Functional Problem Solving (CSC 151 2015F) : EBoards
Primary: [Front Door] [Schedule] - [Academic Honesty] [Disabilities] [Email] - [FAQ] [Teaching & Learning] [Grading] [Taking Notes] [Rubric] [Remote Access]
Current: [Assignment] [EBoard] [Lab] [Outline] [Reading]
Sections: [Assignments] [EBoards] [Labs] [Outlines] [Readings] - [Examples] [Handouts]
Reference: [Setup] [VM] [Errors] - [Functions A-Z] [Functions By Topic] - [Racket] [Scheme Report (R5RS)] [R6RS] [TSPL4]
Related Courses: [Curtsinger (2015F)] [Davis (2013F)] [Rebelsky (2015S)] [Weinman (2014F)]
Misc: [Submit Questions] - [SamR] [Glimmer Labs] [CS@Grinnell] [Grinnell] - [Issue Tracker (Course)]
Overview
This week we'll be talking about Youtube and its efforts to enforce the rights of copyright holders with its ContentID system. Early this month, a federal appeals court in San Francisco "affirmed that copyright holders must consider whether a use of material is fair before sending a takedown notice".
This ruling is the latest development in an on-going battle between copyright holders, Youtube content creators who use the copyright holders content, and Youtube's system---ContentID---to help automatically enforce copyright in its system.
To understand the situation, please read this Gamasutra article---a primer on ContentID for Youtube content creators:
Then catch up on the background of this latest court ruling:
https://www.eff.org/press/releases/important-win-fair-use-dancing-baby-lawsuit
https://www.eff.org/cases/lenz-v-universal
If you are interested in how Google actually checks for copyright infringement of millions of videos, this algorithmic technique is called fingerprinting. Here is a short technical paper from Google on the approach that ContentID takes (based on finite-state transducers versus traditional hashing methods):
http://cs.nyu.edu/~eugenew/publications/icassp07.pdf