On this episode of Spark: Sensors, Predictors, Recognition Software – oh my! Click below to listen to the whole show, or download the MP3 (runs 54:00).
Podcast: Play in new window | Download (Duration: 54:32 — 50.1MB)
You can also listen to individual stories below.
Facial Recognition
Jennifer Steeves
Recent advances in facial recognition technology are giving us the feeling of being much closer to a “Minority Report” reality. But how close are we really? Psychology professor Jennifer Steeves of York University explains how human beings recognize one another compared to facial recognition software. And Alessandro Acquisti from Carnegie Mellon University reveals some surprising research into how regular recognition tech can identify “anonymous” people. (Runs 22:38)
Play audio:
Efficiency Is The New Power
Jonathan Koomey
Ok, Moore’s Law. It is… um… we’ll let you read up on it on your own! We’re interested in someone who proposes a modification of it. Jonathan Koomey is a consulting professor at Stanford University, and his research shows that it’s not processing power that doubles every 18 months, it’s energy efficiency. And in a world dominated by mobile devices and mobile batteries, efficiency may become the new power. (Runs 3:39)
Play audio:
Predictable Us
Jure Leskovec
No two snowflakes are alike, no two people are the same… right? You may think you’re unique, but it turns out you’re awfully predictable. Jure Leskovec is an assistant professor of computer science at Stanford, and he analyses past human behaviour online to predict future outcomes. And he’s discovered he can correctly predict who your next friends on Facebook will be. (Runs 9:57)
Play audio:
Smart and Sensing Cities
Ayesha Khanna
What happens when cities can monitor and respond to the people who live in them? There is no end to the Spark obsession with this question. Ayesha Khanna, director of the Hybrid Reality Institute, talks to Nora Young about the potential, and the challenges of smart cities, and what becomes possible when sensors are embedded everywhere. (Runs 12:20)
Play audio:
Additional Links
- CBC News: Copyright bill reintroduced in identical form
- Nora’s full interview with David Fewer about former copyright reform bill C-32
- Main page image of Facebook relationships visualized by Paul Butler
- APM music used in this episode
Spark Podcast
You can receive Spark automatically by subscribing to any of our totally free podcasts.
Here is a way to link your small C-11 http://c11.ca mention with the rest of the episode.
The only way to have a reasonable conversation about the omnibus bill C-11 that doesn't loop you in circles and burn your brain is to divide it into components.
What I do is divide into 4 sections:
The "Copyright" component and the "Paracopyright" (AKA: technical measures, digital locks) component. Then subdivide each of those into policies proposed as part of the two 1996 WIPO treaties, and policies not related to those treaties.
The most controversial and least understood part of the bill are the non-treaty Paracopyright sections of the bill. Specifically, the access control part of the bill which has no tie to activities previously regulated by copyright (IE: access), and where the prohibition against unlocking is untied in any way to activities that would otherwise infringe copyright.
Details: http://billc32.ca/faq#wipovsdmca
Translated to real-world scenarios, what we are talking about is two types of digital locks: one on content that makes the content only accessable with "authorised" devices, and a separate lock on devices (computing hardware and software) which locks the person who possesses the device (most often its owner) out of the device.
In nearly all real-world scenarios, the person who owns what is locked (copyright or the device) do not control the keys to the locks. Locks, digital or otherwise, protect the interests of the keysholders. It is the law that normally protects the owners, but C-11 protects the interests of keyholders against the interests of owners.
Think about this: this is a device which treats its owner as a threat, and where it is illegal to change the software to meet the needs of its owner rather than a third party.
Having non-owner locked devices won't reduce copyright infringement , but will have implications that go far beyond copyright to the types of other issues discussed in the episode. These devices will by design be spying on those using them. Investigating how much unauthorized distribution of this private data is going on requires an unauthorized circumvention of a non-owner digital lock.
Bill C-11 alleges to allow security research, but prohibits the distribution of tools required to unlock our devices in order to do that independent research.
I wish people spent more time focused on this part of the bill. There are a lot of interesting things that can be said about the Copyright components of the bill, but really dangerous and widely misunderstood components aren't related to Copyright at all.
Ayesha Khanna
used a great term
'Technology Sherpas'
Like you said 'intermediaries',
boundary people, expert yet teacher,
guiding us along those treacherous slopes.