had expired, and she was only able to retrieve partial code from her email. I tried to tweak that, to no avail. Richard’s code was way over my head. After discussing it with him and spending many hours trying to pick it apart, I began to understand some things, but I felt like a Mario Kart noob trying to take on Rainbow Road.

Problem 3

In my desperation I kept making stupid mistakes, like forgetting to import Beautiful Soup when I ran the code. Because windows users seem to be the minorty at NYU/SH/NY/AD, and because Windows- friendly documentation is practically impossible to find, I’ll post a reminder: MAKE SURE YOU’RE ACTUALLY FOLLOWING WINDOWS DOCUMENTATION, AND NOT MAC DOCUMENTATION THAT IS MASQUERADING AS WINDOWS-FRIENDLY.

Problem 4

This is an important yet underrated point: take care of yourselves. Illness, undernourishment, and exhaustion got the best of me. I quite literally crashed at my computer one night and woke up barely in time for class, with nothing to present but a broken soul code.


Nuts and Bolts

I used A-Frame to attempt to create the VR, and to scrape instagram I attempted to use selenium and Beautiful Soup.

This code pretty much just opens Instagram.

Screenshot (59)

this code creates my floating boxes in A-Frame.

Screenshot (60)

These are my floating boxes

Screenshot (61)




A-Frame Github

Another helpful A-frame documentation

Selenium Documentation

Selenium with Python